NASA SP-118 SPACE-CABIN ATMOSPHERES 3 -2 8- Part ;IV,Pngineering Tradeoffs of One- Grsus Two-Gas Systems6

NASA SP-118 SPACE-CABIN ATMOSPHERES 3 -2 8- Part ;IV,pngineering Tradeoffs of One- Grsus Two-Gas Systems6

A literature review by 6 Emanuel M. RothfiA4.D.

Prepared under contract for NASA by Lovelace Foundation for Medical Education and Research, Albuquerque, New NIexico

Scientific and Technical Information Division OFFICE OF TECHNOLOGY UTILIZATION 9 1967 10 c" )NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Washington, D.C.3 PREC€t?iNG PAGE BLANK NOT FILMQ.

Foreword

THISREPORT is Part Iv, the last volume of a study on Space-Cabin Atmospheres, conducted under sponsorship of the Directorate, Space Medicine, Office of Manned Space Flight, National Aeronautics and Space Administration. Part I, “Oxygen Toxicity,” was published as NASA SP47, Part 11, “Fire and Blast Hazards,” as NASA SP-48, and Part 111, “Physiological Factors of Inert Gases,” as NASA SP-117. This document provides a readily available summary of the open literature in the field. It is intended primarily for biomedical scientists and design engineers. The manuscript was reviewed and evaluated by leaders in the scientific community as well as by the NASA staff. As is generally true among scientists, there was varied opinion about the author’s interpretation of the data compiled. There was nonetheless complete satisfaction with the level and scope of scholarly research that went into the preparation of the document. Thus, for scientist and engineer alike it is anticipated that this study will become a basic building block upon which research and development within the space community may proceed.

JACK BOLLERUD Brigadier General, USAF, MC Acting Director, Space Medicine Office of Manned Space Flight

iii FXECEDING PAGE BLANK NOT FILMX.

Contents

PAGE INTRODUCTTON ...... vii Chapter 1: PHYSIOLOGICAL CONSIDERATIONS ...... 1 TOTAL PRESSURE ...... 1 OXYGEN ...... 1 WATER VAPOR...... 4 CARBON DIOXIDE...... 4 DILUENT GAS ...... 5 TOXIC SUBSTANCES AND ODORS ...... 6 DUSTS, AEROSOLS, AND IONS ...... 6 AIR CIRCULATION...... 6 TEMPERATURE CONTROL...... 7 Radiation Heat Transfer ...... 7 Forced Convective Heat Transfer ...... 7 Free Convective Heat Transfer ...... 9 Evaporative Heat Transfer ...... 10 Combined Heat Transfer ...... 11 Chapter 2: ENGINEERING CONSIDERATIONS ...... 17 WEIGHT CONSIDERATIONS...... 18 Structure of Cabin Wall ...... 18 Atmospheric Leakage ...... 19 Tankage for Gas...... 29 Air-conditioning System ...... 71 TRANSIENT PHENOMENA ...... 99 POWER SYSTEM FACTORS ...... 100 ECONOMIC AND OPERATIONAL FACTORS ...... 100 Development Time ...... 100 Uses of Existing Hardware and Equipment...... 100 Maintenance and Convertibility ...... 100 Crew Acceptance...... 101 Contaminant Buildup ...... 101 Qualification Testing...... 101 Environment for Inflight Experiments ...... 101 Complexity of Design and Operation ...... 101 cost...... 102 Chapter 3: COMPARATIVE ANALYSIS OF ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM...... 103 EFFECT OF MISSION LENGTH ON OVERALL ECS TRADEOFFS ...... 107 SUMMARY OF TRADEOFFS IN THE SELECTION OF SPACE-CABIN ATMOSPHERES ...... 109 REFERENCES ...... 115 Appendix A: NOMENCLATURE ...... 121 Appendix B: CONVERSION TABLES...... 125

V Introduction

THE SELECTION of an ideal space-cabin atmosphere requires thorough analysis of physiological, physical, and engineering considerations of the problem. Since the basic function of a cabin atmosphere and its control system is to provide an environment for the optimum function of both crew and equipment, the specific interaction between the two must constantly be kept in mind. In this fourth part of the series on Selection of Space-Cabin Atmospheres, an attempt is made to consider this interaction in establishing valid criteria for engineering tradeoffs. It has been the general practice in this series to avoid relating the analysis of literature to any one specific space mission. This philosophy will be continued in the present study; however, meaningful engineering tradeoffs can be made only with rather well-defined physical constraints on the system. Because the manned- orbiting-laboratory concept allows adequate and realistic constraints to be set, it will be used as a model example in this analysis. Since the success and safety of each mission and crew are to a considerable extent dependent on the choice and design of atmospheric and thermal control systems, synthesis and optimization methods have become essential steps in atmosphere selection. Because every bit of weight and volume must be saved and every fraction of performance extracted from every subsystem, with no loss in reliability, and with economy of effort and at minimum cost, sophisticated analyses have been required for total systems integration. In the past, studies of this type have been of great value to the engineer in presenting heat- and mass-transfer data, as well as chemical process descrip- tions which include the direct influence of vehicle data, system variables, process selection, and reliability considerations. Very often the gas-specific factors have, by necessity, been included in the analyses, but their roles have not been pointedly delineated. It is hoped that the role of the gas-specific factors and the many biases surrounding their choice does become more clear as a result of the present analysis. The study begins with an evaluation of the physiological considerations which set boundaries for the physical environment within the cabin (ch. 1). The engineering analysis of chapter 2 is a review of the interaction between the physiological and hardware parameters of the environmental control system. An attempt is made to compare for each subsystem the effect of several physiologically acceptable gas mixtures on the weight and power penalties for missions of different types and durations. In chapter 3, the tradeoff criteria established in chapter 2 are used in an analysis of a typical mission-a two-man orbiting vehicle. The results of tradeoff analyses performed by several groups are compared to demonstrate the sensitivity of the final product to the physical, physiological, and engineering assumptions which were made. This mission was chosen only because of the availability of several completely independent studies of the tradeoffs by the aerospace industry. These independent analyses offer the opportunity for evaluation of assumptions and biases which usually creep into any tradeoff study. It is hoped

vii ... Vlll INTRODUCTION that this review will provide an adequate basis for unbiased atmospheric tradeoff studies of future missions. Several excellent reviews of analytical methods for atmospheric control processes have been used for basic source materials. 24,20,*0,95 The unpublished data of the Boeing Co. on engineering tradeoffs of different gas systems were also of great value in establishing the gas-specific variables critical to this study.12 At the NASA Manned Spacecraft Center in Houston, there is presently a project on the design of computer-assisted design and tradeoff t00ls.48 Such a program should be of great value in simplifying the approach to many of the problems raised in this review. The nomenclature and units used throughout the paper are presented as appendix A. Tables for use in converting units from one system of units to another are presented as appendix B. CHAPTER 1

Physiological Considerations

THE BASIC physiological requirements for selec- OXYGEN tion of space-cabin atmospheres have been dis- An average oxygen consumption rate of 1.8 cussed by many investigators, as well as by the to 2 lb/man/day can be assumed. Partial pres- first three parts of the present series.m.91992 sure of oxygen in a space vehicle should be ideally Since exhaustive review of these factors is beyond maintained above a minimum point which allows the scope of this present analysis, only a brief a blood saturation of at least 95 percent to insure summary will be presented. The major factors optimum performance. The upper limit of are as follows: oxygen partial pressure is far from clear.g0 Fig- (1) Total pressure ure 1 indicates the operational envelope of pres- (2) Oxygen sure and oxygen which can be considered in the (3) Water vapor space cabin. The curve is based on exposure for (4) Carbon dioxide 1 week or more. To maintain the same degree (5) Diluent gas of oxygen saturation in the blood as in air at sea (6) Toxic contaminants and odors level when total pressure is decreased, the per- (7) Dusts, aerosols, and ions centage of oxygen in the atmosphere must in- (8) Circulation of atmosphere crease as shown by the “sea level equivalent” (9) Temperature control curve. The unimpaired performance zone (center TOTAL PRESSURE clear zone) indicates the range of variation that The boundaries of pressure limitations in can be tolerated without performance decrement. space cabins are more dependent on engineering The maximum oxygen tolerance (definite path- realities than on physiological limits. For the ology) for long periods is currently under investi- upper limit, it is certainly not necessary to con- gation. The role of nitrogen and trace contami- sider pressures greater than 1 atmosphere (760 nants on the symptoms of oxygen toxicity in the mm Hg). The establishment of the lower limit oxygen range of 90 to 100 percent is still open to is determined by the desire to keep the alveolar question, as shown by the right-hand area.m partial pressure of oxygen as close as possible Prolonged exposure to the low oxygen levels to the sea-level equivalent of 104 mm Hg. At illustrated to the left of the unimpaired perform- sea level, this is attained in a dry air free of car- ance zone requires special acclimatization. Accli- bon dioxide and water vapor with an oxygen matization can be accomplished by continuous ambient partial pressure (PO,) of 160 mm Hg. In exposure to successively lower pressures, with the presence of an alveolar pco2 of 40 mm Hg and little intermediate return to higher pressures. Op- alveolar water vapor of 47 mm Hg, the minimum timal acclimatization to allow survival at 25 OOO total pressure of a pure oxygen cabin should be feet requires 4 to 6 weeks. The minimum tolerable 104+40+47, or 191 mm Hg. One must there- total pressure is based upon the effective partial fore examine the engineering implications of pressure of oxygen. Decompression which may total pressures between 760 and 191 mm Hg. occur below a total pressure of 300 mm Hg in the

1 2 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

760

5 000 600

c 0 1 0 000 - i 0 a .-c -c 1 5 000 0 c c 400 -0 0 20 000 .-> > U W 25000

30 000 2 00 35 000 40 000 100 45 000 0 10 20 30 40 50 60 70 80 90 100

Oxygen in atmosphere, percent volume

FIGUREl.-Oxygen-pressure effects. (MODIFIED FROM THE NASA BIOASTRONAUTICS DATA BOOK AFTER THE DATA OF LUFT.~) absence of adequate denitrogenation is disre- At a recent NASA conference on selection of garded. post-Apollo atmospheres, it was reported that Figure 2 represents the approximate time of electron-microscopic changes in mitochrondial appearance of signs and symptoms of oxygen structure have been seen in the liver and kidney toxicity. The nature of the symptoms varies with of several animal species after prolonged ex- the ambient partial pressures of oxygen which posure to 5-psia, 100 percent oxygen.87 There cause them. Above 760 mm Hg, the central were no specific symptoms or clinical chemistry nervous system is the primary site of defect, findings associated with these lesions, and their with symptoms such as nausea, dizziness, con- meaning is not clear. The US. Air Force toxi- vulsions, and syncope. In the range of to cology studies at the Aerospace Medical Labora- 760 mm Hg, respiratory and nervous system tories at Wright-Patterson Air Force Base have symptoms predominate. These are substernal recently revealed that most of the abnormal distress (bronchitis and probably atelectasis), blood chemistries seen in dogs and monkeys paresthesias, and nausea. In the range of 200 to early in exposure to 5-psia, 100 percent 02return 400 mm Hg, reported symptoms are respiratory to normal within 6 months.2 Only blood lactic and, possibly, hematological and renal: atelecta- dehydrogenase (LDH), serum pyruvic glutamic sis, oxidative hemolytic anemia, and protein and transaminase (SPGT), and serum glutamic-oxalic casts in the urine. Studies from the Gemini IV, transaminase (SGOT) remain slightly above the V, VI, and VI1 missions suggest that oxygen upper limits of the normal range. These slight toxicity has not been entirely ruled out as a factor abnormalities may represent adaptation to 5-psia, in the decrease of red blood cells on exposure to 100 percent 02. the 258 mm Hg (5psia) po2 used in these cabin^.^ Recent studies on susceptibility of humans to The role of tocopherol deficiency and plasma lipid pulmonary atelectasis indicate that the ratio of peroxides in the hemolytic process is currently pulmonary air conductance to lung volume ap- under 729 lo pears to be a significant factor in individual PHYSIOLOGICAL CONSIDERATIONS 3

40 80 120 160 200 240 2

Time to onset of symptoms, hours

FIGURE2.-Time of onset of signs and symptoms of oxygen toxicity. (AFTER WELCH1’* AND ROTH.~) sus~eptibility.~~When this ratio, measured in the problem, but this factor has not been clearly liters-sec-’-cm H20-l per liter of lung volume, establi~hed.~~The use of 5.0 psia instead of 3.5 is less than 0.13, atelectasis is seen after exposure psia in Mercury and Gemini cabins appears to to 5-psia, 100 percent 02. Fortunately, most have been dictated by the desire to maintain a “normal” subjects tested have ratios of greater pressure high enough to minimize the chance of than 0.14. This would suggest that selection of pulmonary atelectasis, as well as to reduce the astronauts for resistance to atelectasis in 100 chance of decompression sickness. Early in the percent 02 space-cabin environments may be mission, the residual nitrogen even after several practical. The condition may be ameliorated by hours of preoxygenation could still cause symp- , recurrent deep-breathing exercise. It was con- toms.w Also, the selection of cabin pressure at cluded at this conference that 5-psia, 100 percent 5 psia allows for a backup emergency suit circuit 02 would be acceptable for space missions of operating at 3.5 psia. This backup mode is auto- less than 30 days’ duration. For longer exposures matically initiated when the cabin pressure falls in space, ground-based validation experiments below 3.8 psia. Another factor often mentioned may be required for at least the duration of the in the choice of the higher pressure is the added expected mission. safety factor of a longer decompression time in A partial pressure of pure oxygen at 3.0 to 3.5 case of puncture of the sealed cabin. It is obvious psia would appear to eliminate the problem of that some oxygen pressure between 3.5 and 5.0 oxygen toxicity caused by oxygen at 5.0 psia, psia will be the optimum for most mission types, a poz of 100 mm Hg above normal atmospheric but the exact pressure of choice will be dictated partial pressures of oxygen. The near absence by other physiological and engineering considera- of nitrogen should not play a significant role in tions. 4 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

WATER VAPOR 90

A very important constituent in the atmosphere ! is water vapor. It is customary to plot a psychrometric chart comparing dry-bulb and wet-bulb 80 temperature, saturation temperature, relative

humidity, and dewpoint temperature to delineate LL the comfort zone. Most graphic representations F 70 a c are useful only at one pressure, usually around mI sea level. In figure 3(4 this problem is avoided by .5 plotting dry bulb against dewpoint temperature, e 60 resulting in saturation and relative humidity lines -c13 which are independent of pressure. Figure 3(b) s defines comfort zones more clearly. Other psy- SO chrometric presentations can be seen in figures 73 and 74 as well as in the “Temperature” section (ch. 7) of the NASA Bioastronautics Data Book.” 40 Parameters such as effective temperature and lines of equal comfort are also available.20*’ 50 60 70 80 90 100 The humidity acceptable in space vehicles is Dry bulb temperature, “F thus a function of the temperature, but should lie (b) Summer and winter, (AFTER ASHRAE GUIDE.’) within the vapor pressure range of 5 to 16 mm Hg. The removal of water is a thermodynamic process FIGURE3.-Definition of comfort zone-Concluded. which constitutes .a large percentage of the thermodynamic load on the atmospheric control Btu/hr (resting) to about 10oO Btu/hr (severe system. The amount of water vapor added by the exercise) can be expected as extreme ranges, occupants can be measured by the so-called with an average of 150 to 200 Btu/hr over a 24- latent heat load whereby each pound of water hour period for each person in a multimanned evaporated into the air is represented by about crew. This is a conservative value, considering 1050 Btu. Latent personal heat loads of from 70 the 83 Btu/hr reported for the astronauts in Vostok.109 The biological implications of water Dry bulb temperature, OC metabolism in space vehicles are currently under IO 15 20 25 30 35 40 90 study.”’ 30 - 30 CARBON DIOXIDE -25 *O 25 The carbon dioxide production rate is a func- Y U -20 0 tion of the oxygen usage rate, and the respira- f” 70 E < 20 5 c tory quotient of about 0.82 to 0.85. Since an ? -15 e 1L g60 a average 02 consumption rate of 2 lb/man day 15 $ L c can be expected, the CO2 production rate of .--c 0 10 about 2.2 lb/day can be assumed. d It is generally considered that the best prac- 40 5 tice is to keep the level of carbon dioxide below

-5 4 mm Hg and to have a maximum allowable 0 30 level of 7.6 mm Hg. Figure 4 represents the 50 60 70 80 90 100 110 rationale behind this choice by indicating the Dry bulb temperature, OF general symptoms common to most subjects (a) Temperature-humidity effects. (AFTER COE ET when exposed to mixtures of carbon dioxide in AL.ZO) air at a total pressure of 1 atmosphere. Some FIGURE3.-Definition of comfort zone. adaptation to higher concentration is possible PHYSIOLOGICAL CONSIDERATIONS 5 - 12 90 of electrical equipment in unusual atmospheres ”ac Co.27 f 10 15 is now under study at the Boeing Pre- n E liminary results suggest that there is little or a- f8 60 no difference between gas mixtures in arcing fn V m L tendency at a given pressure unless the amount -E6 45 ; -m 3 of inert gas is above 75 percent. Neon has a 1,4 30 f lower breakdown voltage than the others. .-L -CL m m .E 2 15 E Studies of plastics burning in closed cham- N bers containing 100 percent 02 at 5 psia and in u0 1 zobe I -‘~oeff&t 1 1 1 I1 0 0 0 10 20 30 40 50 60 70 80 other gas mixtures during zero-gravity parabolic Time, minutes 40 days flight maneuvers suggest that the zero-gravity (a) (b) factor in suppressing flame propagation may FIGURE4.-Carbon-dioxide effects. (FROM DATA OF more than compensate for the increased flam- KING,& NEVISON,77 AND SCHAEFER.99) mability in 100 percent 02.100*54 However, the lack of forced convection in the closed but not without biochemical alterations in the system during simulation is somewhat un- body.99 In zone I, no psychophysiological per- realistic and may give a false sense of safety. formance degradation or any other consistent On the other hand, it should be remembered effect is noted. In zone 11, small threshold hearing that in actual space vehicles a nearly zero- losses have been found, and there is a perceptible convection state can be readily attained by doubling in depth of respiration. In zone 111, the merely shutting off circulation fans soon after zone of distracting discomfort, the symptoms are the fire has been discovered. Future zero-g judgment errors, mental depression, headache, studies are being planned to include forced dizziness, nausea, “air hunger,” and decrease convection at levels similar to those expected in visual discrimination. Zone IV represents in operational space cabins. One must, how- marked deterioration leading to dizziness and ever, always keep in mind the fire hazard during stupor, with inability to take steps for self- the positive-g phases of launch and reentry. preservation. The final state is unconsciousness. Presence of inert gases can lead to physiolog- The bar graph at the right shows that for pro- ical problems related to decompression sick- longed exposures of 40 days, concentrations of ness, explosive decompression, ebullism, pre- carbon dioxide in air of less than 0.5 percent oxygenation scheduling, thermal control, and (zone A) cause no biochemical or other effects; voice propagation. These have been covered concentrations between 0.5 and 3.0 percent in great detail in part I11 of this series. sz The (zone B) cause adaptive biochemical changes, general conclusion of this study was that nitro- which may be considered a mild physiological gen, helium, and neon were the only gases worth strain; and concentrations above 3.0 percent considering from the physiological point of (zone C) cause pathological changes in basic view. There was no overwhelming physio- physiologkal functions. logical mandate for the preference of any one of these three diluents. On theoretical grounds only, neon offers DILUENT GAS some advantage by potentially minimizing the The value of an inert diluent gas in decreasing bends and chokes as well as the more serious the fire and blast hazard and possibly the oxygen neurocirculatory collapse symptoms of decom- toxicity hazard in space cabins has already been pression sickness. Only minimum diving data discussed in parts I and 11 of this series.90991 support this theoretical conclusion. From em- More recent work is available on the prediction pirical studies simulating decompression during of ignition and burning rates of different ma- early stages of flight in non-steady state gas con- terials in pure oxygen and mixed gas sys- ditions, it appears that helium is equal to or even tems. 41, 4O A continuing bibliography on fire more dangerous than nitrogen in causing bends and blast hazards is also available. 93 The arcing at altitude. Studies under equilibrium conditions 6 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS are required.92.112, Theoretical and empirical more dangerous from the points of view of studies, on the other hand, also suggest that increased fire hazard 4O and decreased time of helium is less likely than nitrogen to cause neuro- decompression. The 7-psia mixture would have circulatory collapse symptoms after decompres- a greater tendency to cause decompression sion. One must also consider the very low symptoms after total decompression occurred. incidence of bends expected in an astronaut The engineering factors determining the choice population decompressing from a 5- or 7-psia between the three proposed conditions are cabin and the even lower incidence of neurocir- discussed in chapter 2. culatory collapse expected under these conditions?' The incidence for older and less physically TOXIC SUBSTANCES AND ODORS conditioned scientist-observers in the crew The many toxic substances and odors which will no doubt be somewhat greater. can be generated by the hardware or human Previous opinions notwithstanding, the sub- occupants have been discussed in great detail stitution of helium for nitrogen would produce in many st~dies.~.39A review of these sub- no overall increase in the decompression safety stances and their effect on body performance is factor to make helium preferable to nitrogen. beyond the scope of this study. The adverse thermal and voice-distortion It is clear that the chemical constituents of factors predicted for helium have been shown the atmosphere can alter the toxic materials to be minimal at the pressures and composi- generated in the cabin.w The nature of these tions suggested for space cabins. The only alterations and the effect on humans is currently gross advantage of helium over neon or nitrogen being studied by the Aerojet-General Corp. and is in minimizing the hazard of lung damage the Aerospace Medical Laboratories at Wright- from explosive decompression. The extremely Patterson Air Force Base.2 low probability of this event occurring without Provision must be made for continuous reduc- lethal mechanical injury to other parts of the tion of the amount of toxic substances in the body indicates that this factor must be weighted cabin atmosphere. This entails the passage quite low among the selection criteria. of cabin atmosphere through catalytic burners From the preceding discussion it would appear or adsorption devices at relatively low velocities. that the engineering, and not the physiologi- The gas-specific effect of this requirement on the cal, factors will have to play the major role in engineering consideration of cabin atmosphere the selection decision between the three inert selection will be discussed subsequently. diluents. It is also clear that the lower the partial pressure of inert diluent, the safer is DUSTS, AEROSOLS, AND IONS the mixture from the point of view of decompression sickness but not fire. The minimal The zero-gravity environment will increase the partial pressure of inert diluent required for hazard of dust, aerosols, and ions over and above normal metabolic function over long periods that experienced on Earth. Again, the chemical of time has not as yet been determined.g2 constituents of the atmosphere will alter the In the presence of a fixed partial pressure of hazard presented by these agents. Removal of oxygen, the higher the partial pressure of inert these materials requires filters and similar de- gas, the longer is the time of decompression to vices through which the atmosphere must be hypoxic levels, but the more dangerous is the circulated and imposes gas-specific penalties in mixture in production of symptoms once de- atmosphere selection. compression has occurred. As an alternate to AIR CIRCULATION the 5-psia, 100 percent oxygen environments already used in space vehicles, one can suggest Absence of convection in zero-gravity states mixed gas systems ranging from 5 psia with 70 requires that air be circulated through the cabin percent oxygen and 30 percent inert gas to at relatively high velocities, primarily to dissipate 7 psia with 50 percent oxygen and 50 percent local heat from equipment and crew and also to inert gas. The 5-psia mixture would be slightly distribute C02 and contaminants from local PHYSIOLOGICAL CONSIDERATIONS 7 sources. The flow rates must be sufficiently de- single crewman having an effective body surface signed to accomplish these functions without area of 15.6 sq ft with a uniform clothing surface ' producing discomforting drafts in the cabin.62.8 temperature and an enclosure of greater than Because of the relatively high volumes and veloc- 100 sq ft per man, that is, few= eC, a simplified ities involved, the physical properties of the radiation cooling equation can be written as b atmosphere being circulated play a major role in the engineering considerations. Qr-2.65X 10-g E~(T~~-T~~)(3) TEMPERATURE CONTROL A sample graph, assuming E= 0.9, is seen in Major physiological factors bearing on engi- figure 5 where the radiation heat loss to any neering tradeoffs of different atmosphere systems given environmental temperature is given for are the thermoregulatory parameters of man. several clothing surface temperature^.^^ Figure 6 Because of the sensitivity of tradeoffs to these represents the radiation heat-transfer coeffi- parameters, and the somewhat scattered data cients (hr) for different combinations of en- pertaining to this problem, an attempt is made to vironmental and clothing temperature. The analyze in detail the critical data and equations equation used in this figure can be derived from to be used in the presentation of subsequent engi- equations (3) and (5).23 neering considerations. Several excellent re- Forced Convective Heat Transfer views of thermal physiology have been used in the preparation of this section.113 81 38,589 117, 7 The correlation between convective heat- The processes of heat rejection used by man transfer processes and mass-transfer processes include conduction radiation, convection, and has been used by many investigators to develop evaporation of moisture from the skin and lungs. analytic models for forced convective heat ex- Conductive loss in space cabins can be neglected. The convective and evaporative losses may occur 700 under free- or forced-convection conditions. The primary atmospheric parameters that may be varied are temperature, velocity, and humidity 600 of ventilating gas. The primary physiological parameters of the human subject are the metabolic rate; sweating characteristics; peripheral 500 blood flow; body size, position, and temperature distribution; comfort criteria; and the thermal conductivity and water vapor permeability of f 400 clothing. z If one assumes no heat storage or loss, the heat balance equation for man in steady-state condi- uL 300 tions in a zero-g environment may be written as

200 QM- Qw= Qr + Qc + QI= Qs + QI (1)

Radiation Heat Transfer 100 The radiation heat loss equation for man is

Qr = mfcJr(7'c4 - Tw4) (2) 0 40 50 60 70 80 90

Calculation of the actual radiation-heat- Twt OF transfer problem is a complex one involving

numerous thermal radiation sources at different FIGUREJ.--Radiative heat IOSS from man to his temperatures and geometries. By assuming a surroundings. (AFTER PARKER ET AL.'9 8 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 1.05 properties of the fluid, the geometry of the body, and second- or third-order factors of fluid flow. 1.00 The value of h, for convective exchange about 1

0LL the whole human body is a critical coefficient

N= .95 to which the gas-specific engineering tradeoffs L -c are sensitive. Unfortunately, there has been 3 .90 some variance between the values used by L c several different groups in relating the h, of man 1-65 .85 1-65 to the atmospheric gas velocity. Selection of the appropriate film coefficient or actual heat- transfer coefficient is a difficult problem.59 A 55 60 65 70 75 80 85 90 discussion of the implication of different co- T,, "F efficients used in the analysis of forced convection about the human body has been published FIGURE6.-Radiation heat transfer coefficient. recently.53 h,=ea (%12)~=0.9;___ T, is clothing tempera- The early data of Winslow et al.116 suggest that for clothed humans sitting in a turbulent ture; and T, is environmental temperature. air flow the following equation may be used: (AFTER PARKER ET AL.") change in man. In the recent analysis of Beren- - son,7 the following assumptions were made: hc=0.153 VO.5 (1) All sensible heat passes through the clothing by conduction, and the clothing heat-transfer This equation is not too different from that area is equal to the skin surface area. Since derived for rough flat plates.12 sensible heat loss occurs from nonclothed skin and since the clothing surface may be up to 40 percent greater than skin surface, these assump- hc= 1.03 k ($7'. (7) tions may be conservative. It must be remembered, however, that even though the area Figure 7 represents a summary of several ap- increases, the air pockets which are formed proaches to forced convective heat-transfer act as thermal and mass transfer resistances. coefficients (convective film coefficients). The Zero gravity will tend to increase resistance by first three curves represent the h, values ob- eliminating convection currents in the pockets. tained from data on empirical studies of hu- (2) In the absence of conductive heat loss, the man~.~~,m, 76 These are compared with four relationship between clothing surface tempera- theoretical curves: a cylindrical model of man ture T, and skin temperature T8 can be deter- in crossflow, a flat plate with flow perpendicu- mined by the equation: lar to it, a 10-inch diameter cylinder in crossflow, and a cylinder in longitudinal flow (fig. 8). The T, = T, - L (+Qc+Qr value of h, for the cylindrical model of man (4) corresponds closely with those obtained by Nelson76 and is equivalent to h, for crossflow The value of is the useful function of cloth- L/k about cylinders 5 inches in diameter. The specific ing heat-transfer resistance, Clo, where 1 equation used for the flat-plate model in figure 7 Clo = 0.88" F-ft2-hr/Btu. was not stated but appears to differ from the flat- The rate of heat transfer by convection from plate equation noted above 12 which gives results clothing surface can be written as closer to those of the equation of Winslow et al.116 Figure 9 shows the effect of gas velocity on Qe=hcA(Tc-Ta) (5) the convective heat-transfer coefficient based The convective heat-transfer coefficient is on the cylindrical model of man for various actually a complicated function of thermal helium-oxygen and nitrogen-oxygen atmospheres. PHYSIOLOGICAL CONSIDERATIONS

1 4 7.5" IC

Xi N Lz I! I! f s U I

Velocity over man, ft/min

(1) Hall 34 (2) Winslow et a1.lla (3) Nelson et a1.16 (4) Cylindrical model of man in cross-flow (analytical) (5) Flat plate (Hamilton-Standard) (6) 10-inch-diameter cylinder in cross flow (7) Longitudinal flow

FIGURE7.-Comparison of forced convection film coefficient for man at atmosphere. (AFTER PARKER ET AL.'@')

The partial pressure of oxygen of 170 mm Hg is near the sea-level equivalent and is held constant, with the diluent gas ranging from 0 to 400 mm Hg. These curves were generated by taking 8 19.5 ft2 used to include some factor of safety. the heat-transfer coefficient to be proportional b Each finger: 3!$ inch long by 76 inch diameter. to the various fluid properties as follows:

FIGURE8.-CylindricaZ model of man. (AFTER PARKER ET AL.78)

mates the forced-convection cooling rate for The values for neon mixtures will lie between nitrogen-oxygen mixtures:' those for helium and nitrogen. It is clear from comparing physical properties of the gases that - for different mixtures of oxygen and nitrogen, Q, = 0.407d PV( T, - Ta) (9) there is little sensitivity of h, to percent composi- Free Convective Heat Transfer tion of gas (table 4). The following equation, derived from the heat In the presence of a gravitational field, such as mass-transfer analog28 for Prandtl numbers of 0.6 that on Earth, planetary surfaces, or rotating to 15 and Reynolds numbers of 10 to lo5,approxi- space stations, free convection is possible and

261-559 0-67-2 10 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 1.8 With a 10" F temperature difference between clothing surface and atmosphere, the critical 1 .6 velocity is 26 ft/min on Earth but only 10.5 ft/min ' on the Moon. 1.4

Evaporative Heat Transfer 1.2 Y I In addition to the vasodilation caused by rd g 1.0 elevated temperature, the body loses heat by 2 3 evaporation of sweat. At skin temperatures above & .8 91.4" F, the sweat rate appears to depend only on u L intracranial temperatures. Below this skin tem- .6 perature, sweating is reduced by decreasing .4 skin temperature.6 The rate of sweating and humidity loads on the environmental control .2 system have been discussed previously. Krantz 58 employed the data of Winslow et al.l17 0 to establish comfort indices for high temperatures 0 20 40 60 80 100 120 140 160 and related these data to the percentage of evap- Velocity over man, ft/min orative capacity of the body being used. The FIGURE9.-Heat transfer coefficients of man in metabolic rate is estimated for any given level mixtures of oxygen and helium and of oxygen of activity, and the difference between the meta- and nitrogen at different gas velocities. Con- bolic and sensible heat loss is the required evap- ditions: 170 mm Hg of oxyeen; pHs=partial orative cooling rate. Environmental conditions pressure of helium in atmosphere; pN2=partial pressure of nitrogen in atmosphere; based on are evaluated with respect to maximum evapora- cylindrical model of man. (AFTER PARKER tive cooling capacity of humans. Table 1 repre- ET AL.'9 sents the expected comfort level relative to the percent of maximum capacity being used. There is the preferred mode of cooling because no is some controversy as to the absolute validity additional energy need be expended. of this concept in air and especially in unusual Berenson has combined the general free- atmospheres and at high total sweat output.111 convection equations with the assumptions regarding clothing effects to yield an equation for free-convection cooling: Maximum evaporative capacity, Comfort level percent

The handling of mixed free- and forced-con- 0 to 10 ...... Cold vection environments can be simplified by the 10 to 25 ...... Comfortable 25 to 70 ...... Tolerable McAdams rule; that is, both the free- and forced- 70 to 100...... Hot convective heat-transfer coefficients are calcu- Over 100...... Dangerous lated, and the higher of the two values is used.63 Berenson has shown that the critical crossover Approximately the first 10 percent of maximum point of the forced convection velocity (vcrit), capability represents basal insensible loss from where the forced-convection heat-transfer co- respiration and diffusion. These losses are, of efficient is equal to the free-convection coefficient, course, a function of the metabolic output and can be calculated for oxygen-nitrogen mixtures respiratory rate. by equating equations (9) and (10) and solving for Berenson 7 has presented a simplified equation (12) for latent cooling rates as derived from the heat-mass transfer analogy of Eckert and equa- -

PHYSIOLOGICAL CONSIDERATIONS 11 tion (8). One must keep in mind the effect of and T, = 95" F are seen in figure 10. These rates inert gas on the coefficients and second-order should be taken as the very upper attainable ? clothing factors in evaporative transfer."? ll1 levels.lll In the temperature range under consideration, the temperature and the dew point have relatively little effect compared with gas- stream velocity and ambient pressure. Since the heat-transfer properties of nitrogen- For general free convection in nitrogen-oxygen oxygen mixtures are independent of the fraction mixtures, the latent cooling equations have also of each component, equation (12) has been been developed by combining equations for free reduced by Berenson for all oxygen-nitrogen convection, transport properties of air, and mixtures in a forced convection environment to evaporative cooling 7 to yield: yield

The values of evaporative loss under conditions Under high workloads at low pressure, respira- of C = 1 give predicted results slightly higher than tory water loss becomes a more significant factor actually measured.19.111 Since the rate of evapo- in latent heat loss. Recent data are available on ration and the diffusion coefficient for water this problem.120 vapor are inversely proportional to pressure, it Combined Heat Transfer is clear that the latent cooling will increase with decrease in pressure. Sample calculations illus- In any environment, all of the above modes of trating the magnitude of the pressure, dew point, heat transfer may be used. The ambient dry-bulb and gas-stream-velocity effects at Ta= 80" F temperature, humidity, air velocity, and ambient pressure determine the partition of mechanisms actually used by the body. Figure 11 represents the changing partition 4000 of heat-loss mechanisms at rest with increasing drybulb temperature at a constant relative humidity of 45 percent and constant gas velocity.

3000

L 5 z 300 -- 2000 0 U ",200 0 L -S =l 100 1000

-100

0 50 100 150 - Temperature, dry bulb ambient (45% RH), "F V, ft/rnin FIGURE11.-Typical relation between human heat FIGURE10.-Maximum evaporation rate- in oxygen- balance and temperatures for lightly clothed nitrogen mixtures. Q I=246 T.&/P(P.-Po)~p; subjects in still air at sea level. (AFTER JOHN- T.=950 F; To=800 F. (AFTER BERENSON?) SON.~~) 12 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS The regions of primarily metabolic, vasomotor, in a 1-g environment. It must be remembered that and evaporative regulation are noted. Specifica- in zero-g, any velocity can move papers. The tion of the design atmospheric temperature, gas velocity threshold for movement is significant 4 velocity, and humidity will be critical in deter- only before and during launch and during reentry. mining the actual partition of loads on the air- Since the force of a gas stream is proportional to , cooling and dehumidifying systems and therefore pv2, a table of constant force thresholds equiva- will affect the gas-specific tradeoffs in question. lent to 50 to 60 ft/min in air can be calculated. The use of liquid-cooled garments or ventilated Table 2 represents this maximum-force velocity garments inside the cabin will not be considered along with the corresponding ambient tempera- as a basic mode of 0peration.~~*17Tradeoffs in ture T, required to maintain thermal comfort as contingency modes must take into consideration measured by average skin temperatures T, at such interactions between the cabin atmosphere 91" F and 94" F. The development of these and the suit. comfort-zone temperatures follows. Berenson has calculated comfort-zone predic- The thermal comfort zones for different gas tions for many different atmospheric conditions velocities of varied mixtures of oxygen and helium and gravitational states.8 These graphs may be and of oxygen and nitrogen in zero-g were cal- consulted for unusual conditions in space ve- culated by assuming the following conditions: hicles using oxygen-nitrogen atmospheres. For (1) Ta= T, (air temperature = ambient tem- cabins with helium-oxygen atmospheres, other perature = wall temperature) calculations need to be made to determine com- (2) No body-heat storage fort zones. (3) po2= 170 mm Hg in all cases and pN2or The general effect of helium mixtures on ther- pHeincreasing from 200 to 600 mm Hg moregulation in space cabins has been covered in (4) Zero-gravity environment part I11 of this series.92 Since the compilation (5) Evaporative heat loss is the same as at 1 of these data, several studies have been pub- atmosphere and lg, lished on the theoretical and empirical aspects (6) Convective heat loss is for cylindrical of this problem which are pertinent to the present model of man with (Ac= 19.5 ft2) in cross- study. flow (fig. 8) Parker et al.79 have compared theoretical gas- (7) Metabolic heat generation is for a man velocity limits and thermal comfort zones in seated at rest (400 Btu/hr at 70" F) helium-oxygen and nitrogen-oxygen mixtures of (8) T,= skin temperature in the 91" F range varying total pressure. To avoid the movement of (9) Cl0=1; ~2~0.9 papers at 1 atmosphere in air, a velocity of 50 to (10) Ar=15.6 ft2 and Ar/Ac=0.8 60 ft/min is stated as the tolerable upper limit of (11) Partition of heat loss is similar to that velocity above the 40 to 50 ft/min draft threshold seen in figure 11

TABLE2. -Maximum Velocity Over Man in 1-g Environment [AFTER PARKER AND EKBERG T9] [po,= 170 MM HG]

To,"F, required for- Maximum velocity over I man, ftlmin I T8=91"F T, = 94" F

0 0 100 to 120 56.5 to 58.5 66 to 67.5 200 0 94 to 113 65 to 66.5 72 to 73 400 0 88 to 106 68 to 69 74.4 to 75.5 600 0 84 to 100 70 to 71 76.5 to 77.5 0 200 71 to 86 61.5 to 63 69 to 70 0 400 57 to 69 63.5 to 65 70 to 71.5 0 600 50to 60 i64.5 to 65.5 71 to 72 ~~ ~

PHYSIOLOGICAL CONSIDERATIONS 13 (12) The clothing temperature T, is related to 90 atmospheric temperature T, by the relation n 80

LL 1.137 (Ts-Ta) '. 70 T,=T,+ (15) I- (0.8 hr+hc) Clo+ 1.137 60

(13) The relation of hr to Tc is the same as 50 that noted in figure 6; T,= Ta. 0 20 40 60 80 100 120 140 160 I In view of equations (13) and (14) which predict (0) Velocity over man, ft/min that the rate of evaporation is inversely related to the ambient pressure, it appears that assumption (5) is open to question. Since the evapora- -He-0, atmosphere (370 mm Hg) tion probably accounts for less than one-third 80 ----- N,-0, atmosphere (370 mm Hg) T., OF

of the total heat loss at the temperatures in ques- LL tion, the assumption of 1-atmosphere pressure '. 70 does not present too great an error. As for as- I-O sumption (4), it is true that in the presence of 60 adequate forced convection, the gravitational 50 factor in equation (14) would play a minimal role 0 20 40 60 80 100 120 140 160 in evaporative heat loss and can be neglected (b) Velocity over man, ft/min in the solution of comfort zone temperatures. The 0.25-power factor in this equation would in itself reduce the overall weighting of gravity effect. Assumption (12) holds only for Clo 9o I - He-0, Atmosphere (570mmHGj values in air. Where thermal conductivity is high, as in mixtures with high content of helium, there is much less Clo than calculated for air.101.35 Figure 12 represents the results of these calculations. As would be expected, the helium- oxygen mixtures show a narrower zone of com- 0 20 40 60 80 100 120 140 160 fort at higher temperatures, especially at lower (C 1 Velocity over man, ft/min flow rates than do the nitrogen-oxygen mixtures. This is more marked in the cases with higher 90 I -He - 0, Atmosphere (770 mm Hg) content of inert gas (12(c) and 12(d)). The tem- T perature values in table 2 indicate the zone of 80 comfort for the maximum gas velocities calcu- LL lated for each mixture as determined by figure 12. O 70 Secord and his coworkers have tested some +0 of these predictions of He gas effects in space 60

cabins.lo2 The tests were performed in their 50 space-cabin simulator at 5, 7, and 10 psia for a 0 20 40 60 80 100 120 140 160

mixture of helium and oxygen and at 5 and 7 (d) Velocity over man, ft/min psia for a mixture of nitrogen and oxygen. The wall temperature was within 10" F of gas tempera- FIGURE12.-Comfort lines for man seated at rest with 1 (AFTER PARKER ET AL.~~)(a) ture, so no major radiation factor was at play. Clo. PO,= 170 mm Hg. (b) p0,=170 mm Hg and PH~or Figure 13 presents a representative test run for pN2=200 mm Hg. (c) po2=170 mm Hg and one of the four crew members in helium and orpN,=400 mm Hg. (d) po2=170 mm Hg and oxygen at 5 psia, indicating test peaks and nadirs pae or PN~= 600 mm Hg. 14 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS about 65" F. The empirical data of figure 14 for 50 ft/min and 0.7 Clo suggest an average temperature of about 84" F. 4 The discrepancy cannot be fully accounted for by differences in Clo values in although the direction of the error is correct. The comfort temperature of figure 14 would be expected to be 0 30 60 90 120 150 180 210 at most only a few degrees higher than that of Time, min figure 12(b) if Clo were the only factor. Equa- tions (4) and (5) describe the convective relation- FIGURE15. -Represenfa tive thermal comfort.zone with an air velocity of 50 ft/min, a water partial ship involved. Preliminary data lo1 suggest that pressure of 12 mm Hg, a total pressure of 5 the Clo value of clothing is approximately in- psia, a helium-oxygen atmosphere. 0.7 Clo, versely proportional to the thermal conductivity and 1 experimental subject. (AFTER SECORD.~~) of the atmosphere.I0l Since the ratio of k for 5-psia Op-He/sea level air is 0.027/0.015 = 1.8 (table 4), the effective Clo value would only decrease from of the totally subjective temperature comfort 0.70 to about 0.39 Clo. In figure 14, the difference range. Figure 14 represents a summary of the between the comfort temperatures of zero Clo midzone temperatures under several conditions, and 0.7 Clo in a 7-psia helium-oxygen mixture is indicating a significant increase in allowable only 1" to 2". One would therefore expect that the cabin temperature when helium is used as a difference between 0.70 and 0.39 effective Clo diluent. It is also seen that the allowable increase would lead to an even smaller discrepancy. in cabin temperature appears to be a function There appear to be several flaws in the assess- of the Clo about the subject. The value of Clo ment of comfort in the experimental curves. was calculated by assuming air at sea level as The size of experimental group is inadequate. the interstitial gas. At 0.7 Clo, midzone tempera- The peaks and nadirs were entirely subjective. ture differences were noted as atmospheric con- One of the subjects was a Negro from the Deep ductivity was increased; at zero Clo (nude), South who, in spite of his residence in Los minimal differences were noted. Angeles for some time, had a very high average It is of interest to compare these results with comfort temperature and tended to skew the the predictions of figure 12. The atmosphere upper limit of the comfort zone of the four sub- of figure 12(b), containing 170 mm Hg of pop and jects to higher levels.101 Since his tolerance to 200 mm Hg of He and a total pressure of 370 low temperatures was also great, his mean com- mm Hg (7.1 psia), can be compared with the fort temperature was close to that of the other 7 psia He-0, mixture of figure 14. For a gas flow subjects. No average skin temperature for the of 50 ft/min and 1 Clo insulation, figure 12(b) four subjects at midcomfort zone was deter- indicates an expected average temperature of mined. This would help quantify the proposed skewing of comfort temperature curves by helium-oxygen mixtures. 0 Clo If the data of this one subject were eliminated \A LL 10 psia and a larger sample were studied, the theoretical 7 psia He-0, and empirical comfort temperatures would prob- He-0, ably lie closer to one another. This experiment He-0, 5 and 7 psia does corroborate the fact that for equivalent N2-02 partial pressures, the average comfort temperature in helium-oxygen mixtures is higher than in 0.015 0.025 0.035 0.045 0.055 nitrogen-oxygen mixtures but the experimental Conductivity, Btul°F-hr-ft2fft difference of 7" F is much greater than the 2" FIGURE14.--Space cabin comfort zone data. to 3" F predicted, for the 5-psia atmosphere of (AFTER SECORD.~~) 70 percent oxygen and 30 percent helium. PHYSIOLOGICAL CONSIDERATIONS 15 The recent studies of Welch and his co- subjects in loose-fitting surgical clothes of about workers112 tend to shed some light on this 0.25 to 0.50 Clo (sea-level air). problem. Unfortunately, determination of specific These data include varied numbers of different comfort ranges were not part of the Welch pro- subjects being studied under each gas mixture. tocol. Comfort temperatures were recorded as No windspeed measurements were made during the average cabin temperature set over periods these studies. Welch reports, however, that of several weeks by subjects who had control papers were not rustling and no comp!aints of over the thermostat within the cabins. These wind chill were recorded. If anything, he reports, temperature settings are presented in table 3 for gas velocity was on the "low side" in the test

3.7 psia, 100- 5 psia, 100- 5 psia 7.3 psia 7.3 psia percent oxygen percent oxygen par- 175 mm Hg ph- 150 mm Hg po.- 165 mm Hg pHe-74 mm Hg pHe-230 mm Hg h,,-206 mm Hg

Selected temperature, OF...... 69.3 70.9 74.7 75.4 72.7

cabin simulators. According to table 2, this would The predicted differences between 15(a) and indicate maximum expected velocities of less 15(b) are only a few degrees. than 100 ft/min in 100 percent oxygen and 70 When exercise loads are added to the normal ftlmin in the 7.4 psia, nitrogen-oxygen mixture. routine in mixed gas systems, there is some The temperature ranges in these experiments fall in between those predicted by figure 12(c) and the data of figure 14. No measurements of average skin temperature were made. It is diffi-

cult to determine which skin comfort tempera- Y 0 ture (Ts) of figure 14 most closely applies to 2-80 I- these experiments. Welch does report that some d L subjects tolerated air temperature as high as 2 cm 78.1" F in 100 percent oxygen at 5 psia without k 70 n complaint. cE These data seem to corroborate the impression U .-z that the data of figure 14 tend to be high. They E? 60 also suggest that for the 7-psia, 50 percent oxy- -c gen condition, the difference between comfort temperature in helium and nitrogen may be closer 501 0 50 100 150 200 to 2" to 3' F than to 7" to 8" F indicated by fig- Circulation velocity, ftlmin ure 14. The data of Welch for the 5-psia helium- mm Hg oxygen mixtures at 0.25 to Clo were pre- PIOs =173 0.5 PIHe = 75 mm Hg dicted by the comfort chart of Johnson46 in P1H20= 10 mm Hg figure 15(a). It is also interesting that for the PrCOa= 5 mm Hg 1 Clo conditions, the predictions of figures 12(c) and 15(a) are remarkably similar. Figure (a) 5-psia oxygen-helium mixture.

15(b) presents a comfort chart for 5-psia oxygen FIGURE15.--Human comfort chart. (AFTER paralleling the oxygen-helium chart in 15(a). JOHNSON.'6) 16 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

I I I I 0 50 100 150 200 Circulation velocity, ftlmin

P102 =243 mm Hg P1H20= 10 mm Hg P1C02= 5mm Hg

(b) 5.0-psia oxygen.

FIGURE 15.-Human comfort chart-Concluded. advantage in having helium present to de- several pertinent gaseous conditions. The crease the heat-storage index and maintain a importance of controlling Clo values, humidity, lower skin temperature. 29 From this review, and air velocity and establishing more specific one can conclude that accurate comfort-zone comfort zone criteria is apparent. temperatures for an average astronaut popula- With the physiological considerations out- tion under forced-convection, zero-g environ- lined, the next phase of this study considers ments and with variable, atmosphere-dependent the interaction of the biological and engineer- Clo values still remain to be determined for the ing variables in defining the tradeoff conditions. CHAPTER 2 Engineering Considerations

THE FOLLOWING is an outline of the engneering (b) Use of existing hardware and equipment considerations which must be covered in de- (c) Maintenance and convertibility veloping adequate criteria for tradeoffs: (d) Crew acceptance (1) Weight (e) Contaminant buildup (a) Structure of cabin wall (f) Qualification testing (b) Atmospheric leakage (8) Environment for inflight experiments (c) Tankage for gas (h) Complexity of design and operation (d) Weight-power penalty for air-conditioning (i) Cost system: In view of the physiological considerations 1. Cabin ventilation fan presented in chapter 1, the atmospheres ex- 2. Atmosphere processing fan amined from an engineering point of view are 3. Equipment cooling fan limited to those of 5 to 7 psia with pure oxygen, 4. Cooling system pumps, reservoirs, helium-oxygen, or nitrogen-oxygen mixtures. tubes, valves, radiator, and heat Since the physical properties of neon lie between exchangers those of helium and nitrogen and since little (e) Reliability factors is known of their physiological properties, neon (2) Transient Phenomena mixtures will not be covered in great detail. (a) Decompression time after puncture Where the discussion requires that specific (6) Transient overloads from environmental mixtures be compared, the several mixtures rep- control system failure resenting the limits of the physiological bound- (3) Power System Factors aries are used. These limits are 5-psia, 100 (a) Fuel cells percent 02;5 psia with 30 percent inert gas and 70 percent oxygen; and 7 psia with 50 percent (b) Solar cells inert gas and 50 percent oxygen. Table 4 repre- (c) Nuclear systems sents some physical properties of these mix- (4) Economic and Operational Factors tures.46 Physical properties of the individual (a) Development time inert gases have been summarized in part I11

TABLE4. -Properties of Candidate Systems, 540" R (80"F) [AFTER JOHNSON 46] I Molecular X, Btu/ Cp, Btul Atmosphere weight, m ' ft-hr-"R p, Ib/ft3 Ib-"R p, Ib/ft-hr x 10-3 I 14.7-psia air ...... 29 , 0.0151 0.076 0.24 0.0421 0.902 0.67 5-psia 02...... 32 .0154 .0283 .222 .0500 ,935 .72 5-psia 02-N~...... 31 .0153 .0268 .23 .0465 .935 .70 5-psia 02-He ...... 24 .0267 .0198 .278 .0520 1.572 .54 7-psia OZ-N~...... 30 1 .0152 .0362 .23 .0470 .926 .71 7-psia 02-He ...... 18 .0304 .023 .33 .0512 2.15 .496 I I 1 I 17 18 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS of this series and may be used for calculation volume of 2100 cu ft and a vehicle area of 860 of other physical properties of inert gas-oxygen sq ft designed for a mission of 1-year duration 4 mixtures as the need arises. with an oxygen-nitrogen system. This cabin is approximately twice the size of the proposed WEIGHT CONSIDERATIONS cabin of the US. Air Force Manned Orbiting . Laboratory. Structure of Cabin Wall It can be seen in figure 16 that the weight The weight penalty for the cabin wall must be penalty determined by flight loads is independent considered as a function of the thickness re- of the internal pressure of the cabin. The as- quired for safe maintenance of total internal sumed meteoroid criteria are also indicated. pressure as well as a function of the thickness With a 100-percent oxygen environment at 5 psia, required for safety against penetration. The it was considered that a criterion of no-penetra- specific mission in question is therefore a critical tion probability (Po) of 0.997 is required. The factor in this weight penalty calculation. The addition of an inert diluent reduces the cabin- greater the dynamic flight loads and the greater wall weight penalty most strikingly as oxygen the danger of meteorite penetration, the thicker content is reduced to 50 percent. Unfortunately, the cabin wall must be. The greater the prob- the criteria used for the allowable probability ability of penetration, the less tolerable is the of penetration versus oxygen percentage are not 100-percent oxygen environment. The less tol- stated. It would appear that curves of fabric erable a 100-percent oxygen environment, the burning rate versus total pressure and percent greater percent inert gas must be added, and the 02, such as presented in part I1 of this series, greater the internal pressure of the cabin. could be roughly used for generating such a One must obviously define the mission in relationship. In view of the difficulty in assigning question, the size and shape of the cabin, the hazard weighting for specific burning rate, the mission duration, and the specific cabin-gas penetration probability assigned to any burning system before an appropriate analysis of cabin- rate would have to be somewhat arbitrary.91 wall weight tradeoffs can be established. One The recent work covering zero gravity effects on must also define the meteorite environment or burning rates (see section on diluent gas in ch. 1) penetration rate. Figure 16 represents a sample suggests that at least from the point of view of calculation demonstrating these interactions.12 fire and blast hazard, the curve of allowable The basic assumptions are a vehicle with a cabin penetration probabilities in figure 16 may be shifted in the direction of higher percentages of oxygen. Because of the uncertainties regarding c 0 .997 % the absence of forced convection in the zero-g L -Q) studies to date, the degree of this shift is still cQ) n uncertain. 0 .988 In addition to this uncertainty, the relationship 0 ZI between structural weight penalty and prob- .86 .-.E ability of no mass penetration requires that nm .5 $ appropriate frequency-penetration equations be CL specified. Unfortunately, there is no indication 0 2 4 6 8 10 12 14 16 18 of the equations which were used. If the curve Cabin pressure, psi is indeed shifted much to the left, the flight loads may well tend to be the limiting weight factor 100- 50 40 30 23 at pressures closer to or even below 5 psia. Oxygen content, percent by weight It should be stated that meteoroid design criteria influence only very long missions. Puncture FIGURE16.-Structural weight considerations for a vehicle having a volume of 2100 ft3 and an probability is low enough for missions in the area of 860 ft' for a mission of 1 year and an 30- to 60-day class so that special meteoroid oxygen-nitrogen system. (AFTER BOEING.~~) protection is not required. ~ ~~

ENGINEERING CONSIDERATIONS 19 The curve for structural weight versus internal cabin which is a cylinder 120 inches in diameter pressure shows that the weight increase caused and 182 inches long; the meteorite penetration by increasing pressure does not become effective and dynamic-load factors were not c~nsidered.~~ until about 10 psia. The heavy black line indi- The problem of minimum gages and sizes for the cates the locus of weight-limiting factors and sug- cylinder headers and rings prevents the theo- gests that the minimum structural weight is retical value of zero weight for zero pressure from relatively constant at 6 to l0 psia and is deter- being attained. mined by dynamic loads. With less stringent criteria for meteorite protection and allowable Atmospheric Leakage percentage of oxygen, the minimum structural The leakage of atmosphere from the cabin weight may extend closer to 5 psia. It must also must be approached from two points of view. be remembered that the relative weighting of In the acute penetration of cabin wall by me- specific penalties will change as the size and teorites or other missiles, the large ratio of mission considerations are altered, but in general, orifice diameter to wall thickness implies a sonic minimum theoretical structural weight is con- orifice flow. The leakage may well be rapid and stant for pressures up to 6 or 7 psia. The pressure- the overall safety of the mission dependent on dependent weight penalty generally begins above the rate of leak. A slower leak may allow a greater 7 psia but, as in the above case, may not begin period of time for emergency procedures and until 10 p~ia.~~Fortunately, the minumum struc- possible crew survival. tural weight range of 5 to 10 psia also covers the Another consideration is the slow elastomer- physiologically ideal range of 5 to 7 psia. to-metal seal leak which controls the amount of While it is beyond the scope of this report to gas which must be taken on board to maintain present, in detail, the complete structural analysis constant atmospheric pressure over long periods of a typical cabin wall, it appears appropriate of time. In this case, capillary-type flows must to point out several factors which prevent the be considered. The following discussion is an weight penalties from being a simple linear func- extension of the more physiologically oriented tion of pressure. Figure 17 represents a study of review of the problem presented in part 111 of the partition of wall weight for a space station this series. In general, it is necessary to distinguish between five different types of flow: a~fi7 200 (1) Reversible-adiabatic choked flow (sonic 100 orifice flow or isentropic flow) 000 (2) Isothermal frictionless flow g. 900 Ring (3) Isothermal flow with friction I.- weight .-cn 800 (4) Isothermal free-molecular flow - 700 (5) Capillary flow 2 600 Reversible-adiabatic choked flow is the type E’ 500 Header of flow that exists in an orifice nozzle at or above 2 weight g 400 the critical pressure ratio. In this case, the flow at the throat or the minimum cross section is at 300 sonic velocity or at a Mach number of 1. Choked 200 Cylinder weight flow will occur in any leakage where the orifice 100 diameter is greater than wall thickness, pro- 0 vided the pressure ratio is critical or smaller. Internal operating pressure, psi Isentropic flow involves no change of entropy in the system. FIGURE17.-Effect of atmospheric pressure on Isothermal frictionless flow occurs when the pressure vessel weight. Conditions: 120-inch- diameter station, 182-inch-long pressure cabin, diameter of the passage is substantially less than 140-inch-radius headers, and 2219-TtS2 alumi- the length of the passage. The idealized “fric- num material. (AFTER PARKER ET AL.’O) tionless” flow forms the upper limit of flow rates 20 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS where the maximum velocity in the duct is In general, two emergency situations may l/(vc)0.5 times the sonic velocity, or 0.845 times occur which require attention. A cabin puncture * sonic velocity. Where great friction is involved may be small enough to give a slow pressure through small holes or porous media, the flow transient and expose the crew only to the prob- rate is a function of both the passage dimensions lems of decompression sickness. This may in- . and the friction factor and does not lend itself volve isothermal decompression, in that enough to a simple limit. time may be available for thermal equilibration. Isothermal free-molecular flow occurs when the However, even a relatively slow emergency leak critical hole diameter is so small that the mole- from a space vehicle may be isentropic or adia- cules appear to act individually and not as a batic. In a zero gravity environment with no part of a flow stream. In this case the velocity is natural convection, the heat-transfer rate may (8/7ry)O.5(~+l/2)O.5= 2(y+ l/7ry)"5= 1.48 times be inadequate to maintain constant temperature the same velocity found for adiabatic choked and cause adiabatic conditions, although fan flow. However, the area to be considered is circulation should be present. One must also 1/4/(2/y+ l)l/Y-l= 0.395 that for adiabatic choked consider explosive decompression (time con- flow. The actual flow rate is therefore 1.48 stants of less than 0.1 sec) where the transients x 0.395= 0.584 times that for adiabatic choked are fast enough to allow the temperature to vary flow. adiabatically or isentropically with the pressure. Capillary flow is similar in nature to the iso- The physiological implications of these situations thermal free-molecular flow and involves flow have been covered in part I11 of this ~eries.9~ through long capillary passages. The intake Quotient rule differentiation of the ideal gas flow is of laminar continuum type with a transi- law (PV=n,RT) with respect to time yields tion to the free-molecular flow at the zero pressure end. The problem of choice of an equation for flow through porous or capillary passages has been discussed in part I11 of this series.92 There is little empirical data to support any single one. and solving for P, the pressure change in a com- Mason has concluded that a modification of the partment undergoing any form of decompression Knudsen equation appears to best approximate would be this type of flow. This equation may well hold p=- ntRT Pi' for orifices of diameter less than 25 percent of cabin-wall thickness. The equation is presented v +T below in discussion of gas tradeoffs for slow seal-leak conditions. The mass leak rate would be expressed by

EMERGENCYRAPID GAS LEAKS ~L = PCd & &fdy)lb/sec (17) Because of the multivariate conditions of gas leakage, it appears appropriate to review the where development of generalized equations covering those factors pertinent to the problem at hand- gas specific leakage tradeoffs. These may be used to calculate leak rates for specific variables in question. In terms of moleslsec, The theory of sudden rapid leakage under emergency conditions has been reviewed in great detail by Coe and his coworkers.20 Only a general review of the concepts and pertinent equations required for tradeoff studies are pre- Isothermal Decompression sented here. If the decompression takes place isothermally, ~ ~ ~~

ENGINEERING CONSIDERATIONS 21 T=To and 7'0. Then the equation (16b) 1.0 becomes 0.8

0.6 n.0 -n. 0.4 For the purposes of the present analysis, one can assume that the leak rate is always much 0.2 greater than the feed rate and the rate of oxygen consumption. This would be certainly true if 0 gaseous feed from storage can be immediately 0 1 2 3 4 cut out as soon as a sudden pressure drop occurs. UT Substituting equation (19) into (20) leads to a FIGURE18.-Generalized decompression curves for rate of pressure drop space compartments. (AFTER COR ET AL.~O)

metabolic water and COZ production, and oxygen -vp=---' RToFL- 223 (y)J:h(y)P lblft2lsec consumption are not neglected may be found in the analysis of Coe ,et a1.20 These complicated additions to the problem are beyond the scope If of the present study and, except for feed gas, would have little effect on the decompression - curve for acute emergency situations. a=223 ('$) ,/: fi(y) sec-' (22) It can be seen from the isothermal curve and the equations (17) and (19) that the rapid, isothermal, mass-leak rate is an inverse function of the the rate of pressure change becomes square root of the average molecular weight of p=- the gas mixture and the molar leak rate is a direct CWP (23) function of the square root of the average molecular weight. By integrating equation (23), it is seen that at any given second T the ratio of pressure to original Adiabatic Decompression pressure is: If the decompression occurs adiabatically,

and and the time to reach any given pressure ratio PIP0 is:

By substitution of equation (26) and (27) and rearranging, it can be seen that

The upper curve of figure 18 presents a generalized isothermal decompression curve using relationships of equations (22) and (25) with the units To="R, A=ft2, V=ft3, a=sec-', and T= sec. Again assuming no feed into the system during The isothermal decompression curves for the decompression and substituting equation (26) more specific situation where feed gas input, into equation (19) gives: 22 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

Substituting equation (29) into equation (28) with the assumption that there is no feed into the system

r 1

1.0 1.2 1.4 1.6 1.8 where cr is defined by equation (22). Integrating Y equation (31): FIGURE19.-specific heat function. (AFTER COE PV ET AL.~O)

be at play. The interaction between the rapid so that leak rate and an active storage system or gas regeneration system is quite complex both from the point of view of pressure stabilization and weight penalty required to cover the open-feed decompression emergency. It is not felt that the present discussion warrants such a detailed analysis. It is also felt that the complexities of Figure 19 represents a plot of fi(y)versus y as pressure caused by the presence of a fire during indicated by equation (18). accidental or deliberate decompression precludes Assuming specific values of y, generalized the addition of this concomitant event in the adiabatic decompression curves can be plotted tradeoff analysis. The contribution of tempera- using the relationships of equations (22) and (33) ture rise due to heat liberation, oxygen consump- and figure 19, with the units: To="R, A=ft2, tion, and total pressure change due to imbalance Y=ft3, a= sec-1, and T= sec. Figure 18 shows between moles of combustion products formed such a family of curves for isothermal conditions. and moles of reactants consumed have been It can be seen from figure 18 that at low values theoretically analyzed by Coe et a1.20 The reader of (YT,the isothermal pressure drop is slower than is referred to this study for further details. adiabatic, but as CYT increases, the adiabatic What are the gas-specific factors involved in curves level off more rapidly. The molar rate of the fast leak situation? Since the amount of time flow is again inversely proportional to the square required for a cabin to reach physiologically root of the average molecular weight of the gas borderline levels of por is a major consideration mixture (eq. (29)) and is related to the ratio of in fast emergency leak conditions, it would be specific heats of the gases by the complex y and appropriate to compare the time it would take the fi(y)relationship of this equation. several proposed gas systems to reach this The interaction between the human lung and endpoint. explosive decompression of the cabin has been The Boeing Co. has made calculations pre- reviewed in detail in part I11 of this series.92 dicting these critical times.12 Figures 20 and 21 The same general gas-specific factors appear to represent the graphic presentation of variables ENGINEERING CONSIDERATIONS

1.0 .9 .a .7

- .1 0. 1 0.002 0.003 0 0.001 0.002 0.003 rC A sec-ft2 A*- v ft3

FIGURE20.-Zsothermal decompression. (AFTER FIGURE21.-Zsentropic decompression. (AFTER BOEING .I*) BOEING.~~) required for these calculations. From equations --rCdA - 0.000575 (7)sec ft2 (22), (25), and (34), it can be seen that a logarith- V mic plot of P/P" versus rCdA/V will give curves from which the time can be simply determined. and These relations are plotted for the isothermal condition in figure 20 and the isentropic or 0.00575 X 770 - r= -325 sec adiabatic condition in figure 21, with r= seconds, T X (0.25)2 A = ft2, and V= ft3. These represent five specific 144 cases of the generalized decompression curves of figure 18. In order to calculate the physiologi- The time to reach minimum tolerable partial cally critical times, the minimal tolerable total pressure of po2 can be calculated by the factors pressures needed at any percentage of oxygen of figure 1. This reduces the available time in the atmosphere can be obtained from figure 1. considerably. A sample calculation can be shown for isother- Table 5 represents the time in minutes required mal flow using figure 20. For a hol; '12 inch in to reach 3.5 psia for 'In-inch and 3/4-inch holes diameter, an orifice coefficient (Cd) of 1, and a under isothermal and isentropic conditions with cabin volume of 770 ft3, the time to reach 3.5-psia five proposed atmospheres. Table 6 represents total pressure from 0.5-psia oxygen can be the times required to meet minimum tolerable determined from figure 20 by using the ratio total pressures as determined by minimal poz 3.5 to 5.0 or 0.7 to give: levels. It can be seen that in all cases, the 24 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

TABLE5. -Decompression Time to 3.5 psia [AFTER BOEING 121 [CABIN VOLUME = 770 FT3; ORIFICE COEFFICIENT = 11

I I I I I

3.5 psi8 02 3.5 psia 02 3.5 psia 02 3.5 psia 02 5.0 psia 02 3.5 psia Nz 3.5 psia He 1.5 psia NZ 1.5 psia He Leak mode 7.0 peia 7.0 psia 5.0 psia 5.0 psia I 1 I Decompression time, rnin

Isothermal- '/,-inch hole...... 10.0 7.75 5.3 4.59 5.42 Isentropic - '/,-inch hole...... 7.64 5.47 3.9 3.25 3.95 Isothermal - 3/4-inch hole...... 4.5 3.44 2.36 2.04 2.41 Isentropic - Yd-inch hole...... 3.4 2.43 1.73 1.45 1.75

3.5 psia 02 3.5 psia 02 3.5 psia 02 3.5 psia 02 5.0 psia 02 3.5 psia Nz 3.5 psia He 1.5 psia Nz 1.5 psia He Leak mode 7.0 psia 7.0 psia 5.0 psia 5.0 psia

Decompression time, min r Isothermal- '/+inch hole ...... 4.72 2.25 1.93 5.42 Isentropic - Yr-inch hole...... 3.22 1.62 1.35 3.95 I~othermal-~/4-inchhole...... 2.1 1.0 .86 2.41 Isentropic - Ya-inch hole...... 1.42 .72 .59 1.75 I I oxygen-nitrogen mixture at 7 psia takes the operational significance between the maximally longest times and 100 percent oxygen takes the divergent times of 2 minutes and 0.6 minute for shortest time to reach the critical condition. the 3/4-inch hole with isentropic flow. If the The larger the hole, the less the absolute dif- mission requires at least 2 minutes for donning ference between mixtures. There is no difference an emergency suit in a high-risk mission, this between the times for either endpoint criterion difference may well be critical in the selection. for 100 percent oxygen at 5 psia. The lower the The major probability of a penetration producing pressure of inert gas, the less time required to such a hole size is obviously a major mission- reach both endpoints and the greater the dif- specific factor to be considered. ference between the two criteria. From the point There are several other minor considerations of view of the human subject, table 6, of course, in the area of fast-flow systems. These are the presents the more valid endpoint. At equivalent maximum airlock dumping and repressurization composition and pressures, nitrogen has a slight times during extravehicular operations and the advantage over helium. maximum rate of cabin pressure dumping during The relative weighting of this factor in the fire emergencies. The dumping of airlock and overall tradeoff analysis is discussed in chapter cabin would, of course, follow the more isentropic 3. The major question raised at this point is the type of flow. In the present case, the faster the ENGINEERING CONSIDERATIONS 25 flow through the maximum orifice available, the (34) that a lock of 40 ft3 can be isothermally pres- more advantageous the gas mixture. One would surized by air to 99 percent of the main compart- therefore have to weigh the advantage of having ment pressure through a valve of only 0.58 in.2 in a more rapid dumping capability for a suited crew 30 seconds. Doubling the area of the valve can against a less-rapid emergency dumping after reduce this time to about 15 seconds. Since the accidental puncture with an unsuited crew. time required is proportional to the square root The repressurization of an airlock from a of the molecular weight, substitution of air vacuum to the pressure of the main compartment (molecular weight = 29) by the proposed mixture is most rapidly accomplished by opening a valve of lowest molecular weight, helium-oxygen mix- between the two chambers. In most cases, the ture at 7 psia (molecular weight = 18), will reduce pressure and temperature of the main compart- minimum compression time by only a few sec- ment is maintained constant by the gas feed onds. For larger lock systems, the number of system and the compression will be close to seconds to be saved will increase as will the isothermal because of the great flow turbulence physiological significance of the savings. How- in the airlock. The flow across the valve starts ever, the valve size can be increased to meet off as a supercritical pressure ratio and then this demand in a large lock. becomes subcritical when One must also consider the airlock pumping weight penalties. The airlock may be pumped into a separate storage tank or into the main PclPlk = y (L)y+l y-1 compartment. The effect of atmosphere composition on this penalty is currently under study at The approximate time required to recompress Douglas.101 Data are also available on a new a lock isothermally from a vacuum can be deter- elastic recovery principle in the design of mined for air of y= 1.4 by the equation airlocks.'6

SLOWSEAL LEAKAGE (34) The selection of appropriate equations for the description of slow leaks through elastomer-metal This aspect of a space mission will be critical seals is a significant factor in calculation of weight only when a crewman must be retrieved most penalties for gas storage system. In part I11 of rapidly through a lock to a cabin. Since the this series, a preliminary analysis of the problem relatively small volume of the lock suggests that by Mason of AiResearch was discussed.92 More the minimum time for recompression will not recently Mason has expanded his calculations in any practical way limit the survival potential to include analysis for neon and other variables. of the crewman, the effect of atmospheric com- It appears appropriate to review these calcula- position should have little practical effect on the tions and compare them with similar unpublished survival. The difference in time, measured by calculations made by R. K. Moir and J. R. Malcom seconds, which will be given the entering crew- of the Boeing Co.12 man by an optimum gas mixture does not appear The most divergent equations for seal-leak to warrant a thorough analysis of the problem in calculations are those for isentropic sonic-orifice the present context. A general analysis of this flow described above for large holes and those problem is presented by Coe et a1.20 That the for capillary-free molecular flow. Mason's recent gas-specific factor will probably not be critical modification of the Knudsen equation for capillary is indicated by their calculation from equation flow of laminar continuum to free molecular transition at a final pressure of zero is

5.22 D4Pt2+ 7.42 D3P' 7.44 D2p'T'] ( + 23.9 DP' - [ [In E)] (35) q= losp'L 106 L M'+ 108 M'L PI

261-559 0-67-3 26 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS where TABLE7. -Capillary Size and Quantity Required 1.0 lblday Leakage From 5-psia Oxygen pressure x volumetric leakage rate, micron- for q Atmosphere [AFTER MASON ET AL.67] 1 liters/sec D capillary diameter, microns Capillary Number of capil- Number of capil- P' cabin pressure, 14.7 psia diameter, laries of 1-mm laries X diameter, F' viscosity, 1.81 x 10-4 poise microns length microns L capillary length, cm temperature, 527.7" F 2" 0.01 1.o x 1014 3.3 x 1w M' molecular weight, 28.97 .1 1.0 x 10" 3.4 x 1oi .3 3.9 x 109 3700 The number of capillaries required to achieve 1 .o 9.7 x 107 320 a 1 lb/day leakage rate with capillary flow of 5 3.0 2.7 X 1W 27 psia oxygen is shown in table 7 for several 10.0 3.9 x 104 0.06 30.0 5.9 x 102 0.06 holes of 1-mm length. Since the number of cap- i ilaries multiplied by the diameter represents the possible number of leakage paths along spacecraft seal perimeters, it was pointed out that the rate of oxygen at 5 psia; isentropic sonic orifice most probable hole size is in the range of 1 to 10 and capillary flows of the type represented by microns. equations (32) and (35), respectively, are Table 8 represents a comparison of equivalent compared. leakage rates for several physiologically adequate Table 8 shows that the flow type is not critical gaseous mixtures based on a 1 lb/day leakage and that helium leakage is not as excessive as

TABLE8. -Comparison of Capillary and Orifice Leakage Models [AFTER MASON ET AL.67] [BASIS: 1 LB/DAY LEAKAGE FROM A 5-PSIA PURE OXYGEN ATMOSPHERE]

Leakage, lb/day, at pressures of-

5 psia 14.7 psia

Capillary Orifice

Oxygen ...... 1.54 1.29 Helium...... 62 .50

2.16 1.79

Oxygen ...... 0.76 1.10 0.87 Neon ...... 2.23 1.73

Total ...... 3.33 2.60

Oxygen ...... 1.33 0.73 Nitrogen...... 3.74 2.00

Total...... 98 .98 5.07 2.73 -.____ ENGINEERING CONSIDERATIONS 27 often predicted. There is actually little difference For the range of capillary orifices and cabin between the weight of these mixtures lost per atmospheres most suitable from a physiological day in the 5- to 7-psia range. For orifice flow, point of view, the calculated leak rates are the leakage rate is nearly proportional to pres- indicated in table 9. This table also employs sure. At pressures less than 7 psia, the same is as a basis a leak equivalent to 1 lb/day of 02at true for capillary flow. As the pressure and 5 psia. The number of holes required for this molecular weight increase, equation (35) suggests !eak rate through c.pi!!aries of 1-mm length was that the leakage rate becomes proportional to determined by table 7 and the leakage rates for the square of the pressure. other gases were evaluated. Only at the higher

TABLE9. -Leakage Rates for Various Atmospheres [AFTER MASON ET AL.67] [BASIS: 1 LB/DAY LEAKAGE FROM A 5-PSIA PURE OXYGEN ATMOSPHERE]

Capi Diluent, helium Diluent, neon Diluent, nitrogen lary Po, 3 dian mm Hg 180 180 180 180 180 180 180 180 180 180 180 180 PDiluent, eter 79 200 338 580 79 200 338 580 79 200 338 580 mi- mm Hg PTotal? cron mm Hg 259 380 518 760 259 380 518 760 259 380 518 760 (5 psia) (7.35 (10 psia) (14.7 (5 psia) (7.35 (10 psia) (14.7 (5 psia) (7.35 (10 psia) (14.7 psia) psia) psia) psia) psia) psia)

0.01 12...... 0.814 0.947 1.061 1.204 0.739 0.775 0.798 0.818 0.710 0.718 0.723 0.725 Xluent.. .. .045 .132 .249 .485 205 .543 .944 1.662 .273 .699 1.188 2.045

Total..

0.1 12...... Iiluent ....

Total..

0.3 12...... Xluent ....

Total..

1.o 32...... Xluent ....

Total..

3.0 3.2 ...... Iiluent ....

Total..

10.0 32...... Iiluent ....

Total..

30.0 12...... 0.713 1.016 1.356 1.949 0.654 0.853 1.071 1.443 0.718 1.039 1.407 2.055 Xluent.. .. .039 .141 .318 .785 ,181 .398 1.268 2.933 .276 1.009 2.311 5.794

Total.. ,752 1.157 1.674 2.734 ,835 1.451 2.339 4.376 .994 2.048 3.718 7.849 28 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS pressures does the hole size appear significant. flow equation used by the Boeing group was not The gaseous composition does not appear to specified. The Boeing approach was to also be very significant for any hole size in the 5 to 7 assume a given hole size and calculate the num- ' psia range. For the smaller holes, nitrogen has a ber of holes required to leak 2.0 lblday of oxygen- slight advantage; for the larger holes, neon. In nitrogen mixture at 7.0 psia. Mason et al., it the 1- to 10-micron range predicted as most should be remembered, used 1 lblday of 5.0 probable by table 7, there is virtually no dif- psia 02as a basis (table 7). Since the physical ference between the inert gas used. Figure 22 flow properties of oxygen and nitrogen are similar, summarizes these data graphically for a capillary the bases differ only in original pressure (7 3 microns in diameter. Pressure is the major versus 5 psia) and leak (1 lb/day versus 2 lb/day). determinant of mass leakage through seals. A major difference was the assumption of a Unfortunately, there is little empirical data to Y4-inch (6.3 mm) instead of the 1-mm path length corroborate these predictions. Mason reported of the previous study. that preliminary work with a cabin simulator Figure 23 represents the plot of hole diameter suggests little difference between helium and versus number of holes required for the assumed nitrogen in leakage through inflatable elastomer leak. It can be seen that for hole sizes of 30 to metal seals.= The variation in leak rate from microns, 4 X 103 holes are required to support run to run was too great for any definite con- the model leak with capillary flow. Comparison clusion. Only when vacuum grease was applied with table 7 indicates that under the specific to the inflatable seal was there any indication assumptions used in the Mason study, 5.9 x lo2 that the molal leak rates for a 50-percent oxygen holes are required, or only 0.15 the number of and 50-percent nitrogen mixture and a 50-per- holes required under the Boeing assumptions. cent oxygen and 50-percent helium mixture were The Boeing group discarded orifice flow as a similar. Mason feels that large mission-to- valid model because the calculated hole sizes mission variation in hatch-seal leak rates may be were so small that it made the orifice assumptions anticipated in operational vehicles. invalid by virtue of the excessive ratio of path In the Boeing studies of leak rate through seals, length to diameter. Compressible pipe flow was simple isentropic orifice flow, compressible pipe discarded for small hole sizes because the pres- flow, and capillary flow were compared. Because sure drop was assumed to be so great as to make the calculations were approached in a somewhat the flow molecular in nature. Using figure 23 to different way from those of Mason et al., they calculate the number of holes and the hole diam- appear worthy of review. The capillary and isen- eters that will leak 2 lblday of oxygen-nitrogen tropic equations which the Boeing groups used mixture at 7 psia, the mass leakage for the other were the same equations used by Mason et al. atmospheres was determined assuming capillary Unfortunately, the specific compressible pipe flow as indicated by equation (35). Figure 24

lo00 I I I I1 I1

$1- I I Orifice\

=%1 1 10 100 lo00 loo00 loow0 Cabin pressure, psia Number of holes

FIGURE22.-Leakage rate as a function ofpressure. FIGURE23.-Variation of hole diameter with Basis: 1 lblday leakage for 5-psia oxygen atmos- number of holes required for the assumed leak. phere. (AFTER MASON ET AL.~T) (AFTER BOEING.~~) ~

ENGINEERING CONSIDERATIONS ' 29 10 I I I I I mixtures appear to be slightly more favorable I (6 7 psia0,-He 7I psia 0,-N, I than the other mixtures for the 3-micron-diameter I I\ P hole under consideration. =: =: gl Recent advances in sealing technology in I 1 '55 psia 0O,iN, -N and Fpsii5 psia 0, 5 psia 0,-He spacecraft design have been reviewed.lO7 I These principles should be brought to bear on I I I I I the problem. 1 10 100 loo0 loo00 Number of holes Tankage for Gas

FIGURE24.--Sensitivity of leakage to number of The tradeoffs for gas tankage appear to be holes and atmospheric composition for rates equivalent to 2.0 lb/day of 7.0-psia oxygen- most sensitive to differences in spacecraft con- nitrogen mixture. (AFTER BOEING.~~) figuration and mission plan. This arises from the dependence of storage efficiency on the size represents the results of these calculations. and shape of the container, be it for gaseous or Even with the greater capillary length of the cryogenic systems. There are also major ques- Boeing study, it can be seen that the relative tions regarding the basic weight penalties in leakage rates of the different atmospheres is cryogenic tankage for small liquid helium and not affected by the assumption of hole size and neon systems. In the following section an hence number of holes. attempt will be made to define the knowns and Table 10 compares the mass leak rate through the unknowns in storage systems and indicate capillaries of 3 microns in diameter for approxi- how they may condition the several tradeoff mately similar mixtures and pressures calculated analyses in the subsequent sections of this by the two groups using the different basic report. assumptions discussed previously. The leak rates GENERALCONSIDERATIONS appear quite insensitive to the different capillary lengths under study (1 and 6.3 mm). This remark- The storage of gases for atmosphere control able agreement may have fortuitously arisen systems may be classified as (1) high pressure from the interaction between the slightly dif- gaseous storage at ambient temperature, (2) ferent pressures and compositions being studied cryogenic storage at low or moderate pressures, and the differences in path length. In any case, and (3)for oxygen and nitrogen, storage in chemi- these mass leak rates appear to be adequate for cal form. Each of these approaches will be pre- a first-order analysis of the weight tradeoffs sented with the minimum detail required for of the different gas systems. The oxygen-neon tradeoff analyses.

TABLE10. -Comparison of Mass Leak Rates

Mass leak rate, Ib/day, at pressures of- I 5 psia I 7 psia Study

100 percent 50 percent 0250 percent 02 02 50 percent Ne 50 percent N,

1.0 0.811 1 0.702 I ;:it3 1 1.13 0.810 1.70 Mason et al.67 1.05 .76 ...... 1.08 ...... 2.0 Boeing.12 1 .o ...... 1.90 Boeing'z normalized to 5-psia 02=1 lb/ day. 30 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS HIGH-PRESSUREGASEOUS STORAGE The above relation simply states that the per- The basic role of gaseous storage systems centage change in overall system penalty X is appears to be that of supplemental storage or the weighted sum of storage vessel weight and storage for repressurization of the cabin when volume percentage changes. Such a relation long-term storage prior to use makes it more holds true over at least a small system size range, . efficient, especially in smaller cabin systems. with the weighting factors a and b determined The need for high delivery rate in repressuriza- by the type of vehicle geometry and structure tion also favors gaseous storage. A comparison considered. System optimization then involves with liquid systems under these conditions is minimization of X, where presented subsequently. The basic problem in this approach to gas x = WgvTb (36b) storage is the minimization of container volume As can be seen, the relative importance of penalties by the use of elevated storage pressures storage vessel weight and volume is expressed without incurring excessive pressure shell by the weighting factors a and 6. Thus, for a case weight. It can be shown that if the fluid stored where b= 0, weight is all important, while volume acts like an ideal gas, the weight of container is all important for the case a=0. For a=b, designed to hold a given charge is essentially system optimization involves minimization of independent of pressure while container volume the product of weight and volume, WTVT.Storage is inversely proportional to pressure. Very-high- vessel data given in the following section cover pressure storage appears to be the ideal goal. these three cases and also include the cases of However, gas compressibility factors begin to minimum WTGT and VT fl (intermediate limit the weight efficiency of storage. At pressures relative importance levels). above several thousands of pounds per square The optimization of weight and volume param- inch, gases become less compressible. The de- eters to be discussed is taken directly from the crease in compressibility is less serious for study of Coe et a1.20 The physical basis for the helium and neon than for oxygen and nitrogen.46 tradeoff data may be found in this study. In all Thus as pressure is increased, overall vessel cases, it is assumed that the nominal fill tem- volume passes through a minimum and actually perature of the vessel is 530" R and maximum increases because of overall shell-wall thickness. fluid use is 620" R. Storage fluid end pressure Pressure-level optimization studies for oxygen is 30 psia. storage vessels conducted by Jacobson,43 The gas compressibility factors for oxygen Keating?' and Coe et a1.20 indicated an optimum were computed from experimental pressure- storage pressure of 7500 psia for equal pressure volume-temperature data for nitrogen, assuming and volume criteria. This level was used in the the law of corresponding states, the accuracy of Project Mercury system.75 Optimum storage the basic nitrogen data, and the close similarity vessels for pressure up to 9000 psia are currently of the two gases. Container structural analyses under study by several companies.= These were given for simple geometries and were based vessels will be of greater value for helium and upon the assumptions of true geometrical shape neon where compressibility factors play a lesser and of a low ratio of wall thickness to diameter. role. It is to be emphasized that more detailed analyses If the rough sizing of a vehicle volume is than those presented would be required to op- available, the tradeoff between storage weight timize structural design in a specific application. and volume can be made for any vehicle design. Particular attention would have to be given to The total storage system penalty effects are vessel mounting requirements. estimated by using the relation Oxygen Figure 25 represents the variation with nominal charge pressure of the total weight and volume of spherical oxygen vessels for SAE 4340 ENGINEERING CONSIDERATIONS 31 .3

0 m- .1 I - %= .07 r2 2 6 Q .G5 5g 4: c % m .03 3g -E, s .02 2

.01 1 1 2 3 4 56 810 20 30x10~ Nominal fill pressure, psia

FIGURE25.-Weight and volume of spherical oxygen storage vessels for safety factor of 1.88. (AFTER COE ET AL.ZO) steel and titanium alloy C120 AV Ti. The safety of titanium pressure vessels for oxygen storage FIGURE26.-Optirnization of spherical oxygen has been questioned in part I1 of this seriesF1 storage vessels. Material is SAE 4340 steel. but will be included to show the weight savings. (AFTER COE ET AL.~O) A fatigue failure criterion with a safety factor of 1.88 was used in figure 25. sure level for spherical steel oxygen vessels. As discussed above, the weight and volume The use of these factors is explained by equa- penalties show distinct minima. Minimum weight tions (36a) and (36b). The effects of weight and occurs at approximately 2500-psia charge pres- volume weighting factors on optimum charge sure, indicating the deleterious effect of charge pressure level can be seen by comparison of temperature tolerances on fluid load penalties figures 25 and 26. It should be noted that the at low charge pressures. Minimum vessel vol- inclusion of unusable fluid weight penalties ume occurs at a charge pressure of approxi- results in a higher optimum charge pressure mately 20 000 psia for the steel vessels, showing (approximately 10 000 psia) than that computed the effects of increases in vessel wall thickness by Keating.50 (See table 11.) at higher charge pressures, as well as the increasing compressibility factor for the gas under Nitrogen these conditions. From other calculations, Total weights and volumes of spherical nitro- it appears that the pressures at which the weight gen storage vessels are shown in figure 27 as and volume are minimum are apparently inde- functions of charge pressure level for the two pendent of the safety factor used in the design. cases studied. Titanium was used as the vessel However, the actual values of the weight and material for nitrogen. The results are generally volume are directly related to the safety factor. similar to those obtained for oxygen, showing Similar data for Inconel 718, stainless steel 301A minima in vessel weight and volume in the pres- (cryogenically stretched by Ardeforming), and sure range studied. Similar data covering pres- Ti 6A 6V 2s may be found in figure 7-15 of ref- sure up to 3500 psia for Inconel 718, stainless erence si. steel 301A (cryogenically stretched by Arde- The terms WTVT, WT a,and VT % are forming) and Ti 6A 6V 2s may be found in fig- shown on figure 26 as functions of charge pres- ure 7- 15 of reference 51. 32 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

e *-- .lo 10 3= -3 s+ .OB > 82 g .M 6s c 5z .M 4g E .- 2 .03 3s g .a 2

.01 Sl 1 2 3 456 810 20 30x103 Nominal fill pressure, psia

FIGURE27.-Weight and volume of spherical nitrogen storage vessels. Material is Ti C-120 1~10~ AV. (AFTER COE ET AL.~~)

Figure 28 shows the terms WTVT, WTfi, and FIGURE28.-Optimization of spherical nitrogen VFfl for spherical nitrogen vessels as func- storage vessels. Material is Ti C-120 AV. tions ofi charge pressure level derived from equa- (AFTER COE ET AL.~O) tions (36a) and (36b). Here, the optimum charge pressure for minimum WTVT is approximately and on the installation. These weights, in gen- 8000 psia in the case considered. eral, are small; an allowance for accessory weight Table 11 summarizes the optimum values of should be made, however, in the total vessel weight and volume for oxygen and nitrogen weight. vessels. It should be noted here that the weights Helium plotted in figures 25 to 28 and table 11 do not include the weight of the lines, brackets, or The weights and volumes of spherical helium valves; an allowance should be made for these vessels are shown in figure 29 as functions of accessories. The valve weight depends only on charge pressure level, using titanium as the the number of vessels and on the number of pressure shell material. These data are limited valves installed on each vessel for redundancy to pressures below 6000 psia because of the and for installation requirements. (See table 19.) lack of higher pressure-density data. The tend- Mounting bracket design depends primarily on ency of helium to diffuse through the metal may the size of the vessel, on the number of vessels, well limit the usefulness of higher pressures. Compressibility is not a factor with helium. 11. -High-pressure Gas Storage Opti- TABLE Neon mum Design [AFTER ROUSSEAU ET AL.95] In the pressure range studied, the compres- Parameter sibility of neon appears to be the same as that of helium, both acting quite close to that of an Optimum pressure, psia ...... 10 500 ideal gas.44 Since the density of helium at 0" C Weight penalty, W,lW ...... and 1 atm is 0.178 gm/l and that of neon 0.899 Volume penalty, VTIW,,,ft3/lb ...... Optimization criterion...... gm/l, the volume per pound of useful load should be reduced by a factor of about 5 and run parallel ENGINEERING CONSIDERATIONS 33 2.0 CRYOGENICSTORAGE General Considerations 1.5 The cryogenic storage of fluids offers several distinct advantages over high-pressure storage 1.0 100 of the low boilingpoint fluids such as oxygen and c1 nitrogen. These advantages are a higher fluid mz .8 b 70 zz= storage density at low to moderate pressure, 'F- reduced container weight per unit of stored mass, s= .6 s >+ provision of potential refrigeration or cooling

~ .5 50 sources as heat sinks (170 Btu/lb for liquid oxygen m n0)c 5 .4 or nitrogen when heated to room temperature). n E E .-m The major defects are the sensitivity to un- -3 .3 30 5 expected heat leaks and the complexity of de- 0 > livery in zero gravity. These defects require special attention to insulation needs, single- .2 20 phase fluid expulsion, phase separation for venting, and quantity measurement. Cost, develop- 15 ment time, servicing equipment, standby penalties, and limited expulsion capability are 0.1 10 other disadvantages. 1 2 3 4 5 6.X103 Two major classes of cryogenic liquid storage Nominal fill pressure, psia systems are used. They specify either mode of FIGURE29.-Weight and volume of spherical storage or method of pressurization. The fluid helium storage vessels. (AFTER COE ET AL.*~) may be stored as a single phase of fluid or as a two-phase mixture of fluid and vapor requiring to the upper curve of figure 29. Similarly, the total special separation techniques. The pressuriza- vessel weights per pound of useful load should tion may, in turn, be accomplished by use of also be reduced by a factor of 5 and run parallel externally supplied gas or by thermal energy to the lower curve of figure 29. A less expensive added by means of electric power or a heat mixture of 85 percent neon and 15 percent helium exchanger in the storage space. may be economically more feasible than pure Because weight tradeoffs are quite sensitive gaseous neon?2 to the specific form of cryogenic storage involved, it appears appropriate to renew briefly Mixed Gas Storage for the reader not versed in cryogenics, the sev- The availability of mixed gas storage in one eral forms of storage systems available. For container for repressurization purposes appears further details, it is suggested that the reader to be a great advantage of high pressure gaseous refer to the discussions of Vance,'os Cook,2l systems. This system is indeed attractive only Coe et a1.,20 and Christian and Hurlich.18 The for this purpose since the requirement for stable following discussion is taken directly from the use of both constituents precludes its mainte- Coe paper. nance use in cabins where unavoidable erratic The following three types of systems appear leaks occur. Even in the event of constant-leak to be most commonly suggested for zero-gravity systems, the mixed gas form alone is not suitable space cabin use: for cabins where crew occupancy or workload (1) Supercritical single-phase thermal pressuri- can vary from time to time and no parallel con- zation trol of leak rate is feasible. Because mixed gas (2) Subcritical single-phase helium bladder storage is limited to repressurization systems, a expulsion thorough weight penalty analysis does not ap- (3) Low-pressure two-phase vapor or liquid pear warranted in the present context. delivery 34 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS Supercritical single phase. -The operation of this type of storage vessel, which avoids zero- Fill gravity phase separation problems, is shown on a schematic fluid temperature-density diagram on Ground heater .Heating figure 30. Figure 31 shows the thermodynamic coil process operation on a schematic pressure- enthalpy diagram and illustrates one method of reducer tank heat addition. Tank fill conditions are indicated by point 1 on figures 30 and 31. Here, the storage vessel is assumed incompletely filled with liquid at atmospheric pressure. The storage fluid is thus 1 to 3: lhitial pres E a mixture of saturated liquid and vapor. After 2 filling, the tank is capped off. Heating before use a3el t thus results in pressurization at constant average density. If pressurization continues past point 2 in figures 30 and 31, the storage fluid becomes homogeneous, acting as a compressed liquid. In practice, the tank is heated prior to use until Enthalpy the storage pressure is higher than critical (point 3). Tank temperature rises slightly during FIGURE3l.-Pressure-enthalpy diagram for ther- this process, but is below critical at point 3. mally pressurized supercritical storage. (AFTER Fluid delivery starts once supercritical pres- COE ET AL.*O) sure is reached, with pressure being maintained by adding heat to the storage space. Constant figures 30 and 31. Here, as long as supercritical pressure operation is indicated by path 3-4' in pressures are maintained, the storage fluid remains homogeneous. This type of system is thus ideal for the zero-gravity storage of fluid 4 mixtures. As shown in figures 30 and 31, fluid temperatures rise during operations. When the vessel is almost empty, the storage fluid is a compressed gas. Pressure may thus be allowed to fall at the end of operation without incurring liquid dropout, 0- as shown by path 3-4'. Operation on path 3-4' e L0 permits heating requirements to be relaxed dur- E ing the last phases of delivery. The resultant E lower final density reduces the quantity of fluid which cannot be used. One method of tank fluid heat addition is illustrated in figure 31. Here, the delivery fluid is I first heated to ambient temperatures by, for example, heat exchange with cabin air. The waim, high-pressure fluid then passes through Density a valve which senses and regulates tank pressure. When tank pressure falls below the regulated FIGURE30.-Temperature-density diagram for value, the warm gas is directed through a heat thermally pressurized supercritical storage. For P>P,, fluid is single phase regardless of T; exchanger in the storage space, where it provides for T>T,, fluid is single phase regardless of P. energy for pressurization. This gas is then (AFTER COE ET AL.~O) reheated before use. In practice, the pressure ~~~ ~ ~~~

ENGINEERING CONSIDERATIONS 35 control valve usually acts as a flow modulator, High-pressure heI i u m bottle directing a portion of the warm gas to the tank Helium heat exchanger at all times. relief Storage quantity determination is simplified +@-c- Shutoff in a supercritical storage system because the fluid is homogeneous and the mass of fluid left 'nr in the tank is directiy proportional io fluid density. The latter may be determined by the use of a capacitance matrix which measures fluid dielec- tric constant, or by the measurement of the power required to drive a small fan inserted in the tank. Use of such a fan can also eliminate temperature stratification in the fluid and promote higher internal heat-exchanger heat-transfer coefficients. Cabin air Subcritical compressed liquid storage with positive expulsion. - A system using the technique of subcritical compressed liquid storage T '6 with positive expulsion is shown schemati- cally in figure 32 which also illustrates system thermodynamic operation on a pressure-enthalpy 1: Fill diagram. Here, helium from an external high- T6 1 to 3: Initial pres- pressure source is used to pressurize a flexible surizat ion by bladder within the fluid storage vessel and heating thereby to expel fluid from the tank. Low fluid- 3 to 4: Standby venting storage pressures are used, so that, with proper 4to5: Launch 5 to 6: Orbital operation design, the fluid masses stored and expelled are liquid. Generally, it is assumed that the fluid storage Enthalpy space is capped on the ground with saturated liquid in the presence of a certain percentage FIGURE32.-Liquid storage withpositive expulsion. of vapor at atmospheric pressure, and that the (AFTER COE ET EL.*") storage pressure is regulated to some value lower than the fluid critical pressure. The fluid The absolute fluid-pressure decreases due to a fill state is shown as point 1 on figure 32. With drop in ambient pressure, assuming a gage the tank completely filled, a slight heat leak will type of pressure control. System demand and result in compression of the fluid at constant tank heat input determines the exact path fol- density until tank regulated pressure is reached lowed in this period; however, with no fluid (point 3). This initial compression takes place usage, some fluid will be vented during pressure with a small fluid temperature rise, as shown decay. on figure 32, so that the liquid is substantially Operating conditions during flight at altitude subcooled. follow path 5-6 in figure 32. Fluid is withdrawn Heat leak into the storage tank during standby from the tank at constant pressure, with pressure can result in venting of liquid at constant pres- regulation being provided by the helium system. sure, with standby operation between points 3 For stable operation, it is desirable that the and 4. This process is accompanied by a rise in storage fluid be kept as a single-phase liquid the temperature of the fluid mass in the tank, during use. This requires that the tank liquid with the result that the degree of subcooling temperature be kept below the saturation tem- attainable during use is diminished. perature at the regulated pressure or, to express System operation during vehicle launch and it differently, that the fluid vapor pressure be climb to orbit is shown as path 4-5 on figure 32. kept below the regulated pressure. 36 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS It is noted that single-phase storage in this use of capillaries or semipermeable membranes system is possible only if the fluid in the tank or even rotation of the storage vessel to create is at a uniform temperature. If, for example, an artificial gravity field. Magnetic fields may be energy added through the tank walls in zero used for liquid-vapor separation in cryogenic gravity is not transferred to the entire mass of oxygen vessels in view of the paramagnetic prop- fluid stored but is confined to the wall boundary erties of oxygen. layer, local film boiling can occur with subse- There are a number of possible vessel designs quent instability problems. With this system, utilizing two-phase storage. In this report, fluid stored in the tank has fairly constant tem- emphasis is placed upon a particular type of perature and enthalpy during use, giving a con- vessel where the fluid is stored as a liquid-vapor stant heat-sink capability. Storage quantity mixture but designed for automatic vapor de- measurement is difficult for this system, because livery. The fluid delivery method employed can of the variety of bladder geometries possible also be used as a zero-gravity vapor vent in other during operation. The following two possible types of storage system. methods of storage quantity determination are: Figure 33 illustrates the fluid withdrawal proc- (1) The volume of helium gas in the bladder ess used on a schematic pressure-enthalpy can be determined. This could be accomplished diagram and shows one method of tank heat indirectly by determining the resonant frequency addition for pressurization. Preuse pressuriza- of the bladder volume when small pressure pulsa- tion is similar to the process described pre- tions are applied from a transducer. Such a viously. Here, tank operation is of most interest. method could not be used if the storage fluid It is most significant that the storage space fluid contained any vapor bubbles. (2) Radioisotope counting can be used to determine the rates of fluid used and vented, and instantaneous or periodic flow totalization. This To system technique would introduce fluid loading complexities. This positive-expulsion liquid storage system is moderately complex, since it requires two fluid pressure vessels and pressure regulation systems. System reliability at present depends primarily upon the reliability of the thin pressurizing bladder used and its ability to withstand repeated flexing at cryogenic temperature levels. Vessel fabrication problems are also intensified air by the use of a bladder, particularly if the bladder is to be replaceable. It is considered that bladder reliability per se is not a critical development Saturated problem. liquid Subcritical pressure, two-phase storage with thermal pressurization. -For vessels having two-phase storage with thermal pressurization the fluid is carried at subcritical pressure and exists as a mixture of liquid and vapor. For zero- gravity applications, special phase-separation provisions are therefore required to permit pressure stabilization during delivery and to prevent the accidental loss of liquid when venting is necessary. A number of phase-separation tech- FIGURE33.-Two-phase storage with vapor delivery. niques are being considered. They include the (AFTER COE ET AL.~) ENGINEERING CONSIDERATIONS 37 is a mixture of liquid and vapor. In the absence cryogenic fluid, while the outer shell is exposed of gravity, the mass sampled at any point in the to the ambient atmosphere. The shells can be tank may consist of liquid and vapor in any pro- spherical or cylindrical although, when installa- portion. Sampling states may thus range from tion requirements permit, spherical vessels are point 2 to point 2' in figure 33. Withdrawal sys- preferred for minimum size and weight. tem operation is described below. In general, a cryogenic storage vessel for The fluid to be delivered is first passed through space-vehicle use must also incorporate the a valve and throttled to a pressure lower than following components which add considerable tank pressure. Referring to figure 33, states weight and obligatory volume to the overall before throttling may range from points 2 to 2', system: and after throttling from points 3 to 3'. The tem- (1) A delivery line from the inner shell to perature of the fluid after throttling, however, is beyond the outer shell. lower than storage temperature. This passage (2) A venting system from the inner container of the vent fluid through a heat exchanger within with a pressure relief valve; the venting system the storage space permits a transfer of energy can be complicated in a zero-gravity environ- to take place along paths 3 to 4 or 3' to 4 (essen- ment by the fact that only vapor should be tially constant pressure). The fluid can thus be permitted to escape through the vent line. evaporated and superheated slightly before (3) A fill line which, in general, is different being discharged at a temperature close to that from the delivery line for ease of filling and of the storage fluid but at a lower pressure. valve installation. Tank heat addition for pressurization is sim- (4) A quantity gage which will measure the ilar to that for supercritical storage. As shown content of the vessel under all operating con- in figure 33, the delivery fluid is heated to ambient ditions. temperature and used as an energy source. Fluid (5) An internal heat exchanger or electrical quantity measurement is possible by use of a heating coil to insure positive delivery in the matrix-type capacitance gage. case of a thermally pressurized vessel. This withdrawal method provides automatic (6) A bladder in the case of helium pressurized phase separation during venting and is attrac- vessels, with the appropriate lines to the high- tive for longduration, zero-gravity applications. pressure helium source. It is evident that this system has traded a phase- (7) Support members to transfer the high ac- separation problem for a zero-gravity heat- celeration loads usually present at launch or transfer problem. Careful insulation design is reentry to the outer shell and finally to the required to avoid venting during use when the vehicle. fluid withdrawn from the tank may be completely The major vessel weight items, aside from liquid. It should be noted that this system is not the fluid load, consist of the inner and outer suitable for the storage of a liquid mixture such shells and thermal insulation. It is important to as oxygen and a diluent gas because fractional remember that for small vessels, valves and distillation of the two-phase mixture will lead to controls introduce sizable heat leaks and weight fluid-composition variations during operation. penalties. Proper selection of all vessel com- Because much of the weight tradeoff data de- ponents is necessary, however, to insure proper pends on the critical aspects of overall system operation under space flight conditions. Weight size, fixation to the spacecraft heat leak, geom- tradeoffs should, but do not always, specify etry, standby time, etc., it is pertinent to consider these accessories. hardware components and general design pro- Physical properties of the fluid component of cedures which determine the size and weight the system affect much of the subsequent requirements for any given spacecraft cryogenic design procedures. Table 12 summarizes the application. properties in question. Components. -Cryogenic vessels usually con- Design procedures. -An analysis of the gen- sist of two concentric shells separated by an in- eral design procedures gives some understanding sulation space. The inner shell contains the of second-order criteria which must be applied 38 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS TABLE12. -properties of Cryogenic Fluids [AFTER COE ET AL.20]

Normal (14.7 Heat of 7. Fluid Critical Critical psia) boiling vaporization Liquid density, pressure, psia temperature, "R point, OR (at 14.7 psia), Ib/ft3 Btu/lb

Oxygen ...... 736 278 162.3 91.6 71.2 Nitrogen...... 492 227 139.2 85.2 50.4 Neon ...... 395 80 59.2 37.1 70.5 Helium ...... 40.6 9.36 7.6 8.84 7.8

in the subsequent tradeoff analysis. Given a in the storage space of a cryogenic tank. Fill spherical shape and general type of construction, density pf is thus slightly less than liquid density. the first step in design is the estimation of vessel Storage space volume is then fixed as fill load. Here, it is noted that mission and system specifications fix the useful fluid load Wuin V,= -Wf terms of a required delivery schedule wu over the Pf vessel operating time T~.This specification is expressed as For a spherical vessel, pressure shell inner diameter is then given as: wu= IwudTu (37) D = 12 The fluid fill load Wfis then the sum of the use- (31'3 ful load, the amount of fluid lost by venting With pressure-shell diameter fixed, a maximum during the standby and use periods WVand the operating pressure is assumed. Pressure-shell amount of residual fluid at the end of operation weight may then be calculated. A preliminary WR. Fill load is found as follows: estimate of shell support configuration is then made, and the line geometries are established. Following these steps, a thermal analysis is made The residual fluid load depends upon the fluid to permit the selection of the insulation material pressure and temperature at the end of use. and estimation of insulation thickness. These variables fix fluid end density, giving the Determination of insulation thickness requires, following relation for WR: first, that the vessel thermal design criterion be selected. This is often, as explained above, based rdh upon standby considerations. For thermally WR=- 6( 1728) (39) pressurized vessels of long use, insulation may be set by the heat required at minimum delivery Fluid venting losses depend, for a given tank rate. design, upon the duration of vessel standby and A preliminary insulation thickness is first cal- the external temperature environment of the culated using the assumption that vessel heat vessel during standby. Venting losses are also leaks through lines and supports are negligible. possible during operation at low delivery rates. This fixes vessel geometry. Line and support In some cases, if venting losses are arbitrarily heat leaks are then calculated, and a second specified, this specification fixes vessel-insulation approximation to insulation thickness is obtained. requirements. When insulation is based upon flow require- With the fluid fill load fixed or assumed, fluid ments during operation, a thermal design analysis fill density is selected. The filling operation is made to determine if fluid vent losses during usually ends with a small amount of vapor present standby or use result in a useful load less than ENGINEERING CONSIDERATIONS 39 specified. If this is the case, a new value of fill (2) Control and accessory weights are ignored; load is assumed, and the design is repeated. this is an important point. Tank insulation design is thus an iterative proc- (3) Room temperature properties of materials ess. With insulation thickness determined, the are used to give weights which could be lowered outer shell diameter and weight may be obtained, if this factor becomes critical in a design tradeoff. giving total vessel weight, less valves and con- (4) Vessel pressurization is achieved by means trols. These items can be specified independ- of electrical heaters, heat exchangers, or simply ently. by heat leakage from the outer shell, resulting The above indicate the steps generally taken in a uniform temperature throughout the mass in storage vessel design analyses. of the fluid stored. In practice, this condition may Some of the major internal and environmental not be realized unless suitable means are pro- factors determining the design weight of the vided for mixing the fluid inside the container. hardware are: Especially in a zero-gravity environment, where there are no natural convection currents, con- Inner shell: (1) siderable temperature stratification may exist (a) Internal fluid pressures of up to psia 3000 within the body of the fluid if the only mechanism (b) Launch and reentry loads for heat transfer were conduction through the (2) Outer shell: fluid itself. A fan or other suitable mixing device (a) Compression load from buckling pressure is necessary to validate this assumption. Such a of atmosphere device can also be used advantageously to in- (b) Effect of insulation and vacuum beneath it crease the heat-transfer rate from the heat ex- (c) Dynamic loads changer, thus reducing its size and weight. The (3) Insulation: one disadvantage of fluid mixing is the additional (a) Evacuation required to improve insulation heat that is dumped into the cryogenic fluid. and prevent liquefaction of atmospheric Obviously, the solution to this problem of fluid components within the space, with subse- temperature uniformity depends on the vessel quent deterioration of performance size, mission duration, and type of storage con- (b) Temperature and pressure variation inside sidered, and can be found only for a particular the craft application. In a general analysis of the type pre- (c) Compressive loads passing from outer to sented here, temperature uniformity must be inner shell assumed, although in practice, a computer pro- (d) Allowable heat-leak contribution from lines gram can be used to cover the non-uniformity of and support members temperature.12 (e) Ideal operational thermal requirements: no- loss standby for a given holdup with pres- (5) In general, the line and support heat leaks sure buildup from fill pressure to maximum are assumed to constitute a fixed proportion of pressure; constant pressure operation at the insulation heat leaks. This assumption greatly minimum delivery rate with no venting in simplifies the calculations, since these heat leaks thermally pressurized tanks; and no exter- depend on the geometry of the lines and supports nal heat input other than vessel heat leak of a particular vessel and can only be calculated exactly when the detailed design of the vessel It is quite apparent that all of the above factors is performed. Based on previous analysis of lines must be considered in detail before a gas-spe- and supports, it appears that the value of the cific cryogenic weight tradeoff can be made. ratio of line and support heat leaks selected for Minor variation in assumptions about any of these the numerical examples (0.20 insulation heat factors can alter the cryogenic storage penalty leaks) is conservative for large vessels, and can in any specific mission. be achieved for small vessels by careful design Storage vessel weight tradeofs. -In presenting of the lines and support members. typical cryogenic-system storage weights, the (6) Heat exchangers, instrumentation, and con- following assumptions are made: trol valves were not considered in the analysis. (1) Vessels are spherical. They are closely related to mission requirements 40 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS and are therefore treated as separate components; Oxygen Systems as such, these items, together with the storage Weight penalties for the three types of oxygen , vessel itself, form a subsystem. An analysis of storage vessels are presented in figures 34 to these subsystems will be presented below in the 38. The types of vessels considered have been comparison of the total gas system weight penal- described previously and include: thermally ties. The subsystem weight should be relatively pressurized supercritical storage; subcritical, constant for different gases stored. It should be compressed-liquid storage with positive ex- remembered, however, that for small vessels up pulsion; and subcritical, thermally pressurized to 10 inches in diameter, the weight of such items storage with vapor delivery. Useful oxygen load may in certain applications be an important part is taken as the primary design criterion. Vent- of the subsystem weight. ing losses have been ignored. (7) Other assumptions used in the numerical Supercritical pressure - thermal expulsion. - examples, such as constant ambient tempera- Vessel weight and volume penalties for this ture, constant pressure operation, constant rate type of storage are presented in figures 34 and of flow, etc., are clearly stated wherever used. 35 as functions of useful fluid load and min- The design methods outlined here are general imum delivery rate. As shown, delivery rate has and are not limited in any way by the assumptions very little effect except in the case of large made to present typical examples. vessels with low delivery rates. Inner shell The numerical assumptions of table 13 were weight is the major vessel weight penalty for followed. this type of system. One can question the use of magnesium as Subcritical pressure - bladder expulsion. - part of an oxygen storage system in view of the Vessel weight penalties for this type of system meteoroid and fire hazard. 91 The assumption are shown in figure 36 as a function of useful of 14.7 psia as fluid end pressure may also be fluid load. Since thermal design of positive- considered somewhat liberal. expulsion vessels depends greatly upon flow Tradeoff curves using slightly different as- schedule, an arbitrary value of 1 inch was chosen sumptions have been published more recently as the insulation thickness for all calculations. with slight reduction in penalties. 95 The penalty This is a tradeoff assumption which has been data for these advanced systems are discussed questioned. 12 Helium gas stored at 6000 psia below. and 530" R was used as the pressurization

TABLE13. -Sample Data Assumptions for Oxygen [AFTER COE ET AL.*O]

Physical parameter Supercritical Subcritical posi- Subcritical storage tive expulsion 2-phase storage

Design pressure, psia...... 1500 70 150 Fill factor, K, ...... 0.98 0.98 0.98 Fluid end pressure, psia...... 14.7 70 14.7 Fluid end temperature, "R ...... 300 162 Ambient temperature, OR...... 620 620 620 Inner shell material ...... A1 A1 A1 Outer shell material...... Mg Mg Mg Insulation k, Btu/hrft"R...... 5 X 1w 5 x 10-5 5 X 10-5 Insulation density, lb/ft3 4.7 4.7 4.7 Minimum insulation thi ...... 0.50 1.0 0.50 Support and line heat leaks, (qs + qu)/qT...... 0.20 0.20 0.20 Shell structure weight allowance, percent ...... 10 10 10 ENGINEERING CONSIDERATIONS 41

1.70

3' 3' ';c 1.60

x -+ !? a i 1.50 .-0 5

1.40 . 1 2 4 6 810 20 40 60 100 200 400 600 1000 Useful fluid Iqad, Wu, Ib

(a) Supercritical oxygen storage. Vessel pressure: 1500 psia; ambient temperature: 620° R; spherical vessel. (AFTER COE ET AL.~O)

FIGURE34.-Vessel weight penalty.

1.6

3= t- 3 1.4 x. -+ z aal 2 1.2 .-u) 3 1 .o 2 4 6810 20 40 60 100 200 400 600 1000 Useful fluid load, W Ib U,

(b) Advanced supercritical oxygen storage. Note that for flow higher than 4 lb/day, the vessel weight is defermined by the minimum insulation thickness. (AFTER ROUSSEAU ET AL.05)

FIGURE34.-Vessel weight penalty-Concluded.

fluid; helium vessel weight penalties were ob- assumption which has been questioned. 36 tained from figure 29. On the whole, subcritical storage appears to Subcritical pressure two-phase storage. - present a smaller weight penalty than cor- Weight and volume penalties for this type of responding supercritical systems. The volume vessel are shown on figures 37 and 38 as functions penalties are similar. of useful fluid load and minimum delivery rate. Here, the system was assumed to deliver sat- Nitrogen Systems urated vapor at storage pressure for the pur- Data for two types of nitrogen storage vessels poses of thermal design. This is a tradeoff are presented in figures 37 to 41. Vessel

261-559 0-61-4 42 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

Useful fluid load, Wu, Ib

(a) Supercritical oxygen storage. Vessel pressure: I500 psia; ambient temperature: 620" R; spherical vessel. (AFTER COE ET AL.)

FIGURE35.-VesseI volume penalty.

Useful fluid load, W , Ib U

(b) Advanced supercritical oxygen storage. Note that for flow rate higher than 4 Ib/day, the vessel volume penalty is determined by the minimum insulation thickness. (AFTER ROuSSEAU ET AL.85)

FIGURE35.-Vessel volume penalty-Concluded. 4.3

10 20 40 60 100 200 400 600 1000

Useful fluid load, Wu, Ib

(a) Subcritical oxygen storage with liquid delivery. Storage pressure: 70 psia; insulation thickness: 1 inch; spherical vessel. (AFTER COR ET AL.20)

FIGURE36.-Vessel weight penalty.

2.0

1.8 z=

1.0 2 4 6 8 10 20 40 60 100 200 400 600 1000 Useful fluid load, Wu, Ib

(b) Advanced subcritical oxygen storage. (AFTER ROUSSEAU ET AL.w)

FIGURE36.-VesseI weight penalty-Concluded. 44 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

1.20

1.15 9' '+ '+ 3

+x 0 1.10 C Q c L .-m 5 1.05

1 .oo 2 4 6 a10 20 40 60 1 00 200 400 600 1000 Useful fluid load, W Ib U'

FIGURE37.-Vessel weight penalty for subcritical oxygen storage with vapor delivery. Storage pressure: 150 psia; ambient temperature: 620° R; spherical vessel. (AFTER COE ET AL.~O)

,025

.020

,015

.010

FIGURE38.-Vessel volume penalty for subcritical oxygen storage with vapor delivery. Vessel pressure: 150 psia; ambient temperature: 620" R; spherical vessel. (AFTER COE ET AL.20) ENGINEERING CONSIDERATIONS 45 types considered are supercritical storage and alty. Weight and volume of cylindrical, super- two-phase storage with vapor delivery. Assump- critical storage vessels for nitrogen were also tions used were generally those given previously calculated based on the same design conditions. for oxygen, except for those recorded in table 14. The vessel length-to-diameter ratio was taken as 2.0. The results of these computations are shown in figures 41 and 42. These plots show a much TABLE14. --Sample Data Assumptions for Nttr~- heavier design, which is the penalty paid for gen [AFTER COE ET AL."] shape. Subcritical pressure, two-phase storage, vapor Supercritical 2-phase delivery. Weight penalties for this type of ni- Physical parameter storage storage trogen storage are shown in figure 43. Volume penalties are essentially the same as those given Design pressure, psia ...... loo0 150 in figure 40. Here, it should be noted that alu- Fluid end temperature, OR... . 300 140 Inner shell material ...... Ti Ti or AI minum is used as the inner shell material where Outer shell material ...... Mg Mg minimum gage thickness is desired, giving a broken weight-penalty curve. Air mixture.-It is possible to store liquid air for spacecraft use. Supercritical air-storage Supercritical pressure. -Vessel weight and vol- weight and volume penalties are shown plotted ume penalties are shown in figures 39 and 40 in figures 44 and 45. The calculations were per- as functions of useful fluid load and minimum de- formed for the same conditions as for the nitro- livery rate. The vessel weights are somewhat gen vessel discussed above. It should be remem- lower than shown for oxygen in figure 34. This bered that if uniform gas is desired, the two-phase results from the permissible use of titanium as storage is impossible for oxygen-nitrogen mixtures the inner shell material for nitrogen storage. with the throttling system discussed previously. A change in the shape of a vessel from spher- The bladder and other two-phase systems could ical to any other form increases the weight pen- be adapted for subcritical two-phase systems.12

1.35

1.30 ?' '+ '+ 3 x % 1.25 C aJQ * I .-m 2 I .20

1.15

Useful fluid load, Wu, Ib

(a) Supercritical nitrogen storage. Vessel pressure: 1000 psia; ambient temperature: 620" R; spherical vessel. (AFTER COE ET AL.*O) FIGURE39.-Vessel weight penalty. 46 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 1.6 , 32 1.5 \ 3-- 1.4 . -,X 1.3 U C 0) 1.2 + r .-0 2 1.1 I I IIIII 1 .o.- 2 4 6810 20 40 60 100 200 400 600 1000 Useful fluid load, Wu, Ib

(b) Advanced supercritical nitrogen storage. (AFTER ROUSSEAU ET AL.05) FIGURE39.-Vessel weight penalty-Concluded.

.04

,035

.015 2 4 6 810 20 40 60 100 200 400 600 1000 Useful fluid load, W,,, Ib

(a) Supercritical nitrogen storage. Vessel pressure: 1000 psia; ambient temperature: 620" R; spherical vessel.

FIGURE40.-vessel volume penalty. . (AFTER COE ET AL.20) ENGINEERING CONSIDERATIONS 47

> c

(b)Advanced supercritical nitrogen storage. (AFTER ROUSSEAU ET AL.~~) FIGURE40.-Vessel volume penalty-Concluded.

FIGURE41 .-Cylindrical vessel weight penalty for supercritical nitrogen storage. Vessel pressure: 1000 psia; ambient temperature: 620’ R; spherical vessel. (AFTER COR ET AL.*) 48 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

Useful fluid load, Wu,Ib

FIGURE42.-Volume penalty for supercritical nitrogen storage. Vessel pressure: 1000 psia; ambibnt temperature: 620' R; spherical vessel. (AFTER COE ET AL.*~)

1.25

1.20

3' 3' \ $ 1.15

x -+ 0 0 Q L %.- 1.10 5

1.05

1 .00 2 4 6 810 20 40 60 100 200 400 600 1000 Useful fluid load, Wu,Ib

(a)Subcritical nitrogen storage with vapor delivery. Storage pressure: I50 psia; ambient temperature: 620° R; spherical vessel. (AFTER COE ET AL.*O)

FIGURE43.-VesseI weight penalty. ENGINEERING CONSIDERATIONS 49 2.0

1.8

\2 is j, 1.6 m c P,(I 1.4 s.-m 1.2

1.0 2 4 6 8 10 20 40 60 100 200 400 600 1000 Useful fluid load, WU, Ib

(b)Advanced subcritical nitrogen storage. (AFTER ROUSSEAU ET AL.05) FIGURE43.-Vessel weight penalty-Concluded.

1.8

1.6

1.4

1.2

1.0 1 2 46 10 20 40 60 100 200 400 600 1000

Useful fluid load, Wu, Ib

FIGURE44.-Vessel weight penalty for supercritical air storage. Vessel pressure: 1000 psia; ambient temperature: 620" R; spherical vessel. (AFTER COE ET. ALZ0) 50 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

.05

s .04 L*

3’ ‘c * .03 -+x 0 0 n 0 -s >” .02

. 01 1 2 46 10 20 40 60 100 200 400 600 1000 Useful fluid load, W,, Ib

FIGURE45.-VesseI volume penalty for supercritical air storage. Vessel pressure: 1000 psia; ambient temperature: 620” R; spherical vessel. (AFTER COE ET AL.~~)

Sensitivity of 02 and Nr Cryogenic Storage Penalties to tion, the effect of standby time on tankage sys- Design Assumptions tem dry weight was analyzed for several usable Sometime after the cryogenic tradeoffs dis- fluid weights in supercritical and subcritical cussed previously were published, advanced storage with no ~enting.3~The tanks had a spheri- systems were analyzed using different basic cal inner shell of Inconel and an outer shell of assumptions. The pressures of the systems and aluminum. The glass paper-aluminum insulation materials were the major alterations. Table 15 had a conductivity of 3 x 10-5 Btu/hr-ft-OR. represents these new design parameters. Figures Figure 46 presents the results. The systems can 341b) and 35(b) represent the weight and volume tolerate a 48-hour standby without weight penalty. penalties of the supercritical and subcritical Initially, subcritical storage offers weight ad- systems for oxygen. Figures 39(b), 40(b), and vantages over its supercritical counterpart; how- 43(b) represent the weight and volume penalties ever, this advantage diminishes in cases of long for supercritical and subcritical storage of standby and small useful fluid payloads. This is nitrogen. It is of interest to compare these alter- due to the fact that the quantity of heat required nate designs with the corresponding figures to pressurize the fluid from 1 atmosphere to an 34(a), 35(a), 36(a), 39(a), 40(a) and @(a), and see operating pressure at 100 psia is only about 30 the effects of these changes in design on the rela- percent of that required for supercritical storage tion between useful weight and weight penalties. at 875 psia. This effect is especially noticeable at This advanced series probably represents the small payloads where insulation presents a larger current state-of-the-art for these systems. part of the total weight of the system. For long Sensitivity of 02 and Ns storage penalties to systems having long standby times, venting can mission-specijic variables. -As was discussed be used in both supercritical and subcritical above, mission-specific variables have a signifi- systems. In such cases, a tradeoff between vent cant effect on tankage tradeoffs. Several ex- fluid and insulation must be made. It should be amples of these effects will be presented. stated that anticipation of such long standby In a study by Hamilton Standard of oxygen times for oxygen systems is probably unrealistic, tankage for a large, manned, orbiting space sta- but may be realistic for inert gas systems.24 51

Parameter Oxygen Nitrogen

Design pressure, psia ...... 800 ...... 725 Maximum pressure, psia ...... 875...... 850 Inner shell material ...... Reni 41 ...... Reni 41 Minimum insulation thickness, in ...... 0.75 ...... 0.75 Insulation density, lb/ft3...... 5.0 ...... 5.0 Outer shell material...... A16061-Tfj ...... A16061-T6 Liquid fraction at fill...... 0.95 ...... 0.95 Vessel shape...... Spherical ...... Spherical

Design pressure, psia...... 100...... Maximum pressure, psia...... 120...... Inner shell material ...... A1 ...... AI Insulation thickness, in...... 1.0...... 1.0 Insulation density, lb/ft3 ...... 5.0...... 5.0 Outer shell material...... A16061-T6 ... Liquid fraction at fill ...... 0.95 ...... Accessory weight (gas bottle). lb...... 7.0 ...... Helium ...... Helium ...... 3...... 3 ...... Spherical ...... Spherical

The sensitivity of total weight of a nitrogen 500 system on rate of usage and usable fluid weight 2. can be seen in figure 47. The same assumptions + I as those for figure 40 hold true, except that the .-m 9, inner tank was of Inconel instead of titanium. In 3 t. 100 the case of nitrogen, the use rate is quite low D (leakage and repressurization) and becomes the E z 50 determining factor in the total system weight x

9) penalty. 0) Y In the case of weight tradeoffs for cabin re- C I-0 pressurization systems, the mission duration, 10 repressurization rate, and cabin size are critical 10 50 100 500 1000 determinants for various cryogenic and gas Standby time, hr storage systems. The Hamilton Standard Corp. has studied these repressurization system trade- FIGURE46.-Weight variation cryogenic oxygen of offs for two given cabin sizes, 7000 ft3 and 70 000 storage system with standby time. Use rate per tank =8.5 Iblday. (AFTER HAMILTON-STAND- ft3, at 7 psia in very large stations where repres- ARD.16) surization penalties become severe unless lock 52 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS systems are used. Three general methods were compared- high-pressure gas storage, super- . critical cryogenic with thermal pressurization, and subcritical cryogenic with thermal pressurization. The study also compared storage of separate . gases and mixed gases. The characteristics of , the tanks were the same as above with fluid delivery at 30 psia. Figures 48(a) and 48(b) represent the results for the 7000 ft3 cabin, and figures 49(a) and 49(6) the results for the 70000 ft3 cabin. It can be concluded from these figures that for small fluid payloads and long standby times, gaseous storage looks attractive. The

(a) Subcritical nitrogen storage. FIGURE47.-Total system weight versus fluid. (AFTER HAMILTON-STANDARD.36) 0 1 2 3 4 5 Standby time, yr (0)

(a) Minimum storage weight.

FIGURE48.-Weight of repressurization storage system versus standby time. Cabin volume: 7000 ft3; total pressure: 7.0 psia; useful oxygen required: 138 Id; useful nitrogen required: 120 lb; spherical tankage. (AFTER HAMILTON STANDARD. 38)

I I I I A

=- 800 L 0, 5 700 I

2 600 h '. 500 I OI * Supercritical N, + 0,. 1 tank 2 400 'Subcritical N, + O,, I tonk 01 - I I 0 1 2 3 4 5 Standby time, yr (b)

(b)Total storage weight. (b) Supercritical nitrogen storage. FIGURE47.-Total system weight versus fluid- FIGURE48.-Weight of repressurization storage Concluded. system versus standby time-Concluded. ENGINEERING CONSIDERATIONS 53 gaseous-cryogenic crossover point in figure 48(b) that percent left after standby. For short standby appears at 2.6 years. For larger systems, the and large fluid payloads, the subcritical storage crossover point is shifted to longer standby times methods have weight advantage over supercritical (fig. 49w. (figs. 48(b) and 49(b)). Because the subcritical system has a lower For repressurization, a considerable weight average specific heat input requiring more insula- saving can be attained by mixing the 02 and Nz tion, the optimum weight of the supercritical and storing both fluids in the same tank (figs. 48jb) systems is less than the subcritical system for and 49(b)). In general, system weight increases long periods of standby (fig. 48(b)). The average with the increase in the number of tanks needed specific heat input can be defined as the average to store a given payload. Care must be taken in heat input required to maintain tank pressure programing the withdrawal to avoid compositional and vaporize the fluid per pound of fluid with- changes in the final gas output. drawn between the 100 percent full condition and Helium Systems I I I I -0, Gas storage- i I I Experience with small flight-rated cryogenic helium systems is limited. Because the rate of use of helium will be so low and the heat-leak 0, Supercriticol I 1~~. factor so great in determination of weight tradeoffs in helium systems, it appears appropriate to review in detail some of the second-order factors involved in the design of small cryogenic helium tanks. It should be kept in mind that improved Itechnology may alter some of the numerical factors used. Figure 50 reviews the state of the art in low Standby time, yr (0) heat-leak systems. The curve represents the (a) Minimum storage weight. minimal heat leak which can be expected in systems. The modified Gemini oxygen system is FIGURE49.-Weight of repressuriza tion storage of AiResearch design. The helium and nitrogen system versus standby time. Cabin volume: systems are estimates made by the Cryogenics 70 000 ft 3; total pressure 7.0 psia; useful oxygen required: 1380 lb; useful nitrogen required: Group of the Boeing Co. The nitrogen and helium 1202 lb; spherical tankage. (AFTER HAMILTON tanks are evaluated for 5-psia and 7-psia cabin- STANDARD .36) Gemini 0, Apollo 0, 6000 I I I I 20 Btulhr 20 Btulhr 14.01 I IIIIII Ill Gor storagef 12.0 state-of-the-art /‘. Mddified Gemini -& 10.0 -(existing) / oxygen system 3 c * 8.0 ~ I x- A Helium,psia basic tank m 6.0 I m ~ j Nitrogen, 700 psia basic tank 0) I 1,) 4.0 5.0 psia rn Optimized nitrogen tank

2.0 0 1 2 3 4 5 Standby time, yr (6) 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 (b) Total storage system weight. storage volume, ft3

FIGURE49.-Weight of repressurization storage FIGURE50.-State-of-the-art tank heat leak. system versus standby time-Concluded. (AFTER BOEING.~~) 54 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS compartment pressures. The helium tanks rep- first glance, one would think that the heat leak resented are a basic 400-psia system and an could be avoided by operating at a higher tank optimized venting tank. This is the minimum pressure or by venting fluid. pressure at which the helium tank can operate Because of the complex nature of the weight and yet provide, by allowing pressure delay, the tradeoff, however, the total system penalty does flow rates required for compartment repressuriza- not get smaller with increasing pressure. Figure tion. The nitrogen tanks are represented by the 52(a) represents the role of insulation thickness basic 700-psia tanks, as well as an optimized tank on the weight tradeoff presented by the Cryo- of nonventing type operating at 2000 psia. The genics Group at Boeing. These leaks were deter- “basic” tank is at the minimw pressure for mined by utilizing the optimum secondary supercritical storage. It can be seen that these support structures for heat-leak resistance small tank systems for helium and nitrogen are available at Boeing. certainly pushing the state of the art. The small Total system weights for helium in a two-man volumes require relatively large percentage orbiting laboratory system is presented for penalties for the fill tubes and secondary hard- different insulation thickness (spacing between ware weight as well as for excessive insulation shells) in a 5-psia compartment. Both unvented weight to cover large obligatory heat leaks storage at 2000 psia and vented storage at 1000, through fill lines and support structures. Improve- 1500, and 2000 psia are plotted for spacings of ment in secondary hardware could certainly effect gross changes in these figures. Figure 51 represents the effect of pressure on the maximum allowable heat leak and suggests that supercritical operation is required. As the pressure is increased, a higher design heat leak can be tolerated. The nitrogen tanks can take a much higher heat leak than can helium tanks of corresponding volume at any given pressure. The smaller the tank, the smaller the maximum heat leak. The 1.9 ft3 tank, for instance, could probably serve the Apollo mission and sustain a 1.8 Btu/ hr heat leak at 700-psia pressure. The basic 400-psia cryogenic helium tank of this same size (a1 Space between inner and outer tolerates a maximum 0.25 Btu/hr leak rate but shells, in. could tolerate 1 Btu/hr heat leak at 2000 psia. At

6.0 -2 5.0 3 c m 4.0 J m 3.0 m0) S c 2.0 .-CI,* d 1.0 0 loo0 ZOO0 3000 4ooo 0 (b) Tank operating pressure, psia Tank operating pressure, psia FIGURE52.-Helium storage for 30 days. (a) FIGURE51.--Maximum allowable heat leak for Supercritical s forage. (b) Low-temperature supercritical nonvented tank. (AFTER BOEING.~~) gaseous storage. (AFTER BOEING.~~) -~

ENGINEERING CONSIDERATIONS 55 2 to 12 inches between the shells. The system for the 5-psia compartment, and this is the major . weight includes usable fluid, vent fluid, fill fluid, factor in the weight difference. tank, and the weight equivalent of electric power The design uncertainties brought _about by the required for expulsion of the liquid helium. It acute sensitivity of the small helium systems to can be seen that the nonvented storage presents heat leak cannot be overemphasized. Figure 54(4 a greater systems weight penalty than does the represents the sensitivity of system weight vented system. There appears to be an optimum penalty of cryogenic helium to errors in heat-leak range of 3 to 4 inches of inner-to-outer wall calculations. The same components of system spacing for vented tanks and 10-inch spacing for weight are included as those in figure 52(a). The minimum system weight for nonvent design. system was designed for a 30-day mission at Increasing the pressure of vented storage sys- 7.0-psia cabin pressure and is used for the op- tems has little effect on overall system weight. timized helium tank of figure 50 at lo00 psia. It There appears to be a weight minimum at 1500 psia. Calculations for 3000 and 4000 psia indicate 80 progressively greater penalties along a family of curves paralleling the 2000-psia curve. 60 -n In figure 53 the effect of compartment pressure c L on cryogenic storage can be seen to be interacting .-rn $ 40 with insulation thickness to alter total system of e nitrogen storage. The lower pressure of 5 psia rn appears more favorable than 7 psia from this - 20 point of view. It should be remembered, however,

that the usable quantity of nitrogen is much less n 0 0.2 0.4 0.6 0.8 1.0 1.2

(a) A Heat leak, Btuhr

20.0 1 60 15.0 E -6- 140 c c 10.0 II w .- -.-E, r” 120 a 5.0 +i 100 v) 0 0 0.2 0.4 0.6 0.8 1.0 1.2 80 I ( b) A Heat leak, Btuhr -- Vented storage I I FIGURE54.-Helium tank sensitivity to heat leak 60 I .. for30 days at 7.0psia. Basicsystem (optimized) was used: 1000 psia vent tank; 11.8 pounds of I I I I I I usable helium with 4.4 pounds of vent helium; 0246 8 10 12 1.15 Btu/hr heat leak; 2.42 ff 3 storage volume; 110-pound system weight. (a) System weight Space between inner and outer shells, in. if heat-leak design margin (from 1.15 Btu/hr) is selected. (b) Tank vent fluid if tank heat FIGURE53.-Nitrogen supercritical storage for 30 leak exceeds design value of 1.15 Btu/hr. days. (AFTER BOEINC.~~) (AFTER BOEING.~~) 56 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS has an 11.8-pound usable helium weight in a structural design of the helium tank the faillire 110-pound system with a 2.42 ft3 storage volume. mode with the hydrogen-tank pressure of zero , To account for a predicted heat leak of 1.15 must also be considered. For the quantities of Btu/hr, 4.4 pounds of vent helium is expected. helium stored in this study, however, the internal Figure 54(a) shows that an error of only 1 Btu/hr helium maximum pressure resulted in a vessel . in heat-leak calculation of the design can increase capable of withstanding this condition. the total system weight by 50 pounds or nearly Table 16 indicates the weight tradeoff of the 50 percent. Figure 54(b) indicates that an actual several helium systems studied by the Boeing heat leak of only 1.0 Btu/hr above design value Co. for a manned orbiting laboratory for a 30-day brings about an increase in the vent helium from mission. The system weight includes vent fluid, a predicted 4.4 pounds to 14.6 pounds, thus unavailable fluid, fill tolerance fluid, dry tanks, reducing the usable fluid quantity. valves, heaters, and the energy-weight penalty Another approach has been studied by Boeing. for electrical thermal expulsion. The tradeoffs If the vehicle has a hydrogen tank for a fuel-cell for the cryogenic helium system proposed and the reactant supply or for a propulsion engine, con- heat-leak uncertainty factors are noted in figures sideration should be given to mounting the helium 52 and 54. The penalty is about 9 pounds of total tank within the hydrogen tank. This method ,system weight per pound of useful fluid. These would result in low-temperature gaseous storage are compared with the more certain tradeoffs for of helium with a fluid storage density comparable nitrogen. The leak values of study 2 in table 10 to that of liquid helium. The advantage is that the were used to determine usable fill weights. helium tank does not require insulation and there- It can be seen in table 16 that the supercritical, fore the tank design is simply a high-pressure nonvented storage of helium in a 5-psia cabin gaseous storage vessel. A thermodynamic analy- requires 71 pounds, the lowest total storage- sis must be made for each mission to establish system weight penalty for the cryogenic systems the minimum expulsion rate, final density, and under study. The low-temperature gaseous stor- optimum storage pressure. For the same usable age in the liquid hydrogen tank gives the lowest fluid quantities and mission as the supercritical value of 53 pounds. tanks shown in figure 52(a), figure 52(b) shows the Another major factor which must be considered resulting system weight versus helium storage in the helium tradeoffs is the duration of mission. pressure. This penalty also includes the penalty As discussed in the case of nitrogen systems, to the hydrogen tank for installation of the helium both the size of the helium system and the stand- tank. The hydrogen tank for this study was a by times interact in determining the specific 30-cubic-foot supercritical tank supplying a fuel weight penalty. Each cannot be considered alone. cell during the 30-day mission. The energy needed The Cryogenics Group at AiResearch has re- to expel1 the helium is transferred into the helium cently reviewed this problem.66 Four different from the hydrogen; therefore, no heaters or mission durations were selected for study: 30 blowers are required in the helium tank. In the days, 60 days, 90 days, and 180 days. They are

TABLE16. - Diluent Tankage System Weights for 30-Day, 2-Man Mission [AFTER BOEING 12]

I Nitrogen system weight, Ib I Helium system weight, Ib Storage method I I I 5.0-psia cabin 7-psia cabin 5.0-psia cabin 7-psia cabin 1 with 32 Ib I with 85 Ib I with 4.5 lb I with 11.8 Ib of useful fill of useful fill of useful fill of useful fill I I I I

Supercritical- vented...... 108 174 121 180 Supercritical - nonvented...... 113 175 71 110 Low-temperature gaseous storage in liquid-hydrogen tank ...... 53 105 ~~

ENGINEERING CONSIDERATIONS 57 listed in table 17 along with the corresponding mode, the tank pressure is allowed to increase crew size, minimum use rate, and standby times. slowly during the mission. A portion of the energy The constant 0.005 lb/hr is determined by the transferred into the liquid is used to expel the leak rate which is assumed to be the same in all demand and thereby to reduce the insulation vehicles. The first three missions are identical requirement. For the 30-, 60-, and 90-day missions except for differences in duration. The fourth is the pressure-variant tanks have a maximum pres- for a lunar shelter with 90 days of standby time sure of between 850 and io00 psia and have the before use. same weight penalty. At first glance it would appear that tanks for the longer mission would be larger and would entail a greater weight penalty because of the same demand flow to cover a constant leak rate. However, the greater quantity of fluid stored in the longer mission allows a greater amount of energy to be absorbed Number Usable Standby by the stored cryogenic per unit increase in pres- Mission length, days of men weight, Ib duration, days sure. This counteracts the other factors mentioned above. It should be pointed out that utilization of the 30...... 2 5.8 2 pressure-variant mode may not be acceptable if 60...... 3 11.6 2 90...... 3 17.4 2 helium is to be capable of supplying the high 180...... 3 34.8 a 90 flow rate for compartment repressurization in the launch or orbit-stabilization phases of a space

a Lunar stay. mission. Figure 55 represents the data for the isobaric Because the Cryogenics Group at AiResearch and pressure-variant cryogenic systems and has had the most practical experience with small shows that their weight penalties are more favor- cryogenic storage systems for helium, a close able than that of the gas-phase storage at 3000 analysis of the approach appears warranted. The psia. The 180-day mission has a 90-day standby general designs of cryogenic helium tankage for time which must really be considered apart from all these missions are the same. They are flight weight, vacuum-jacketed, supercritical pressure 18 vessels with superinsulation and a vapor-cooled - 16 - shield. The inner vessel is of Ti-5A1-25%-ELI. 3000-psia, ambient temperature-/ The vapor-cooled shield is of A1 6061 and the story system outer shell and mounting ring is of A1 2219. Com- pressed fiber glass is used for the inner-shell l4 ' support and the insulation is aluminized Mylar. The delivery systems for all the tanks are similar. The fluid leaving the tank intercepts heat being transferred through the insulation at the vapor- cooled shield. The fluid is then heated to atmospheric temperature in a warmup-heat exchanger before reaching the shutoff valve. A pressure relief valve operating at atmospheric temperature is employed.

There are two modes of operation for this n _. 60 90 120 150 180 cryogenic system -isobaric and pressure variant. Mission duration, days The pressure-variant mode leads to a lighter FIGURE55.-Weight penalty for helium storage system. In operation of the pressure-variant systems. (AFTER MASON AND POTTER.^^)

261-559 0-67-5 58 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS the other three missions which have only 2-day difference in the assumption regarding min- standby times. imum heat leaks which can be tolerated by Because of the long standby time, venting optimized insulating support systems for the during the later portion of the standby period tankage. By improving this parameter, the heat must be considered. In view of this venting, leak may be decreased, and the total system a pressure variant operation proposed for the weight penalty can be greatly improved by the shorter missions is not applicable. The tank relationships presented in figures 54(a) and represented in figure 55 for the 180-day mission 54(b). undergoes approximately 35 days of non-vented Work in progress on the ground-support standby and then vents at lo00 psia for the re- helium tank for the Apollo lunar excursion mainder of the 55 days before beginning mission module and on small cryogenic vessels may operation. give a more definitive answer to the problem. In these considerations, flow rates are critical. Because of the lack of operational experience If the use rates were to be increased, the weight with small cryogenic helium systems in space- penalties will decrease. To indicate the change craft and the indicated sensitivity of the system in penalty involved, the 30-day mission with to small design errors, it would be wise to isobaric tank was further analyzed. The ref- approach helium tradeoffs with great caution. erence quantity of usable weight shown in Neon Systems table 16 was increased in proportion to the increase in minimum demand flow. Figure 56 Unfortunately, little work has been done on represents the improvement in WT / Wu from flyable neon cryogenic systems. The Linde 3.8 to 2.75 as demand flow increases from Division of Union Carbide Corp. has made small 0.005 to 0.15 lb/hr. cryogenic neon units for laboratory use.68 It is of interest to compare the weight ratio AiResearch is currently studying neon cryogenic WT/W~of 3.8 presented by AiResearch (fig. 56) systems but no published data are available.M with the 9.3 value predicted for the 7-psia cabin Neon appears to be a much more favorable by the Cryogenic Group at the Boeing Co. (table liquid for cryogenic storage than is helium. 16). The chief difference is probably the fact Matsch reports that because of the high heat of that the Boeing group includes vent fluid, heater vaporization and liquid density (table lo), much power, accessory structures in their penalty; less boiloff of neon occurs.68 In commercial con- while AiResearch includes only the basic tank tainers of 25-liter size (8 to 12 lb of liquid), the and useful fluid. (See table 19.) There is also a normal evaporation rate of nitrogen is 1.9, neon is 6.3, and helium is 18.1 ft3 (STP) per day. The percentage of boiloff per day is 0.33 per- 4.0 cent for liquid nitrogen, 0.54 percent for liquid neon, and 3.0 percent for liquid helium. The 3-= '+ '+ gross weights for lo00 ft3 (STP) of gas are 92 3- 3.0 pounds for nitrogen, 63 pounds for neon, and 31 x -c pounds for helium. On a volume basis, neon also 0 0 offers 3.5 times more refrigeration than does a 2.0 -c liquid hydrogen and 40 times more than liquid ._m 2 helium. 1 .o A preliminary review of the data at AiRe- .005 .010 .015 search 66 indicates that a subcritical system Minimum demand flow, lb/hr designed for 30 days of minimum leakage at 0.012 lb/hr with an initial charge of 20 pounds FIGURE56.-Weight penalty change due to mini- and a pressure-variant operating mode from 450 mum demand flow increase for 30-day mission, 1000-psia isobaric operation. (AFTER MASON AND psia to 1500 psia will have a dry weight of only POTTER.^ about 17 pounds. Therefore, wT/wu=37/20= 1.85 ENGINEERING CONSIDERATIONS 59 for neon compared with an optimized 3.8 for (2) Alkali and alkaline earth chlorates and . helium and about 1.2 for nitrogen. perchlorates Since the boiling point of liquid neon is above (3) Hydrogen peroxide that of liquid hydrogen, gaseous storage in a (4) Water electrolysis liquid hydrogen tank is impossible. Because of Table 18 shows some of the pertinent physico- the favorable aspects of neon from a physiological chemical properties of oxygen-producing chemi- point of view, inore work on the cryogenic storage cals suitable for space cabin use. Lithium of this gas appears appropriate. The problem peroxide is not available commercially, and of storage of the technical grades of neon con- calcium superoxide, because of its low yield per taminated with 15 percent helium also requires pound in commerically available material (50 further study. percent impurity), is of value only in extravehicular suit backpacks where its resistance to fusion SOLIDCHEMICAL STORAGE is of merit. Because relatively stable forms of chemical Potassium and sodium peroxides are com- compounds containing a high percentage of pounds of primary interest in the first category. oxygen and nitrogen are available, this mode of They absorb water and carbon dioxide and pro- storage appears particularly suitable for cabin duce carbonates, bicarbonates, and oxygen. In pressurization, erection of inflatable structures, terms of oxygen storage capacity, the ozonides emergency breathing supplies, spacesuit back- are superior to corresponding superoxides. (See packs, and nitrogen supplies for missions requir- table 15.) The potassium and sodium ozonides ing small units with long standby time prior to are readily prepared.83 As with the superoxides, operation. Several excellent reviews of the sub- lithium ozonide theoretically has the most desir- ject are available.20,65, 83, 81, 13, 74, 14, 64, 82, 84 able characteristics in terms of oxygen availability (0.73 lb/lb of compound), but all attempts Chemical Oxygen Supplies - General Considerations at synthesis have failed.= Lithium peroxide has Oxygen producing chemicals can be divided been synthesized. Chlorate candles are stable into four major groups: materials which can be burned in generators to (1) Alkali and alkaline earth peroxides, super- produce oxygen at a constant rate. Hydrogen oxides, and ozonides peroxide is a strongly oxidizing liquid which can

TABLE18. -Comparison of Oxygen-Producing Chemicals [AFTER COE ET ALJO AND PETROCELLI

I KO? I NaO? I,i202 NaO, LiNO, LiC104 NaC10, HzOz HzOt

Available 02(theoretical), weight percent ..... 33.8 43.6 34.8 56.3 23.2 60.1 45.1 47.1 47.1 Purity...... 0.90 (9 ...... 1.00 1.00 ...... 0.90 0.98 Available OzrIb/lb ...... 0.32 0.392 0.375 0.56 0.232 0.601 0.40 0.423 0.461 Density, Ib/in.3 ...... 0.0237 ...... 0.0774 ...... 0.0861 0.0878 0.0815 0.0502 0.0515 Heat of reaction, Btu/lb ...... 415 635 e -363 +1515 -488 - 5% + 422 + 1106 1214 H20 balance, lb/lb ...... - 0.0207 - 0.0246 ...... -0.136 0 0 0 +0.577 + 0.539 HtO balance, Ib/lb 02.....- 0.0862 -0.0862 -0.225 0 0 0 + 1.34 + 1.17 Cot balance Ib/lb ...... 0.31 0.40 0.96 0.31 ......

a 10 percent LiZOd. + Indicates exothermic reaction: - indicates endothermic reaction. '2 KOz+ 1.23 COtt0.23 H20-=0.77 K~CO3+0.46KHCOa+ 1.5 02. 2 NaOt + 1.23 Cot + 0.23 HzO = 0.77 Na~C03+ 0.46 NaHC03+ 1.5 02. e Li202. 60 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS be decomposed catalytically to generate oxygen, utilization (about 80 percent) of the superoxide water vapor, and heat. charge. The inefficiency of such canisters can be attributed to the formation of a hard crust of Superoxides, Ozonides, and Peroxides potassium hydroxide on the reaction surface of The reactions of superoxides with water vapor the superoxide, thereby preventing water vapor and carbon dioxide to form oxygen have been in the exhaled breath from contacting the unre- reviewed by Petrocelli and much of the following acted superoxide. The discovery that bicarbonate discussion is based on his work.*' These reactions does form under certain conditions of tempera- can be expressed by the following equations: ture and relative humidity has shown that the problem of oxygen overproduction, anticipated 2MOz(s)+ HzO(v)=2MOH(s)+ 3/202(g) (42) when only carbonates were thought to be formed, and is insignificant. Semipassive superoxide systems have been designed to incorporate control of flow rates and relative humidity to achieve better than 90-percent oxygen recovery from the superoxide where supply.69*52, 51 In effect, the following stoichiometry can be s solid achieved in a properly designed superoxide v vapor reactor. g gas 1 liquid 2MOz(s) 1.23COn(g) 0.23HzO(v) M alkali earth element + + = 0.77MzC03(s)+ 0.46MHC03(~)+ 1.5Oz(g) (46) In turn, carbon dioxide is removed from the breathing atmosphere through reactions with the Lithium peroxide (LizOz) is of interest as an air product hydroxide which cause the formation of vitalization material because in the presence of carbonates and bicarbonates: moisture it can be caused to react directly with carbon dioxide to yield oxygen and lithium 2MOH(s) + COz(g) = MzC03(~)+ HzO(1) (44) carbonate: 74 ~MOH(S)+~CO~(~)=~MHCO~(S)(45) On the basis of these stoichiometries, the theoretical respiratory quotient (RQ), capable Thus, it is possible to remove 0.96 pound of car- of being obtained with superoxide systems, bon dioxide with each pound of lithium peroxide ranges from 0.67 (carbonate formation only) to from a closed breathing system and, at the same 1.33 (bicarbonate formation only). With ozonide time, to return 0.35 pound of oxygen to the sys- systems, the theoretical RQ range is 0.40 to 0.80 tem. The RQ for a system employing only lithium for the corresponding stoichiometries. peroxide would be 2.0. As a result, the use of this The early concern about RQ mismatch with chemical would require an additional source of humans has been resolved by analysis of alter- oxygen. The theoretical capacity of lithium per- nate reaction mechanisms. At first, superoxides oxide for carbon dioxide is about 4 percent greater were evaluated on the basis of a stoichiometry than the capacity of lithium hydroxide for carbon which involved the formation of the metal car- dioxide. bonate only (eq. (44)). Thus, the RQ of the system In 1964, Markowitz demonstrated that, in the was expected to be 0.67 and oxygen overproduc- presence of water vapor, carbon dioxide absorption was expected. The other factor which con- tion and oxygen evolution by lithium peroxide tributed to doubts about the superoxides is does occur, but oxygen generation lags far be- based on the experience gained from the use of hind the amount anticipated on the basis of potassium superoxide canisters in self-contained equation (47).74 Yet Markowitz was able to ex- breathing apparatus for firefighting and mine plain his results by showing that the absorption rescue. Such canisters resulted in very inefficient of carbon dioxide and the evolution of oxygen ENGINEERING CONSIDERATIONS 61 proceed by two different reactions; lithium perox- calculated from estimated values of the heat of ide and water vapor first reacting to yield the formation, entropy, and heat capacity. The esti- active carbon dioxide absorbents, LiOH, LiOH mates were based on graphical comparison with H20, and hydrogen peroxide: properties of other oxides. (See fig. 57.) The LizOZ(s)+ 2HzO(v)= 2LiOH(s)+ HzOz(l) (48) heat of formation was also determined from calculation of the lattice energy by means of and the Born-Haler cycle, as -65 kcai as compared LiOH(s) + HzO(v)= LiOH * HzO(s) (49) to -38 kcal for lithium peroxide and about Carbon dioxide is then absorbed via: -20 kcal for the other superoxides (fig. 57). From the Born-Haber result and other estimated ZLiOH(s) + CO&) = LizCOs(s)+ HzO(Z) (50) data, the free energy at room temperature is and - 53 * 5 kcal. Consideration of the free energies of various 2LiOH * HzO(s) CO&) = LizCOa(s) ZHzO(1). + + decomposition reactions showed that the tend- (51) ency to decompose corresponds to 15 kcal from Oxygen is evolved as a result of the decomposi- 100" to 300" K. This tendency is so much greater tion of the HzOz: than the uncertainty of the estimates that lithium superoxide can be considered unstable at all temperatures. Furthermore, none of the usual methods of promoting stability are sufficiently- Markowitz points out that in order to achieve theoretical yields of oxygen, it will be necessary to develop a catalyst to insure the decomposition -50 of all the H202 formed in equation (48). In 1964, Dobrynina published a monograph on lithium peroxide in which she thoroughly re- - 40 viewed the state of the art with respect to its chemistry.22 Ducros and Beranger have also reviewed their preliminary respiratory exchange studies with this compound.26 These studies -30 show that the state of the art of lithium peroxide, as an air revitalization material, is not nearly ii5 -20 as advanced as it is for superoxides. Continued I basic research is necessary in order to optimize 9 lithium peroxide as a carbon-dioxide absorber Y0 and oxygen source. J--10 Lithium superoxide (LiOz) if it exists in a I a stable form would be of great value for air regeneration. Lithium superoxide potentially represents the lightest alkali metal oxide in 0 terms of weight of agent per weight of oxygen produced. Experimental efforts to produce this compound have given ambiguous results. An IO effort has been made to estimate the thermodynamic properties of this compound to determine whether further experimental efforts are 20 worthwhile, to predict suitable experimental 20 40 60 . 80 conditions, and to draw conclusions about the Atomic number of metal stability of the compound. 1°3 FIGURE57.-Heats of formation of oxides of alkali The free energy of lithium superoxide was metals and hydrogen. (AFTER sNoW.1O3) 62 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS effective to overcome this instability. Substance? absorption are assumed to be the same as those can be stabilized by putting them into solid solu- for LiOH systems in figure 59. Using these , tion. For example, phase data have shown the flows and assuming that all the heat of reaction existence of solutions of sodium superoxide in is all dumped into the process airstream, the sodium peroxide. It has been shown that, theoreti- temperature rise of air flowing through the cally, no significant concentration of lithium superoxide bed can be determined for any cabin superoxide can be stabilized in this way.103 This pcoZ. This relationship is recorded for both conclusion might be different if a mixed compound that has a definite heat and free energy of CO? partial formation is formed. Such compounds do not pressure sensor P usually have sufficient free energy to overcome the instability of lithium superoxide. Further attempts to prepare lithium superoxide do not From humidity appear promising. Even if the compound were control subsystem ! i prepared, it would tend to decompose spon- :I j :: ; taneously. It would not be safe to carry such an .__>:; I unstable compound in a manned space cabin. e..-..’Filter Flow Weight and Volume Tradeoffs of Alkaline Earth Superoxide controller and Peroxide Systems FIGURE58.-AIkali metal superoxide subsystem Because of several operational interactions diagram. (AFTER ROUSSEAU ET AL.Oa) with other systems, determination of weight tradeoffs for superoxide systems is most difficult. In the presence of leaks and the resultant waste of oxygen, supplementary oxygen systems will probably be required. The simultaneous requirement for extra cooling capacity, on the one hand, and provision of carbon dioxide absorbtion, odor control, and sterilization of air, \ on the other, alters the requirements for weight 1Cabin pressure, and power ordinarily allocated for these pur- psia __ poses. Inclusion of these factors in tradeoffs is feasible only in final systems-integration steps. \ The use of superoxides is very similar to that of LiOH in that they are packaged as granular Y \ beds through which the cabin gas is circulated. Figure 58 is a schematic of the system with the appropriate water and carbon dioxide controls required for adequate control of system RQ and COz absorbtion. The filtered granular beds must be of stainless steel or other metal pro- tected by polyethylene or polyfluorocarbon coatings. Test results for the superoxide systems suggest that utilization efficiencies of 0.90 are at- 3 4 5 6 7 8 tainable.20 Since the purity of the NaOz and C02 partial pressure in cabin, mm Hg KOZ are about 0.95, the consumption rates for COZ control are 5.46 and 6.9 lb/man day, re- FIGURE59.-Lithium hydroxide subsystem HOW spectively. The flow characteristics for COz requirement. (AFTER ROUSSEAU ET AL.05) ENGINEERING CONSIDERATIONS 63

NaO;? and KO, at several cabin pressures.95 20 The additional cooling power to handle this heat 18 must be accounted for in final weight tradeoffs. The design equations for superoxide canisters 16 may be found in the studies of Coe et a1.20 and Rousseau.95 Figure 60 represents the canister 14 weights of NaOZ and KO, required for different s +- 12 I numbers of crew members. Since the expendable .-0) superoxide weight of the potassium subsystem is : lo i-l considerably greater than the sodium, only the c .: .: 8 latter is discussed further. Potassium superoxide 6 give slightly lower pressure drops in the canister 6 and heat rejection load, but they account for only a small part of the overall subsystem equivalent 4 weight. The total subsystem equivalent weight 2 is the total of the sodium superoxide consumption, the canister weight, accessory weight, 0 power-loss penalty, heat-rejection penalty, and 123 4 5 6 7 8 9 10 material balances weight. Then a deficit of water Number of crew members exists, the material balance requires additional water and causes a penalty. However, oxygen FIGURE60.-Alkali metal superoxide canister which is added by the system can be subtracted weight. (AFTER ROUSSEAU ET AL.OJ) from the consumption weight by a factor of Wo, = 2.28 N7, where N is the crew size. The system equivalent weight penalty is

We = ( wNao2 - WO,+ wd+ wcan + Wac, + WP+ w, We = (5.52- 2.28 + 0.185)NT + 3.423 Nz'3 + (5.2+ 1.79d%)+ [(PL)t(PP)]+1.70N(RP) (53)

This equation has been solved for a typical set of conditions defined by the assumptions: cabin pressure = 10 psia; pcOs = 7.6 mm Hg; heat rejection = 10 percent vehicle power penalty in lb/watt; pressure losses other than NaOz bed=0.8- in. of H20. The subsystem accessory weights are in the neighborhood of 8 pounds for a three-man system. Figure 61 is a plot of the subsystem equivalent weight for sodium superoxide as a function of time for one to five crew members. There was no material balance credit for water or oxygen and power penalties of 0.1 to 0.4 lb/watt were used. In figure 62 the same parameters are FIGURE61.-Equivalent weight for sodium super- plotted with credit for oxygen and a penalty oxide subsystem with no credit for oxygen for water upon assuming a 0.171 lb/man-hr water production. (AFTER ROUSSEAU ET AL.BJ) 64 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 400

300

-n E ._07 a, 3 E: 200 -a, ._m5 /I LiOH 3 w0-

100

I I I I 0 0 10 20 30 0 10 20 30 Mission duration, days Mission duration, days

FIGURE62.-Equivalent weight for sodium super- FIGURE63.-Comparison for carbon dioxide ab- oxide subsystem credited with oxygen produc- sorption subsystem (superoxide credited for tion. Power penalty is 0.1 lb/watt. (AFTER water). (AFTER ROUSSEAU ET AL.05) ROUSSEAU ET AL.05) Lithium perchlorate decomposes into lithium consumption rate by the superoxide process. chloride liberating oxygen (60.1 percent by Any less water produced by the crewman would weight). The following equation describes the be considered a penalty. chemical reaction. In any tradeoff study, a comparison must ultimately be made between the equivalent LiClO4 + LiCl+ 202 (54) weight of carbon dioxide absorption by superoxide and LiOH with water and oxygen credita- tion. Figure 63 shows this plot with the same The temperature necessary to initiate the assumptions as above and for a three-man reaction is approximately 720" F, which is cabin. This plot indicates that when oxygen is approximately 280" F higher than the melting credited to the superoxide system, the weight point of the perchlorate. A heat input of 991 penalty for equivalent carbon dioxide absorption Btu/lb of oxygen produced is required to sustain is still greater than that for the LiOH subsystem. the reaction. Much remains to be done on the optimum A problem of separating the gaseous oxygen design of the canister system. Several new from the liquid compound in the decomposition approaches to the problem show some promise in chamber arises in a zero-gravity environment. specific applications.83, Separation can possibly be achieved by use of a porous diaphragm. The rate of oxygen pro- Chlorate Candles duction appears difficult to control, since little This discussion of oxygen generation by de- is known about the decomposition mechanism. composition of lithium perchlorate and hydrogen This can probably be achieved, however, by peroxide has been taken directly from the dis- controlling the heat input into the decomposi- cussions of Coe et al. 20 and Rousseau. 95 tion chamber. ENGINEERING CONSIDERATIONS 65 Chemical sources of oxygen do find a special- Oxygen Generation by Decomposition of Hydrogen Peroxide ized application in portable life support systems Hydrogen peroxide is available commercially where storability and compactness are decisive at a concentration of 90 weight percent or advantages. In such applications perchlorates lower. A concentration of 98 percent is now are potentially promising because of high oxygen available on a semicommercial basis. Higher yield, but to date they have shown an unfavor- concentrations are desirable since they have a able side reaction ieading to the evolution of higher content of oxygen and greater stability. chlorine gas. Present chemical oxygen develop- To generate oxygen, hydrogen peroxide de- ment effort is concentrated on sodium chlorate, composes according to the following equation: NaC103. This chemical will be used as a source of oxygen in an advanced portable life-support backpack now under development. 66 A rough estimate of the weight of a simple The water produced in the reaction may be lithium perchlorate decomposition system is useful in the vehicle water-management system. shown here as figure 64. The calculations from Catalysts are required for smooth and rapid which this plot was obtained are based on the decomposition of hydrogen peroxide. These following data assumptions: materials have been thoroughly investigated in (1) Ullage: 50 percent connection with propellant uses of hydrogen (2) Decomposition chamber wall: stainless peroxide so that few problems remain. Silver- steel, 0.050 inch thick screen packs appear to be the most advanced .(3) Decomposition chamber insulation: fiber catalyst at the present time. Zero-gravity condi- glass, 1.0 inch thick tions increase the complexity of the decomposi- (4) Porous diaphragm for phase separation: tion and feed system. ceramic, 0.15 inch thick Disadvantages of hydrogen peroxide include Even with this simple system, lithium per- its high toxicity. Concentrated peroxide blisters chlorate does not appear on a weight basis alone the skin on contact. Vapors and aerosols en- to be competitive with other sources of oxygen trained with the oxygen are deleterious to the for use aboard space vehicles. respiratory system. On contact with most structural materials, the decomposition of peroxide is catalyzed. Storage in pure aluminum appears practical for long durations in vented tanks. However, accidental contamination could be disastrous. Figure 65 shows an arrangement of a peroxide decomposition system. Hydrogen peroxide is

Coolant Water separator

exchanger Pressure ?--regulator Coolant Water out out

0 50 100 150 200 Pressurizing Useful oxygen weight, Ib gas bottle

FIGURE64.--System weight for oxygen generation FIGURE65.-Chemical oxygen supply using hydro- by lithium perchlorate. (AFTER COE ET AL?O) gen peroxide. (AFTER RWSSEAU ET AL.05) 66 ENGINEERING TRADEOFFS OF 3NE- VERSUS TWO-GAS SYSTEMS stored in a positive-expulsion tank and expelled Oxygen Generation by Electrolysis of Water through a silver-screen gas generator. The water Electrolytic processes have been treated in vapor and oxygen produced are then circulated detail in the study of Coe et al.*O The electrolytic through a heat exchanger where the water vapor cell considered here is an ion membrane type is condensed and subsequently removed as a cell which appears promising for zero-gravity liquid. operation. The temperature of the gas at the outlet of the Reported parameters for a system satisfying gas generator depends on the heat leaking from the oxygen metabolic requirements of a three- the generator and on the purity of the hydrogen man vehicle are as follows: power input, 702 peroxide used. The adiabatic temperature of watts; weight, 112 lb; and volume, 1.97 ft3. decomposition of a 90-percent purity solution is It is believed that the voltage across the cell estimated to be 1364" F. electrode was on the order of 1.8 volts, and that An estimate of the weight of the hydrogen the gases were delivered at approximately 50 peroxide storage vessel and pressurizing system psia. The heat rejected by the three-man cell was made based on the following assumptions: is estimated to be 447 Btu/hr, and the cell Spherical storage vessel operating temperature is estimated as 122" F. Hydrogen peroxide purity: 0.90 Water is pumped to the electrode by means of Tank material: aluminum a wick. Utilization efficiency: 0.98 The weight of an ion-membrane electrolytic Pressurizing gas: helium cell, including a positive expulsion type water Pressurizing gas subsystem control valves: storage subsystem, has been estimated from the 7.0 pounds data given above with the following assumptions: System is credited for the water evolved (1) Water storage vessel pressurizing gas: in the reaction helium The results of these calculations are given in (2) Pressurizing gas subsystem control valves: figure 66, where the hydrogen peroxide storage 7.0 pounds vessel weight penalty is plotted against the useful (3) Hydrogen is discharged overboard and no oxygen load. It should be noted that the weight credit is given for its production penalty plotted is lower than the tankage and (4) Water storage tank material: aluminum pressurization subsystem weight by 1.125 pounds The results of these computations are plotted in per pound of oxygen generated due to the credit figure 67. given to the water of reaction which is potable. This oxygen production technique does not appear competitive on a weight basis with the 20 2.0 18 3= 1.8

16 -&2 0 J = 1.6 E 14 P EK 1.4 s !z 12 s 1.2 1.0 2 4 6 810 20 40 60 100 200 400 600 1oM _.Iin Useful owen load, Wu, Ib 2 4 6 810 20 90 60 100 200 400 603 lCW Useful oxygen load, Wu, Ib FIGURE66.-Storage weight penalty for hydrogen peroxide. Hydrogen peroxide tankage is cred- FIGURE67.-Electrolytic cell subsystem weight ited for the water of reaction. W=,o/W,=1.125. penalty. Weights include valves and control (AFTER ROUSSEAU ET AL.~) system. (AFTER ROUSSEAU ET AL.OS) ENGINEERING CONSIDERATIONS 67 other storage methods discussed previously, a large percentage of the total system weight of especially for short-duration missions. In addi- the gas supply, comparison of the various storage tion, the high power requirements of the elec- techniques discussed previously can only be trolytic cell are presently a serious disadvantage made on an integrated basis. for space-vehicle installation. The weight and size of the accessories are, in general, independent of the size of the storage Nitiogen Gexeration by Y"ecorn;tcsiticn cf Lithium .4zide vessel and, in most cases, of the delivery flow On decomposition at 550" F, lithium azide rates. While this is true of valves and sensors, it yields 85.8 percent nitrogen and 95 Btu/lb. To does not apply to items such as heat exchangers provide heat for initiation and maintenance of where weight is a direct function of the flow rates decomposition, a reactant can be provided which in the system. Table 19 is an example of current produces a more stable lithium compound as system component weights for several gaseous reaction product. Oxidants such as lithium nitrate and cryogenic systems. These must be added to or fluorocarbons have been used to form nitrogen- the appropriate storage tradeoffs for total-system producing solid propellants. In the case of lithium integration. Sometimes these weights are in- nitrate, the chemical reaction is given by the cluded in the basic tankage.12 following equation: Subsystem Comparison of Weight Penalties

5LiN3 + LiN03 + 3Li20 - 8N2 + p720 Btu/lb (56) Since weight is usually the determining factor in the selection of a space vehicle system, a This reaction produces 71.5 percent nitrogen. plot of several subsystem weights for gas supplies For practical use, extensive filters must be pro- considered in this report is presented for com- vided to remove lithium oxide; and heat ex- parison in figure 68. This plot has been prepared changers, to cool the product nitrogen. Because using the accessory weights of table 19 and the the reaction operates more smoothly at elevated total storage vessel curves of figures 34(a), 36(a), pressures, the nitrogen-producing mixture would 66, and 67. The curves are given for oxygen only, be used to pressurize a small storage tank, the since the conclusions drawn from it also apply to required nitrogen being obtained through a nitrogen storage vessels. pressure-regulating valve on the tank. A major Other parameters of importance in system disadvantage is the fact that the nitrogen-produc- selection are listed in table 20. These include ing azide mixture burns like a solid propellant maximum subsystem pressure (excluding the and is not amenable to simple control. pressurizing gas components), maximum esti- Material balance alone, without considering any weight penalty for the storage of the azide 3.0 or the disposal (or storage) of the lithium oxide, 2.8 shows a weight penalty (WJWNJ equal to 1.14 and --.%= 2.6 is not competitive with the cryogenic methods 2.4 described previously. In addition, the reaction is g 2.2 difficult to control and presents a safety problem 2.0 E 0. which makes the process prohibitive for space- 'f 1.8 vehicle use. -5 1.6 2 1.4 COMPARISONOF GAS STORAGEAND SUPPLYSYSTEMS VI 1.2 Component Integration 1.0 2 4 6 8 10 20 40 60 1W 200 400 6W 1wO In addition to the storage vessels and their Useful fluid load, W,. Ib pressurization subsystem, other components, FIGURE68.-Comparison of gaseous oxygen supply such as valves and heat-transfer equipment, are subsystem weight. Weight penalty for high- integral parts of the complete gas-supply sub- pressure gas storage subsystem=3.46+7.7/ W.. systems. Since these accessories can contribute (AFTER ROUSSEAU ET AL.05) 68 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

TABLE19. -Typical Gas supply Subsystems Accessory Weights [AFTER ROUSSEAU ET AL.S~]

Accessory Weight, Ib Accessory Weight, lb

1. High-pressure gas storage, fig. 26: 8. Subcritical cryogenic storage, figs. 36(b) (a) Pressure relief valve ...... 0.3 and 43(6): (b) Fill valve ...... 1.1 (a) Helium pressurization subsystem (ac- .4 counted for in storage vessel weight) .2 (b) Transducer...... 0.4 (e) Filter ...... 1 (c) Fluid quantity indicator...... 4 (f) Check valve ...... 2 (d) Vent valve ...... 2 (g) High-pressure regulating valve ...... 9 (e) Fill valve ...... 3 (h) Demand pressure regulating valve ...... 1.4 (f) Vessel shutoff valve ...... 2 (i) Partial pressure sensor...... 2 (g) Heat exchanger ...... 7 fj) Flow controller...... 2.5 (h) Check valve ...... 2 (k) Shutoff valve ...... 4 (i) Demand press e ...... 2X1.4 fj) Partial pressure sensor ...... 2 Total weight ...... 7.7 2.3

2. Supercritical cryogenic storage, figs. 34(b) and 39(b): Total weight ...... 8.3 (a) Vessel vent and pressure relief valve 0.2 (b) Pressure controller ...... 4 . Hydrogen peroxide storage, fig. 66: (c) Electrical heater ...... 8 (a) Helium pressurization subsystem (ac- (d) Internal heat exchanger ...... 9 counted for in storage vessel weight) (e) Fluid quantity capacitance sensor .5 (f) Capacity indicator...... 6 (g) Fill valve ...... 3 (h) Vessel shutoff valve ...... 2 (e) Flow control valve ...... 1.5 (i) External heat exchanger .5 (f) Partial pressure sensor ...... 2 6) Flow control valve ...... 1.2 (g) Flow controller...... 2.5 (k) Check valve ...... 1 (h) Gas generator ...... 2.2 (I) Check valve...... 2 (i) Heat exchan ...... 2.5 (m) High-pressure ...... 2x .9 ...... 1.3 (n) Demand pressure regulating valve ...... 2 X 1.4 (k) Pressure regulator ...... 1.4 (0) Partial pressure sensor ...... 2 (p) Flow controller 2.5 12.3 (4) System shutoff valve ...... 4 (r) Pipes and fittings ...... 50 (s) Bosses ...... 30

Total weight ...... 14.4

mated temperature in the subsystem, power considered, except possibly superoxides. This requirements, the heat evolved, and the heat- weight advantage increases markedly as the sink potential of the subsystem. Also listed in capacity of the supply system increases. At a table 20 is the water consumed (or produced) by total fluid load of 100 pounds, the weight of the the subsystem. Complete system evaluation can- two cryogenic subsystems is about the same. not be made without taking into account these Above 100 pounds of fluid storage capacity, the parameters. Normally, the gas supply subsystem subcritical system is slightly higher than its would be penalized (or credited) for each one of supercritical counterpart. Below 100 pounds, this the items listed above. weight picture is reversed. The weight difference On a weight basis, cryogenic fluid-storage sub- is so small that system selection must be based systems are superior to all the other subsystems on other considerations, such as the mission du- ENGINEERING CONSIDERATIONS 69 TABLE20. -Oxygen Gas supply Subsystem Characteristics [AFTER ROUSSEAU ET AL.95] I Maximum Maximum Power Heat evolved, Heat sink poten- Subsystem system pressure, system tempera- requirement, Btu/lb 02 tial, Btu/lb 0, psia ture, "F watts/lb 02 I High-pressure gas...... Ambient...... None None None Supercritical storage...... 875 Ambient...... None None 90 at 270" R a Subcritical storage...... 100 Ambient...... None None 155 at 170" R Hydrogen peroxide 50 1360 None 2610 None decomposition. Water electrolysis...... 50 122 I 117 75 None

a Average. Pressurizing gas stored at 2000 psia. ration, standby time, repressurization sequence, Considerable savings may be gained by inte- and cabin volume, which were discussed above grating both gaseous (table 16) and cryogenic in the appropriate sections. storage with the other cryogenic materials stored The weight penalty of the high-pressure gas on the spacecraft for engineering needs. A very storage subsystem for oxygen and nitrogen is comprehensive review of this approach is avail- approximately three times as large as that of able.I04,57,33,45 cryogenic subsystems at large fluid loads. For Several tradeoff studies comparing gaseous this reason, high-pressure gas storage vessels and cryogenic oxygen storage with superoxide are not very attractive for space vehicle appli- systems have been recently performed. The one cations other than emergency or repressurization using the most recent superoxide technology is gas supply. In this case, maximum reliability, shown in table 21, which compares the weight indefinite standby periods, and short-duration and volume characteristics for 90-day missions usage are the design criteria, and weight is a sec- of several different systems. These are based on ondary consideration. At a stored weight lower a man consuming 1.87 lb/day. The fixed weight than approximately 6 to 8 pounds, the high-pres- and volume estimates were said to be based sure oxygen and nitrogen gas subsystems show on the same criteria as the present study?O*959 98,84 a lower weight penalty than all the other subsys- but no specific choice of penalties was stated. tems analyzed.95 The sodium superoxide system compares favor-

TABLE21. - Weight and Volume Characteristics for Advanced State-ofthe-Art, Nonregenerative Air Revitalization Systems [AFTER PETROCELLI 81]

System Fixed weight, Ib a 90 man-day mission Fixed volume, ft3 a 90 man-day mission total weight, Ib total volume, ft3

LiOH (alone) ...... 12 269 8 Liquid oxygen (150 psi) (50 percent 20 209 11 loss). 02spheres (3000 psi) (SAE 4.340 10 534 13 steel, safety factor 1.88). Liquid (50 percent loss)/LiOH...... 32 667 19 02 spheres/LiOH ...... 22 803 21 KO,...... 12 769 17 Na0 ...... 12 598 13.5

a Fixed weight and fixed volume estimates include blowers, manifolds, regulators, control device, and miscellaneous piping and tubing. 70 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS ably with the liquid-oxygen (50-percent loss)/ quirement of nonregenerative systems is on the LiOH combination. However, a 50-percent loss order of only 0.1 kilowatt, and only 30 pounds for of liquid-oxygen is assumed with no justificatio; power supply need be added to their basic system for this value. From the discussion of cryogenic weight. Figure 70 presents the system volume systems, it would appear that little or no venting figures as a function of mission length. would be necessary for 90-day missions with even These figures imply that an active chemical the older subcritical oxygen systems. Subcritical system employing sodium superoxide offers the vessels with vapor-cooled shields, designed for least weight and volume penalty of any of the two men, would definitely require no venting nonregenerative systems considered and may be over a %-day period.= Also, from consideration competitive with regenerative systems for mis- of the section on gaseous oxygen spheres, use sions up to 125 to 180 man-days in length. The of a higher pressure (loo00 psia instead of 3000 same objections to the assumed 50-percent loss psia) for the oxygen in the oxygen spheres/LiOH in liquid oxygen and 3000 psia storage hold as weight column may make this system more above. Comparison of the superoxide versus re- competitive. generative system appears more valid. In view of the uncertainties of the data, it is Except for the superoxide system discussed felt that final judgment should be reserved until previously, none of the chemical systems are more concrete tradeoffs regarding the superoxide competitive on a weight basis with either of the system are available. Table 21 has been included cryogenic storage techniques. No suitable chemi- to show the great importance of one’s initial as- cal-generation method for nitrogen supply has sumptions in estimating storage-systems weight been found to date. penalties when making comparison with other Oxygen generation from hydrogen peroxide systems. decomposition could find a use in vehicles in Figure 69 represents a time-dependent study which the water-management subsystem shows of the same type with estimates for a three-man a water deficit. This situation would occur in total regenerative system in the 280- to 600-pound missions of short duration when water is not range. Included is the estimated power require- recovered from the waste or wash products. ment of 2 kilowatts. Although it is not stated, it In this case, however, the small requirement is presumed that a nuclear power source is be- for gas-storage capacity and the high weight of ing used to give a power penalty of 300 lb/kw. the hydrogen peroxide subsystem offset the ad- Approximately 600 pounds must therefore be vantages of water generation as a byproduct of added to the basic estimated weights of the re- oxygen production. In addition, hydrogen- generative systems. The estimated power re- peroxide systems for a breathing oxygen supply

24 0,- 20 16 -3 2 12 585 m regenerative systems, 4 200tp+*u

0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100 120 140 160 180 200 Mission length, man-days Mission length, man-days

FIGURE69.-Variation of weight penalties with FIGURE70.-Varia tion of volume penalties with mission length. (AFTER PETROCELLI.~~) mission length. (AFTER PETROCELLI.~~) ENGINEERING CONSIDERATIONS 71 , are comparatively underdeveloped, are not these parameters, the optimum combination of easily controlled, and lack safe operational parameters for a zero-g, shirtsleeve, artificial characteristics. atmosphere environment differs from the com- Because of the high electrolytic cell weight, bination of parameters considered standard for oxygen production from water by electrolysis a sea-level, 1-g, comfort environment. In addition is attractive only for missions in excess of 1 to weight penalties for average comfort tem- year. The weight of this system (fig. 67) greatly peratures, other factors to be considered are the improves if, in, the overall vehicle material requirement for accommodating high metabolic balance, an excess of water is produced which rates for short periods of time; the need €or simple can be used for the electrolysis process. The and responsive controls; and the need for air weight plotted can, in this case, be reduced by the circulation for subjective comfort, as well as amount of excess water production. In a mission for removal of waste products and contaminants of this type, it is very likely that vehicle electrical from the atmosphere. power would be derived from nuclear or solar Several previous studies form a basis for the sources and that a very low penalty would be present review. The analytic procedures devel- paid for supplying power to the electrolytic oped by Coe et al.?O Rousseau et al.,959 96 Pase1k:o cell. This type of gas supply system, therefore, Boeing Co.,12 and Johnston48 form a foundation offers potentialities for long mission duration, for the present study. Computer codes using especially if hydrogen is required for carbon many of these analytic techniques are available. dioxide reducti0n.~5 It is obvious that many of the weight and power tradeoffs depend on the integration of the en- Air-conditioning System vironmental control system with other systems The design of the “air-conditioning” system of the spacecraft.80 The type of spacecraft of a space cabin depends quite clearly on the radiator system, the degree of integration of nature of the atmosphere being studied. Comfort cooling modes for equipment and crewmen, of the crew is the prime consideration. Once the the power source, and myriads of other factors physical requirements for the comfort zones determine both the absolute power and weight have been established, one can proceed in an penalties and the relative penalties for different orderly fashion to optimize the equipment gas mixtures. Because such a complex analysis required for establishing these requirements. is far beyond the scope of the present study, it The basic physiological considerations have was felt that a valid approximation would suffice already been covered in chapter 1. In this section, if typical integration factors and weight penalties the interaction between these physiological for power could be entered into the study at requirements and engineering variables is appropriate points. It is clear that estimation of discussed. A parametric analysis of the several these factors is the weakest part of the analysis. weight-limiting functions of the system is made Wherever possible, the errors and biases pre- for the five acceptable gas mixtures. The final sented by these integration factors are pointed product is a specific power and weight tradeoff out and the sensitivity of the specific tradeoff for the air-conditioning subsystem of a two-man to these factors is estimated. orbiting laboratory system with 30-day To simplify the gas specific tradeoff factors, capabilities. the Boeing Co.12 has broken down a typical environmental control system into a few major GENERALCONSIDERATIONS component parts. Figure 71 represents this Chapter 1 demonstrated that the parameters of breakdown. The atmosphere control fan moves radiant wall temperature, air temperature, rela- air through a water removal, C02 removal, and tive humidity, and air circulation may be com- trace contaminant absorption system. The suit bined in several ways to produce a comfortable circuit runs parallel to this. Characteristics of crew environment. Because of the relative power this system are low flow and high pressure drop penalties and equipment weight associated with (AP). The AP of the suit is one of the most 72 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 0 (1) WH is the system hardware weight com- prising heat exchangers canisters, valves, ducts, etc. This weight is the actual system weight, Suit including all its components and associated

Atmosphere control fan - Low flow hardware such as sensors and system-flow HighAP controllers. Cabin temperature control fan - LJ r- High flow (2) WPis the weight of the vehicle power source Low A P chargeable to the system under consideration. Ventilation fan - It can be expressed as the product of the system High flow Low A P power requirement by the vehicle power penalty. The system power requirement includes FIGURE11.-Typical environmental control sys- the power expended to circulate the process air tem. (AFTER BOEING.~~) through the circuit, the power necessary for system control, and the power required for heat- critical factors. If there is an emergency mode ing or any other process used in the system. with the requirement for the cabin blower to The vehicle power penalty depends mainly handle the suit loop with its high AP, the weight on the size of the vehicle power installation and penalty increases greatly. The cabin temperature- on the duration of the mission. Nuclear or large control system has a fan of high flow and low solar power sources for short missions have AP passing air through a heat exchanger. The relatively high specific weight, with a minimum cabin ventilation circuit is also of high-flow, of 400 lb/kW of installed power. On the other lowdP type. The major problem at hand is to hand, for long mission durations, the nuclear or determine the weight, power, and volume trade- solar power penalty is lower, in the range of offs of such a system and to relate them to dif- 200 to 300 lb/kW. This difference is a major ferent atmospheres in question. The criteria of factor in overall tradeoffs using these systems. cost, reliability, development status, and inter- (3) WQ is the weight of the vehicle cooling faces with other systems must also be eval- system that can be charged to the particular uated. Rousseau et al.S5 have summarized the system or subsystem considered. WQis the prod- major criteria in the following paragraphs. uct of the vehicle cooling system specific weight in lb/W, and the system heat rejection load in Weight watts. However, this penalty depends not only Weight must include the weights of fixed on the size of the heat load but also on the tem- equipment, ducts and connecting fixtures, any perature level at which the heat load is rejected supplies, such as activated charcoal, and silica to the cooling system. This temperature level gel, necessary to the operation of the system, as must be taken into account when determining the well as related control mechanisms and instru- term WQ. mentation. In addition, the power requirement As mentioned above, final systems integration is considered in terms of the weight required for is a most critical aspect of this problem. An the power sources, of whatever kind. As seen example of the mission dependence of vehicle below, the weight equivalent of power is the com- cooling system is shown in figure 72. Tl& figure mon base for gas-specific tradeoffs. Competing represents different heat sink potentials as a systems and subsystems are usually compared function of mission duration and thermal load. on an equivalent weight basis. The equivalent Hydrogen must be treated as an expendable when weight is made up of several terms and is defined it is also used as a fuel. In the shaded area, active algebraically by the equation: radiators can be used if power is supplied by fuel cells; passive radiators if power is supplied by solar cells. Thermal transport may be provided by a The terms of equation (57) are in turn defined coolant fluid circulated between the heat pro- and discussed as follows: ducing equipment and the heat sink, with the ENGINEERING CONSIDERATIONS 73 00 The preferred type of power will depend upon the design of the equipment and upon the relative availability of the different types. For contin- uously rotating devices, such as compressors, electrical power may be better than pneumatic, while for periodically actuated devices, such as T-Heat of fusion I water separator sponges or control valves, pneumatic power may have distinct advantages. One must consider off-design as well as on-design Passive radiator systems modes. Both the maximum rate at which power IO 100 1000 10000 will be used and the average rate must be con- Mission duration, hr sidered. The penalty imposed by any power source will be a combination of the influences of FIGURE72.-Heat sinks as a function of mission the maximum rate and the average rate times duration and thermal load. (AFTER ROWLETT the use. AND LEE.87) Two philosophies can be used in computing heat transferred from the equipment to the power.95 The first, which is machine oriented, is coolant through cold plates or through circulating to set up the flow circuit and then compute the pressurization gas and then transported to a radi- total gross power required by the components ator or evaporator. Electrical equipment alterna- to maintain operation of the circuit. In practice, tively can be cooled by forced convection of the one assumes a flow rate, composition, density, pressurization gas, which, in turn, generally is and temperature. The pressure drops of the cooled in a heat exchanger using an expendable individual equipment and ducts are computed evaporant as its heat sink. and added to give the total pressure drop in the (4) WWAT depends on the system material circuit. A compressor capable of providing a balance. As an example, if a nonregenerable specified flow rate at a pressure rise equal to absorbent is used for the removal of water the computed pressure drop is then chosen. from the cabin atmosphere, the weight of the The power required by the compressor (an elec- absorbent must be charged against the system. tric motor) is then said to be the power required Also, if water is used in a system, either per- to maintain the desired fluid circulation in the manently absorbed or evacuated overboard, then circuit. the system, in certain cases, must be charged The second method, which is function-oriented, for this amount of water expended. On the other determines with respect to each equipment and hand, if a system produces water or oxygen, it duct section, the power equivalent, at 100 percent can, depending on the application, be credited efficiency, of the pressure drop in the device. for the production of these materials. The resulting total power equivalent is then modified to reflect attainable power conversion Power Requirements efficiencies, and the required total power is Power requirements include mechanical or found. This function-oriented method has two pneumatic power for circulation of the atmos- drawbacks when compared with the machine- phere, heat power for use in a catalytic burner, oriented method. The resulting calculations are mechanical or pneumatic power for a water more complex without being more accurate. separator, and pneumatic or electrical power for Also, unless great care is used, violation of the operation of control elements and instrumenta- principles of continuity and conservation of tion. Mechanical power may come from electric energy may occur, resulting in meaningless motors, and heat power may come from electric values. resistance elements; that is, the entire power supply may be electrical. Pneumatic power is Volume customary in capsule pressure controls and pres- The volume of an atmospheric control system sure relief valves. is relatively difficult to determine at an early

261-559 0-67-6 74 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS stage of the program. This is due to the fact that does not present any disadvantage, since the a substantial percentage of the total volume is requirements for relative humidity are very necessarily devoted to ducts and fittings, the broad, as seen by the comfort zone definition in actual sizes of which are very dependent upon figure 3. the layout or arrangement of components. The The water recovered from the cabin atmos- total volume of the atmosphere control system phere is relatively pure when compared with can include the following: urine or wash water. To be made potable, how- (1) Core volume or volume of heat exchanger ever, it must undergo treatment. element, or volume of reacting substance, such Figure 73 represents psychrometric charts for as lithium hydroxide several pressures of oxygen which are of value (2) Volume of the supports for the core in establishing humidity-control design. Figure (3) Volume of the pans and manifolds 74 represents similar charts for gas systems of (4) Volume of associated or integral ducts nitrogen-oxygen and helium-oxygen mixtures. (5) Volume of auxiliary items, spare parts, Moisture can be removed by two methods: tools, and replacement chemicals such as lithium (1) Adsorption on to silica gel or molecular sieves hydroxide or absorption by chemicals, or (2) cooler-con- (6) Volume, space clearance necessary for densation methods. Rousseau et a1.95 have access to equipment and for repairs, on the demonstrated that humidity control by solid ground or in flight adsorbents, such as silica gel or molecular sieves, Because of the complexity of the geometries is more attractive than water removal by chemical involved, it is essential that a layout, or better a absorbents. No heat of reaction is involved in mockup, be used to arrange circuit components the process, and the heat of adsorption released for minimum volume. This requirement pre- is roughly the heat of vaporization of the water. cludes a precise evaluation at an early stage The saturated adsorbents can be regenerated by necessary for atmospheric selection, and there- adding heat to the bed at a much lower tempera- fore systems volumes are not usually consid- ture level than that required to regenerate the ered in gas-specific tradeoffs. chemical absorbents. A temperature of 250" F is usually quoted for silica gel. Regeneration also Humidity Control can be partially achieved by evacuating the bed As is discussed below, the design of the humid- to vacuum. This process, however, is relatively ity control system appears to determine much of slow, and its dynamic characteristics are not the hardware and power weight penalty of the well known at the present time but are under atmosphere control loop. A brief comparison study. It appears that heat addition to the sat- of the two most advantageous systems appears urated bed, coupled with evacuation to vacuum, to be in order. It is taken directly from Rous- would be very satisfactory for systems in which seau et aLg5 water is dumped overboard. A desorption tem- Humidity control of a space vehicle cabin perature of 150" F is sufficient in this case. atmosphere involves the removal of the water Water absorption by chemicals was covered vapor produced by the crew members. As dis- briefly in the section on superoxides. Because cussed in chapter 1 of this report, the rate of of the general inefficiency of the chemical ab- production of water vapor by respiration and sorption process from a systems point of view, perspiration varies greatly, depending on the no further discussion is necessary in the present occupants' metabolic rate and also on their context.95 activity. For normal operations the rate of water vapor emitted is taken to be an average of about Silica-Gel Adsorbent System 4 to 6 lb/man-day. If water is produced at a rate It has been shown that in general, silica gel higher than average, as in decompression modes is far superior to molecular sieves for space with ventilated suits at about 10 lb/man-day, cabin dehumidification 0perations.~5 A typical cabin relative humidity will rise slightly. This silica-gel system may be seen in figure 75. Two 75

8 < z < 76 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

I I I I I I I J ENGINEERING CONSIDERATIONS 77

\o h

hs E E h 2 c0 8 Y h .: v.: 4 8 1 0 c 2 n W TF 78 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

0 0% 0*o 0v\ w0 0rn 8 9. ~ ~~ ~~

ENGINEERING CONSIDERATIONS 79 80 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

~r~~~~~gszs0 ENGINEERING CONSIDERATIONS 81 0000 13 mMb%5:8s

hs E E O O h W .$ a, 2 u 91 zE e % E us e & m u) a E 91 hs E E O g, W1 .9 8

zu 0 1 Ln

‘Po5

ui P w E 2 a, h w0 w h va

mui

m0 0 0 0 0 0 0 0 0 0 0 0 0 \o In U pr\ N 13 se6 hpql/su!e~6 ‘lualuo~ ainls!oW 82 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS ENGINEERING CONSIDERATIONS 83 ~g~~~Hs: I. 84 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

RH ?RH sensor sensor Cmling 1 1-Heater (on) ? Water I1fluid 0cfbTrol lei separator -, 11

Moist air t-

Dry air- 4 from cabin outlet Flow control Flow control valve valve with manual override

'LHeater (off) FIGURE?5.-Humidity control subsystem for (AFTER ROUSSEAU ET AL.05) regenerable silica gel. -Power input \Solenoid actuator identical silica gel beds are required, one ad- sorbing and the other desorbing. The operation FIGURE76.-Cooler-condenser humidity control is fairly simple. When the process water concen- subsystem. (AFTER ROUSSEAU ET AL.05) tration at bed outlet reaches a certain preset value, all the valves of the system are turned 90" the surface of the cooler-condenser and is blown from the position shown and heat is applied to downstream by the air flowing through the heat the saturated bed, which is then evacuated to exchanger. Part of the liquid water droplets are vacuum. The process air is routed through the separated from the main airstream in a water other silica-gel bed. Water vapor from the satu- separator. The air is then returned to the cabin rated bed is dumped overboard. The valves are or to another subsystem for further processing. usually automatically switched at fixed time inter- The condensate is channeled to a reservoir vals. The valve-actuating mechanism can be a (shown as a bellows), pumped to the water cam shaft driven by an electric motor. management subsystem, and dumped overboard The heater provided for bed desorption can or returned to the cold side of the cooler-con- serve a dual purpose: It can be used for removing denser where it is evaporated at low pressure to the heat generated in the bed during the adsorp- provide part of the heat sink for humidity tion period, thus increasing the capacity of the condensation. silica gel. However, this adds to system compli- The system diagram illustrates a possible ar- cation and is possible only when the heat of rangement for the collection and disposal of desorption is provided by a hot fluid. Often, no the water separated, suitable for operation in suitable fluid loop at the temperature required a zero-gravity environment. The condensate is for desorption (150" F) is available aboard the ducted by means of wicks from the water sep- vehicle and electrical power must be used. arator to a bellows-type reservoir. As the water It is to be noted that water is not easily recov- in the reservoir accumulates, an electrical con- ered from a saturated bed and must be evacuated tact activates a solenoid actuator which com- overboard. If the regenerable silica-gel system presses the bellows and thus pumps the water is considered for installation aboard a vehicle through a check valve to the water management in which no excess water is produced, the system subsystem. As the bellows are depressed, the must be penalized by the amount of water contact is broken and the cycle repeated. dumped overboard. Comparison of Silica-Gel us. Cooler-Condenser Systems Cooler-Condenser System Equations for the operating parameters and A relatively simple method of controlling the subsystem weight penalties of these two systems humidity of the cabin air is to condense the are beyond the scope of this present study. The moisture in a heat exchanger and to remove the analyses of Coe et a1.20 and Rousseau et al.95 condensate from the process airstream. Figure cover these most adequately. Comparison of the 76 is a schematic diagram of such a system. ..system equivalent weights of these two processes Water from the moist airstream condenses on IS of value. Table 22 presents the comparison ~

ENGINEERING CONSIDERATIONS 85

TABLE22. -Comparison of Subsystem Charac- Rn I teristics [AFTERROUSSEAU ET AL.~S]

Parameter subsystem I I Hardware weight, lb ...... 28.1 10.3 Pumping losses, W ...... 1.55 4.81 Heat rejection load, Btu/hr, 439 ...... at 70" F. " Heat rejection load, Btu/hr, ...... 569 0 0.1 0.2 0.3 0.4 at 45" F. Vehicle power penalty, Ib/wott Heating requirement, Btu/hr, 439 at 150" F. Water balance, lb/day...... a-6.6 b+6.6 (1) No heat-rejection, heating, or water-consumption penalty. (2) No heating or water-consumption penalty. a Dumped overboard. (3) No heating penalty. Recovered. FIGURE77.-Humidity control subsystem comparison. (ROUSSEAU ET AL?') of the hardware weight, heat rejection load, heating requirement, power consumption, and subsystem valves requires a complex mechanism. water. balance for the cooler-condenser and the The number of valves, in itself, makes the system silica gel subsystem for the following typical unreliable. In addition, all the valves seal against vehicle and-mission parameters: the vacuum to which the bed is desorbed; this (1) Cabin pressure: 7 psia presents a serious safety problem. In practice, (2) Cabin relative humidity: 60 percent two valves in series would be installed every- (3) Number of crewmembers: three where. Although single valves are shown in the (4) Cooler-condenser subsystem air outlet subsystem diagram of figure 75, the accessory temperature: 45" F weight estimate is based on the use of two. The equivalent weight of the subsystem is Another undesirable feature of the regenerable plotted in figure 77 for various penalties con- silica-gel subsystem is the temperature cycling sidered. Here it is assumed that the heat rejection of the process air at subsystem outlet. At the penalty (RP), in lb per watt, is 10 percent of the start of the adsorption period, the bed is hot, power penalty (PP). near 150" F, and the process air temperature From this plot, it is seen that even in the best will rise through the bed, approaching the tem- light, the silica-gel subsystem is heavier than the perature of the bed at the outlet. As the bed is cooler-condenser subsystem for vehicle power cooled, the air temperature will decrease. The penalties below 300 lb/kW. If heat-rejection load cyclic temperature of the outgoing air depends is taken into account, and more so if the water on the bed dynamic characteristics. balance is introduced, the silica-gel subsystem Removal of the moisture from the cabin air by is not competitive with the simple cooler-con- cooler-condenser offers the possibility of inte- denser subsystem on a weight basis. gration of the humidity control and cabin-tem- Even at high vehicle power penalty, the slight perature control subsystems. This greatly re- weight advantage of the silica-gel system (in duces the installation of a number of components ideal conditions) is not enough to offset the ad- as well as control complexity. In actual practice, vantages of the simpler cooler-condenser system. these two functions, humidity control and tem- The cooler-condenser subsystem also is orders perature control, are unified and effected in the of magnitude more reliable than the regenerable same atmospheric control loop. For the purpose silica-gel subsystem. Operation of the silica-gel of clarity and to assess better the penalties in- 86 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS volved in the process of controlling cabin humid- nitrogen at 7 psia is taken as 1 and the other ity, humidity control was assumed to be divorced mixtures normalized to this value. Thus, let us from temperature control. consider the second requirement k, AP. For the same mass-flow rate and pressure drop, the PARAMETRIC ANALYSISOF GAS SPECIFIC FACTORSIN THE power required for the reference mixture varies AIR-CONDITIONINGSYSTEM . with that required for any other mixture as the Often in a study of air-conditioning tradeoff ratio of the density of the reference mixture, values, there are several core requirements pl, to that of any other mixture, p2. Since the which form the basis of the study. The more product of the factors CFM and AP both occur common factors such as equal comfort, minimum in the horsepower equation weight, availability of equipment, and reliability are usually considered. In the case of the air- 62.3(CFM) AP(in. H2) conditioning system, power and the resulting Fan horsepower = 33 0007, x 12 weight are usually most crucial. In designing for minimal horsepower required, several factors have to be held constant in any tradeoff analysis. These design requirements are volume flow, pres- this requirement has no gas factor dependence. sure drop, mass-flow - pressure drop, air velocity, Derivation of the critical gas factors is, of mass flow, heat-flow - temperature gradient, and course, crucial to the analysis. Assuming that specific humidity. 7) = 1, equation (58) indicates that fan power Table 23 represents these system require- - CFM . AP. ments. Critical for each of these requirements Parker et al.79 have shown that a system can are design factors in the form of physical prop- be analyzed for either constant CFM or constant erties of the gas and the configuration of the heat transfer rate, Q. ducts. These are also shown in table 23 along Constant CFM, constant size system. -The with the effect of the different gas mixtures on fan power is - AP. In turn, AP=the sum of the the fulfillment of the requirement. The value for duct friction and resistance due to flow transi- a mixture of 50 percent oxygen and 50 percent tions. This sum can be expressed as

TABLE23. -Horsepower Equation [AFTERBOEING 12] [Values are normalized for 50 percent oxygen and 50 percent nitrogen at 7 psia = 11

Factors Oxygen-nitrogen Oxygen-helium 3xygen system system system Requirements

Gas Duct configuration 7.0 psia 5.0 psia 7.0 psia 5.0 psia i.0 psia

CFM . AP ...... 1.0 1.0 1.0 1.0 1 .o W.AP ...... 1.0 1.43 1.68 1.82 1.38

Ai DI v...... -_ 1.0 .70 .59 .55 .72 AI Di w ...... 1.0 2.05 2.82 3.32 1.9

Q. AT...... 1.0 2.12 1.05 2.46 2.56

CFM, AP ...... -D, 1.0 .70 .59 .55 .72 D2 @...... 1.0 .73 .61 .41 .77 ENGINEERING CONSIDERATIONS 87 Since

where

L pvz hP ?f-- Fan power - CFM AP - -* D 2g PCP'

is the pressure loss due to duct friction, and and if equation (59) holds true:

k-PP AP - fpF+pF. 2g For turbulent flow, the relationship of equation is the pressure loss due to flow transition-el- (60) holds again. bows, diameter changes, and bends. The factors Since L, D,7, g, and k are constant; therefore

Fan power -fp+p.

For turbulent flow, P P

Since the first term (duct friction) should be Therefore, small compared with the latter term (flow transitions), pO.l6 can be neglected, and Cp2.M= Cp3. p0.16 Fan power - po.'g p + p - pOJ6pO.84+ p 1 Fan power - - Since the pressure drop due to flow transitions P2CP3 (latter term) should be larger than the pressure drop due to duct friction (first term), the varia- Fan power generally will be required to action of power with viscosity, p0.l6,can be neg- complish two functions: (1) contaminant re- lected and pO.8.2 can be approximated by p. moval which requires a constant CFM output, Therefore, and (2) the cabin cooling which optimally requires a constant heat-rejection capability by the ven- Fan power - p for constant flow systems. tilating fan. Thus, for the contaminant fan, power-p, and for the cabin heat exchanger, For constant heat transfer, constant size sys- fan power - l/pzCp3. tem. -Heat removal rate can be expressed by: It should be pointed out that different conclusions with regard to the role of gas-specific factors can be reached if different assumptions are made. For example, in considering flow in where AT = temperature difference of (air in - ducts and manifolds, the assumption of Parker 79 air out). Since AT and Q are constant, that friction pressure loss is small compared to duct transition pressure appears reasonable. -A For flow in heat exchangers or sorbent beds, WCp= constant =A and W = -e CP friction may be a major factor in power analysis.66 88 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS The value of specific or absolute humidity One can compare the different gas mixtures in arises in evaluation of power required for water convective cooling capacity by use of equations removal (eq. (67)). The value of r$ is most sensi- (6) and (7). If h,, is the convective heat-transfer tive to R', which is the gas constant divided by coefficient of a baseline gas and h,, is the con- the molecular weight (table 23). The R' value vective heat transfer coefficient of any other for air = 53, 02= 48, N2 = 55, He = 386. The sig- mixture, nificance of these values is presented in subsequent paragraphs.

Analysis ofthe Gas-Specik Factors of the Cabin Ventilating Fan The equations for the convective heat transfer The cabin ventilating fan must be considered of several gas mixtures can be obtained by the from several points of view: crew comfort, equip- relations in table 25. It is assumed that a mixture ment cooling, and wall temperature. Each of of oxygen and nitrogen at 5 psia has the same h, these in turn is affected by many design variables. as oxygen at 5 psia and that a mixture of oxygen Those most pertinent to the present analysis are and nitrogen at 7 psia is known. Table 25 also shown in table 24. It should be remembered that shows the calculation for the velocity, v,of the radiant cooling affects both man and equipment. several mixtures required to give the same con- Integration of devices controlling the tempera- vective heat transfer coefficient as a mixture of ture of the wall and the instrument cold plates oxygen and nitrogen at 7 psia. This is obtained with the external heat sink can be a major design from the relationships indicated by equation (6) factor. that:

TABLE24. -Major Factors Influencing Power Requirements of Cabin Ventilation Fan [AFTER BOEING 12]

Crew comfort Equipment cooling Wall temperature

Sweat rate ...... Tg P Tg...... Distribution of walls

Relative humidity...... TCO"d.?"SW Relative humidity Interaction with external radiator Distribution of crew......

TABLE25. -Convective Heat Transfer Coeficients of Several Gas Mixtures and Velocities Required for h,, = h,, [AFTERBOEING 12]

Mixture I Heat transfer coefficient I Velocity

- - - Oxygen and helium at 7 psia ...... h,= 2.13 hcaN2-, (-y'5 V=0.215

Oxygen and helium at 5 psia ......

- - - Oxygen and nitrogen at 5 psia, or 100-percent oxygen at 5 psia...... h, = 0.85 h, V= 1.36 VaNP-, ENGINEERING CONSIDERATIONS 89 tive horsepower penalties for the different gas (a) mixtures. Analysis of the Cas-.Speci$c Factors Limiting the Atmospheric By determining the different velocities re- Control Loop quired for a constant convective transfer coef- In the analysis of gas-specific factors influenc- ficient, one can calculate the power required for ing the atmosphere control loop (fig. 71), one is equal convective heat loss by the following concerned with a system having a low constant- reasoning: flow rate and a high AP (about 8 inches of water). From equations (58) and (6), it is evident that The major considerations are water absorption, fan horsepower HP is - CFM * AP and CFM - 7. trace contaminant removal, carbon dioxide re- Since AP - p, HP - pr Table 26 indicates the moval, and suit circuit. Table 27 outlines the values of k, p, and 7 for constant h, and the rela- major factors which influence these considera-

TABLE26. -Parameter Values for Different Gas Mixtures [AFTER BOEING 121

O,-Nz O2-He 02

7.0 psia 5.0 psia 7.0 psia 5.0 psia 5.0 psia k. Btu/hr-ft-"F...... 0.0153 0.0153 0.0386 0.0286 0.0155 p,- lb/ftJ...... 0.0365 0.0268 0.022 0.0206 0.0279 V. ft/min ...... 47 64 12.5 25 60 Power, watts ...... 63 62 10 19 61 Relative power ...... 1 0.98 0.16 0.30 0.97

TABLE27. -Major Factors Influencing Power Requirements of dtmosphere Control Fan [AFTERBOEING 121

Water Trace contaminant coz Suit

W"*O Metabolic rate k't, Material, man WC, Diet TIC Chemical processes in Metabolic rate Tg, P, k, R man and materials Leakage Clo Atmospheric gas v r) Type equipment Radiation r) Leakage CFM Relative humidity Equipment COZ level Leakage Gas

AP PI Gas Relative humidity TIl

Type separator w Relative humidity

261-559 0-67-7 90 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS tions, the required mass gas flow (Wg),and, ultimately, the horsepower requirements of the atmosphere control fan. The general equations which determine the relationships are also indicated. There are many obvious interrelationships between components within the equation. For instance, the efficiency of the water separator 7 will ultimately influence WHz0and AH. Since the cabin calls for a shirtsleeve environment, no suit factors are considered in the analysis of the cabin tradeoffs. It is assumed that the suit cir- t * 3” cuit will have its own fan and power. A specific analysis of the absolute weight of cabin gas needed to handle the different loads of the atmospheric-control system is beyond the requirements of this paper but may be found in Coe et a1.20 and Rousseau et al.95~96From these considerations, the estimated flows of an oxygen- nitrogen mixture at 7 psia required for the several control functions are presented in table 28. It can be seen that the water removal requires FIGURE78.-Mass flow of gas required to re- the greatest mass flow of cabin gas. The value move 6 pounds of water per day at dif- for trace-contaminant removal from a “dirty ferent gas temperatures and relative humi- dities. T,,.~=5O0 F. (AFTER BOEING.~~) cabin” is the highest expected. The clean cabin represents the “average space cabin.” specific relative humidity curves noted. The It is clear, therefore, that the removal of water higher the relative humidity, the less the mass from the cabin atmosphere will require the flow of air required per mass of water removed greatest mass flow of gas and will determine the per day and the lower the power requirement. power requirements of the atmosphere control The relative power requirement for water re- loop. Figure 78 indicates that as the temperature moval from the several gas mixtures is the next of the gas increases, the weight of gas which step in the tradeoff analysis. A baseline power must be blown through a condenser-cooler with value for an oxygen-nitrogen mixture at 7 psia fixed outlet temperature of 50” F to remove 6 can be determined from the following assump- pounds of water per day decreases along the tions and calculation of power:

TABLE28.-Estimated Mass Flow of Gas Re- W= 1 lb/min, empirical resistance quired per Day for Atmospheric Control in a constant = 3.9 2-Man Cabin [AFTER BOEING 12] AP = 8 inches of H20 Power = 0.0001575kg, AP= 100 watts Function Mass flow of gas, W,,Ib/min These assumptions should handle the assumed water load of 6 lb/day for two men which is a Water removal...... 0.6 to 1.0 borderline low value. Comparative power for Trace contaminant removal: other atmospheres can be determined from equa- Clean cabin ...... 0.4 tion (58) which indicates that HP - CFM AP. Dirty cabin ...... 1.0 The mass flow of gas required to remove water COt removal...... 0.23 to 0.40 Suit loop (liquid cooled- gas aug- 0.5 from any atmosphere is inversely related to the mented, 6 CFM). specific humidity of the atmosphere. Therefore, I comparing any two gases: ~~ ~

ENGINEERING CONSIDERATIONS 91 50-percent nitrogen at 7 psia. It is obvious that many second and third order interactions have been omitted. However, except for very unusual The relative AP for a gas flow system is related design limitations, these interactions should not to the kgas follows: greatly influence the relative tradeoff values.

Absolute Weight Penalties for the Air-ConditioningSubsystem AP-- w; The next step in the process of establishing total system weight penalties for different gas mixtures in the design of air-conditioning systems is to determine the absolute penalty for Since relative power for any gas may therefore the reference mixture, oxygen and nitrogen, at be determined by the relationship, power 7 psia. At this stage of the tradeoff analysis, - WAP/p, some basic assumptions must be made about the volume of the craft, crew size, and mode of heat sink and power sources. The complex relationships between these variables have been discussed above. In order to establish these interactions, a systems analysis of the entire vehicle is required. The Boeing CO.'~ has performed such an anal- The final step of this equation is developed in ysis for a two-man, 30-day, orbiting laboratory. equations (68) to (72) on page 105. The study has included the integration of cryo- The specific humidity, 4, is directly propor- genic, power, and environmental control systems tional to R'. The power required for any gas to with an analysis of the sensitivity of the power- remove water from the stream under the assumed weight penalty of multiple changes in design. conditions of flow can be calculated from equa- A critical factor in this analysis was the determi- tion (67) relative to an oxygen-nitrogen mixture nation of a conservative value of 1.25 lb/watt for at 7 psia=100 watts. Table 29 shows this rela- a 30-day mission. Depending on specific design tionship. limitations relating to power source-cryogenics Table 29, along with tables 23 and 26, shows interactions, the calculated 30-day penalties the relative factors required to perform power ranged from 0.7 to 1.14 Ib/watt. An analysis for and other tradeoffs in the design of an air-con- a similar system by AiResearch Corp. arrived ditioning system for a space cabin of any type at 1.2 lb/watt as the power weight penalty. Since once the appropriate values are known for a these two independent groups agree so well, reference gas such as 50-percent oxygen and it is safe to proceed with a conservative value of 1.25 lb/watt. TABLE29. -Power Required To Remove Water It is felt that a step-by-step analysis of this From a Gas Stream [AFTER BOEING 12] absolute power penalty will be of great value. [Relative to 7-psia oxygen-nitrogen system = 100 watts] The internal tradeoffs between the critical thermal control, dehumidifying, and ventilating Power, watts loops have been very graphically presented by System the group. It is only through such an analysis that one can appreciate the sensitivity of the

I I total air-conditioning power penalty to on-design assumptions and off-design contingencies. Oxygen-nitrogen...... 100 72 Since these factors ultimately play a significant Oxygen-helium...... 60 53 Oxygen ...... 72 part in the evaluation of any inert gas tradeoff a detailed review of the study appears warranted. 92 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS The model of physiological heat transfer in for on-design, average cabin conditions. The chapter 1 was used to calculate the heat transfer lower Clo value tends to increase the comfort- . by each mode for a range of parameters of radiant zone temperature and decrease the power penalty wall temperature, air velocity, air temperature, of the atmospheric cooling system (fig. 15) for and effective wetted surface under a fixed met- all gas mixtures. It is certain that the region of abolic rate. A power weight estimate was then from 0.05 to 0.15 Clo will give a minimum sys- prepared for each mode of heat transfer. The tem weight for air conditioning. equipment weight was considered to be the same Heat balance. -Figure 79 represents the radi- for the range of parameters investigated. The ant, convective, and evaporative heat rejection most promising alternate combinations were partition which will be required for a crewman examined to determine their sensitivity to higher working at 520 Btu/hr with a wetted skin fraction and lower metabolic rates for off-design peaks of 10 percent and a clothing temperature of 89" F in metabolic loading. The power weights then in a 7-psia oxygen-nitrogen atmosphere for sev- provided a means for evaluating the weight impli- eral air velocities. Lung losses were not included. cations of the various gas mixtures. It can be seen that increasing the air velocity The following assumptions were made: from 20 to 80 ft/min at 70" F will double the con- (1) A balance is required between body heat vective heat loss (fig. 79(b)). Also, decreasing production and rejection to keep the body at an average skin temperature of 91" F without sweating or shivering. 400300 (2) To cover a variable exercise load in the - 11 orbiting laboratory, the average heat rejection capability of 490 to 520 Btu/hr must be provided in the design. The value of 400 Btu/hr accepted " as the average sedentary level was too low for 70 75 80 the laboratory operations specified in the mis- (a ) Wall temperature, OF sion description. (3) During normal activity the humidity must be maintained between 40 and 60 percent. f (4) Crew comfort .must be provided for in z 200 either a 5-psia or 7-psia environment. The lab compartment in a cabin is about 750 ft3; the lock a" 100 compartment is 450 ft3. I I (5) Body surface of a crewmember is 20 ft2. 70 75 80 (6) The clothing insulation of the crew can be (b) Air temperature, OF adjusted from 0.05 to 0.15 Clo. This selected value was on the low side. The underwear worn in Gemini 7 had a Clo value of "less than 0.25 C~O."~~During the waking hours, this was the preferred mode of dress for general comfort. For sleep, the crew preferred a double layer of underwear and a coverall to protect them from the cold in the dark portion of the orbit. Gen- 250 175 80 eral sensitivity of the comfort zone to Clo value 70 can be estimated from the oxygen-helium pre- (C ) Air temperature, OF dictions of figure 15(a). Since future space cabins FIGURE79.-Heat rejection modes. Q,=520 with more operating volume will probably have Btulhr; f,=lO percent; T,=89" F; P=7.0 psia. astronauts in coverall garments, a Clo value of (a) Radiation. (b) Convection. (c) Evapora- 0.5 would probably have been more appropriate tion (lung loss not included). (AFTER BOEING.~~) ENGINEERING CONSIDERATIONS 93 the air temperature from 80" to 70" F at constant function of air temperature and wall temperature air velocity will double convective heat loss. are ghen in figure 81. The increase in flow re- Air temperature has little effect on evaporative quirements with a decrease in air temperature heat loss in the temperature zone of 70" to 80" F is caused by a decrease in the specific humidity of vasomotor regulation. This can be seen more of the lower temperature air. Because of the in- clearly in figure 11. Figure 79(c) does show that creased heat lost via radiation and decreased loss increase in air velocity from 20 to 80 ft/min re- of latent heat, the inflow requirements are de- sults in a twofold increase in evaporative loss creased with a decrease in wall temperature. (see also fig. 10). The values of figure 79(a) ap- Both the power weight required to provide the pear very much lower than the corresponding necessary air flow through the humidity control values of figure 10 but it must be remembered to achieve the necessary evaporative heat rejec- that the former is assuming 10 percent skin wet- tion rates and the power weight to provide the air ting while the latter assumes 100 percent wet- circulation to achieve the necessary convective ting. A mass transfer coefficient of heat rejection rates at three different wall temperatures are shown in figure 82. The power weight penalty of 1.25 lb/watt for the 30-day mission was used. The combined power weight was used for these calculations (see eq. (12)). penalty is shown as a function of air temperature The analytical approach is to select an air and wall temperature in figure 83. An air tempera- temperature, wall temperature, and relative ture of 73" to 75" F results in the lowest power humidity and to determine the radiant loss (fig. penalty for the range of wall temperatures 79(a)) and the air velocity required to provide a considered. balance of heat loss through convection (fig. 79(b)) and evaporation (fig. 79(c)). The resultant air velocity requirements are shown in figure 80 3.0 as a function of air temperature and wall temperature. Minimum air velocity requirements are realized with low air and wall temperatures. Air flow requirements through the humidity control system (outlet temperature of 45" F) as a 2.0 100 m E .-c .E 80 P c -d .- L -...E ._I ;= 60 a 1.0 .--s -8 % 40 .-L a 20

0 65 70 75 80 " ~ ~~~~ 65 70 75 80 Air temperature, "F Air temperature, "F FIGURE 81.-Humidity control requirements. FIGURE80.-Air velocity requirements. Q,=520 T,1=89" F; heat production=590 Btulhr; P= Btu/hr; RH=50 percent; P=7.0 psia; f,=lO 7.0 psia; T,,,d=45° F; and f,=lO percent. percent. (AFTER BOBING.~*) (AFTER BOEING.~~) 94 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

300 - 240 I\ I I I 85

Gntrol tolerance- 250 - 200 Maximum mean radiant temperature control tempemture P -- 200 - 160 6 zrn Humidity control 75 8 I Minimum mean radiant temperature control tempemture 150 - 120 0LL 8m .- : E Control tolerance7 t n. 2 70 7 g 100- 80 D t Allowable temperature spread below minimum L 2$ 1 meon radiant temperature 50- 40 ----- Minimum local equipment/wal I 60 temperature Condensation margin OL 0' I I I 65 70 15 80 >I Maximum dew pint temperature Air temperature, "F 55 FIGURE82.-Humidity and air circulation power FIGURE84.-Tempera ture control margins. (AFTER requirements. Q,= 520 B tulman-hr. (AFTER BOEING.~~) BOEINC.~~)

Wall temperature. -Figure 83 indicates that equipment. From an independent study of wall the power weight penalty decreases with de- heat leak, the minimum surface temperature creasing wall temperature. Even at an optimum allowable in the compartments was selected at value of T,, 50 pounds or more of weight are at 65" F. This temperature allows a margin of 5" stake. The total weight penalty could be doubled. between equipment temperature and the maxi-

However, in the selection of wall temperature mum dew point. A k4.O allowance was made for one must consider, in addition to weight implica- locations above and below the mean radiant tions, the effect of higher, off-design metabolic temperature and a control tolerance of ?laF rates and the system control concept. Figure 84 was selected. This establishes a minimum tem- shows the allowable range of temperatures to perature of 70" F for air and mean radiant prevent condensation on the vehicle walls and temperature. Figure 85 compares the capability of two sys- 300 - tems to accommodate high, off-design metabolic rates. System A is designed for a Tu of 75" F and 250 - a T,of 70" F; system B is designed for a Tu of 5 75" F and a T, of 75". A comparison of figures -E 5 200- 85(a) and 85(b) shows that system B can accommodate higher metabolic rates than system A at 06 0 150 - equal values of fs (effective sweating area frac- 1 8 .F tion). The crew duty cycles showed that the 3 - system chosen must have the ability to absorb & loo nB 780 Btu/hr for 1.5 hours. System B accommodates this rate (fig. point with approximately 50 - 85(b), P) 20 percent of the body covered with perspiration if the wall and air temperature is 70" F. At the OL 0' 1 I I 65 70 15 80 same heat load, system A requires that approxi- Air temperature, "F mately 50 percent of the body perspire. In addition, system B accommodates the heat output FIGURE83.-Combined humidity and air circulation power requirements. Q,=520 Btulman-hr. during exercise (1400 Btu/hr). (AFTER BOEINC.*~) Evaporative losses during normal activities ~

ENGINEERING CONSIDERATIONS 95 Maintenance Exercise 100 c

V0)c L % 15 .-i xv) 50 -al 3 -0 25 3 a I I I I I 0 400 600 800 loo0 1200 1400 evaporative heat-loss zone Heat production, Btulman-hr for prolonged periods 0 400 500 600 700 800 900 IO00 Exercise Heat production, Btulman-hr 100 c eal 7 FIGURE86.-Heat rejection distribution. T.=75" z F;P=7.0psia; V.=47 ftlmin;TCi=89"F. (AFTER 75 .-r' BOEING .la) Yv) 50 s heavy sweating occurs. The exact quantitation of b 0 the fractional wetness factor in exercising indi- 5 25 viduals in altered environments requires further E U study. These periods of high activity are expected to be short, such as those in a programed exercise 0 400 600 800 loo0 1200 1400 period. During periods of lower activity, the crew- Heat production, Btulman-hr man may have to don more clothing above 0.15 Clo in order to remain comfortable, especially FIGURE85.--Wetted skin plotted against heat during sleep. production. System A: T,=75" F; T,=70" F; Va=25 ftlmin. System B: T,=75" F; T,=75" It was assumed that wall temperature can be F; Va=47 ftlmin. (AFTER BOEING.'~) controlled by varying the temperature of radiant panels mounted in the crew compartment. The with air velocity at 47 ft/min and air and wall surface temperatures of many of the consoles temperature at 75" F represent approximately and equipment cabinets can be controlled by 20 percent of the total cooling requirement indi- mounting cold plates on the cabin side of the cated in figure 86. Table 1 suggests that lightly electronic packages. Any surfaces without cold- exercising men are comfortable when evaporative plate temperature control such as storage cab- capacity is between 10 and 25 percent. The effect inets can, if necessary, be shielded from the of lowering the wall temperature during increased crewmembers by radiant panels which are kept activity is also shown in figure 86. Lowering the at the desired temperature. Since the pressure wall temperature from 75" to 70" F increases the drop in the panels is very low and the panels are radiative losses from 220 to 295 Btu/hr, thereby incorporated into the water loop, no power allowing man to work at an increased rate without penalty is assessed against the crew temperature increased evaporative losses. Beginning at a control system for radiant temperature control. work activity corresponding to a heat production The weight of the panels does not depend on the of approximately 605 Btu/hr, the evaporative operating temperature, which is controlled by losses increase to maintain a heat balance. At varying the water-flow rate and temperature. 605 Btu/hr the man's skin may be approximately Therefore, the weight penalty for radiant tem- 10 percent wet. Further increase in activity perature control is constant and does not in- results in increased wetness and eventually fluence the optimum radiant temperature. How- 96 ENGINEERING TRADEOFFS OF -ONE- VERSUS TWO-GAS SYSTEMS ever, if the wall-temperature design of 70" F is Air at 75" F and 60 percent relative humidity selected, obviously the wall-temperature could has a dewpoint temperature of 60" F. The lower not be lowered if condensation on the wall is to limit for wall temperature has been established be avoided (fig. 84). In this instance control of as 70" F as noted in figure 84. Consideration of higher metabolic rates would require a variable these two design parameters provides a 5" F air velocity. This design approach requires a margin of safety before condensation on the wall multiple-speed blower or multiple blowers which can occur. To permit humidity control over a would introduce problems of hardware avail- range of metabolic activities, it is desirable to ability, development cost, reliability, and system control the condensing temperature. As illus- complexity. trated in figure 87, a system designed for a rela- Returning again to figure 85, one sees that tive humidity of 50 percent and a condensing under design conditions, the system for Tw= 70" temperature of 45" F has twice the capacity when F requires an air velocity of approximately 25 relative humidity is increased to 60 percent and ftlmin and the system for Tw=75" F requires condensing temperature is lowered to 40" F. an air velocity of approximately 47 ftlmin. The Air circulation. -The power requirements for weight penalty of 20 pounds for the 75" F design the air circulation system are given in figure 82 is largely due to the power penalty associated with for a 7-psia mixed-gas atmosphere. The circula- the higher air velocity. A pertinent unresolved tion system consists of two fans and a single-fan physiological problem is the determination of heat exchanger combination. One fan is directed minimum velocity requirements under zero over each crewmember at the control console gravity for subjective comfort, aside from thermal to provide the air velocity required for cooling. balance considerations. Although there is no The fans are assumed to provide 5 ftYmin of air experimental evidence to substantiate a conclu- flow for each watt of power consumed, a value sion, it was assumed that either system A or typical of the fans considered for this application. system B provided adequate air velocity because The cabin ventilation fan-heat exchanger unit both systems met or exceeded normally accept- provides overall air circulation in the cabin and able sea-level air-velocity values (15 to 25 ftlmin). maintains the air temperature at the desired The factors influencing the selection of wall level. The heat exchanger removes the heat temperature are summarized as follows: transferred by convection to the air from the crewmembers and the heat dissipated by the System A System B two fans. The air-flow rate through the fan-heat Tp75";Tp70" Te75";T-75'

I ~ ~~ exchanger unit is determined by allowing a tem- Weight penalty, Ib ...... 0 + 20 perature drop of 10" F across the unit. Equations fi at 780 Btu (maximum 45 20 for fan design and power tradeoffs can be found sustained level), per- in the study by Coe et a1.20 cent. A single-fan circulation system is being con- Control concept for Air temperature Air and wall tem- sidered as a substitute for the three-fan system. peak metabolic rates. and additional perature equipment Air velocity, ftlmin...... 25 47

0 0.002 0.m 0.006 0.008 0.010 0.012 0.014 Water removal, Ib H?Ollb air

FIGURE87.-Humidity control flexibility. T,= 750 F; P= 7.0 psia. (AFTER BOEING.~~) ENGINEERING CONSIDERATIONS 97 Preliminary scale-model tests at Boeing have system power, total power, and total weight as- shown that a circulating-flow pattern can be suming the conservative power-weight-penalty obtained by employing a centrally located ceiling of 1.25 and the most optimistic value of 0.70 inlet with a single outlet located on the opposite lb/watt. face near the outer wall. The air from the inlet As predicted from the Q, AT requirement of flows along the ceiling at a relatively high ve- table 23, table 30 indicates that the power for locity, entraining flow from the center of the the suit and humidity control is much less eensi- chamber. The resulting circulation pattern is tive to gas specific factors than is ventilation maintained and is not destroyed by the outflow power. Analysis of the total power of the air- when the outlet is located near the outer wall conditioning system indicates that under the on the wall opposite the inlet. specific design assumptions in question, oxygen- Conclusion.-Based on this study an air tem- helium mixtures at both 5 and 7 psia require perature of 75'25" F, a relative humidity of equal power and that this power is less than that 50210 percent, a mean radiant temperature for the other gas mixtures. At least 90 watts may (wall temperature) of 75'25" F, and an air velocity be saved by going from an oxygen-nitrogen mix- of approximately 47 ft/min were recommended. ture at 7 psia to any one of the helium mixtures. This recommendation is based on the crew This represents 114 pounds for the conservative wearing clothes with an insulation factor equal power weight penalty but only 55 pounds for the to 0.10 Clo. During periods of normal activity, optimistic penalty of 0.7 lb/watt. Thus, the power the air and wall temperature are maintained at penalty conditions the absolute weight differ- 75" F and, relative humidity is a nominal 50 per- entials between gases. It is also clear that the cent. At periods of higher activity the air and wall design contigency of a suit-in-the-loop increases temperatures are decreased to 70" F. During the weight differential between gases. By in- periods of extremely high activity, such as creasing the total power penalty for two men exercise, the crew is permitted to sweat. At pe- from 112 lb/mission (fig. 83) to 163 lb/mission riods of lower activity, the crew has the option of (table 30), the suit contingency increases the adding more clothing or increasing air and wall baseline from which the other values are deter- temperature. In all cases, relative humidity con- mined. In comparing gas-specific tradeoffs, one trol is maintained by adjustment of condensing must keep such seemingly minor, yet sensitive temperature in the humidity control system. factors in mind. Thus, the mission subtask of high exercise A new approach to gas circulation in space conditions the selection of air temperature and cabins has been suggested by Keating.49 This wall temperature and mode of control during technique makes use of the energy obtained by peaks. The basic power penalty of figure 83 for isentropic expansion in the circulation process. T, of 75" F can then be assumed as the on-design The technique may decrease the fan-power point for the mission. Figure 83 indicates that penalties which can be a considerable part of these assumptions give a total power penalty of the overall weight penalty of the environmental about 56 wattslman for the water removal and control system. cabin ventilation systems. This is equivalent to RELIABILITY- WEIGHT INTERACTIONSAND GAS CONTROL about 68 lb/man over the 30-day mission or about SYSTEMS 136 pounds for the overall mission. Reliability is an important criterion in evalu- Unfortunately, another requirement on the ating the weight specific factors in atmospheric designer was to include the possibility of a suited control systems. In certain cases, there is little man in the atmosphere-control loop during emer- information on which to base component failure gency situations. This would impose a AP of rates. However, the use of good engineering 8 inches of water which was assumed in table 29. judgment will tend to give reasonably valid Therefore, it was felt that the tradeoffs should system reliabilities, especially when these results include the suit-in-the-loop values of tables 26 are to be used primarily in a relative rather than and 29. Table 30 compares the values for sub- an absolute comparison. 98 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS TABLE30. -Weight and Power Tradeo#s of the Air-Conditioning System for a 2-Man Orbiting Laboratory [DATA FROM BOEING 121

7 psia 5 psia

3.5 psia 02 3.5 psia 02 3.5 psia 02 3.5 psia 02 5 psia 02 3.5 psia N2 3.5 psia He 1.5 psia N2 1.5 psia He

Power, watts ...... 100 60 72 53 72 Ratio...... 1 0.60 0.72 0.53 0.70 Power penalty, watts ...... 0 -40 - 28 - 47 - 2%

Ventilation power

Power, watts ...... 6; -7L-6l-T Ratio...... 1 1 0.16 0.98 0.30 0.97 Power penalty, watts ...... - 53 -1 -44 -2 -

Power, watts ...... 163 Ratio...... 1 0.43 0.90 0.44 0.80 Power penalty, watts ...... 0 - 93 - 29 - 91 - 30

Weight, Ib ...... Ratio...... 1 Weight penalty, Ib ...... 0 - 114 - 34 -111 - 35

Weight penalty (0.70 Ib/watt)

Weight, Ib ...... 104 49 Ratio...... 1 0.43 0.82 Weight penalty, Ib ...... 0 - 55 - 10 - 54 - 11

Reliability studies should be made with full systems. Advantage should be taken, and re- cognizance of the critical nature of specific mal- flected in the reliability calculations, of the functions and the results of failures of parts and possibilities of redundancy and of replacement ENGINEERING CONSIDERATIONS 99 and repair of components when a human operator no reliability data are available.60 Flyable hard- is present. Unfortunately, reliability factors ware is now under development.110 cannot always be included in the preliminary Two new approaches to flyable oxygen sensing design phase of a program where tradeoffs of devices appear encouraging. A zirconium-oxide gas-specific factors need to be made. solid-electrolyte cell with high temperature The reliability of partial-pressure sensing operation is under development by the Westing- instruments is an issue of great pertinence to the hotise E!ectric Corp.lnS and a thin-film mctal selection of cabin atmospheres. Little reliability oxide process is under study at the Research data of this type are available on the most recent Triangle Institute of North Carolina.119 No sensing systems. reliability data are as yet available. The cabin with 5-psia oxygen has a very reliable Sensing inert gas components is another ap- control system based on a simple sensor for total proach to the problem. Helium, by virtue of its cabin pressure. As oxygen is consumed and unusual physical properties, presents the greatest carbon dioxide is absorbed, the cabin pressure opportunity for flyable instrumentation. Such drops and more oxygen is allowed to enter the physical approaches as thermal conductivity, cabin to offset this pressure drop. Mixed-gas sound resonance, mass and coincidence spec- cabins require partial-pressure sensors for one trometry, and others, offer good potential, but of the two gases in order to maintain a constant no flight hardware has been developed. An percentage of both gases in the face of simul- ionization gage has been developed for analysis taneous oxygen consumption by the crew and of helium-oxygen mixtures in gas dynamics variable, mixed-gas leakage from the cabin. laboratories.70 In spite of the complexity of the Many different ph sensors are available, but circuitry, the modification of such a device for no device with ruggedness and long-term reli- spacecraft use may be a fruitful approach. ability of the simple anaeroid sensor of the 5-psia A thermal conductivity meter has been used oxygen system has been developed.86 A flyable, by Meneely and Kaltreider 71 in physiological ultraviolet-absorption po2 meter is currently under experiments to separate helium from other development for the NASA by the Perkin Elmer respiratory gases and contaminants. An acoustic C0.42 There are still some unresolved problems gas analyzer of the National Instrument Labora- in the area of the interference by water vapor tories has also been used by Faulconer and and carbon dioxide in the ultraviolet band being Ridley 30 in respiratory physiology. sampled. Polarographic sensors all appear to The weight penalty and reliability factors have a limited duration of performance without associated with the additional controls as well adjustments or replacement of the sensor ele- as the sensors in mixed gas systems must also ment~.~~Chromatographic techniques are avail- be accounted for. Several control instruments able but these are costly in terms of weight and for mixed-gas control are available.88.15373 It are not as reliable as might be desired in flight has been estimated that additional weight for a equipment. A flyable chromatograph is under mixed-gas control above that for 5-psia oxygen development by the Beckman Instrument will range from 12 to 15 pounds.46, 79 It has also CO.~,114, 115 Time-of-flight mass spectrometers been estimated that for the Apollo spacecraft, also have the same problems of reliability and substitution of a 7-psia oxygen-nitrogen system flight w~rthiness.~A coincidence mass-spectrom- for the present 5-psia system would increase the eter suitable for flight operations is also under total gas systems weight penalty, including sen- development by the Johnson Laboratories of sors, controls, and tankage, by only 52 pounds Baltimore for the AiResearch Corp. but no reli- or about 10 percent.66 ability data are available as yet.% TRANSIENT PHENOMENA The General Electric Co. is developing fuel- cell sensors which may operate as part of the Several transient phenomena influence gas- hydrogen-oxygen fuel cell of the main power specific tradeoffs. The first has already been supply or be self-contained instruments. Again, covered in the discussion of rapid leakage of gas 100 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS from the cabin in a previous section of this for the 30-day, two-man, orbiting laboratory chapter. Figures 20 and 21 and table 5 compare from 1.25 to 0.7 lb/watt has been discussed. the approximate time available to the crewman There is presently at the Manned Spacecraft prior to the onset of physiological symptoms of Center in Houston, a project on the design of hypoxia after exposure to the several different computer-assisted design and tradeoff t00ls.48 gas mixtures in question. The problem of lock This study should be of great value in system repressurization rates in emergencies was also integration. covered in a previous section of this chapter. ECONOMIC AND OPERATIONAL FACTORS Failure of the environmental control system can produce transient problems. Failure of the In an overall evaluation of atmosphere selec- control system for poz may result in periods of tion, economic and operational factors must progressive hypoxia before repairs can be made. always be considered. These factors are treated In such a case, the time available for action by very briefly here and specifically noted for each the crewman will depend only on the original mixture in chapter 3. poz in the cabin. Under such circumstances, the cabin with the highest pozrpure oxygen at 5 psia, Development Time will allow the longest time. There will be no time The development status of each component difference between the other mixtures since they and subsystem should be determined to provide all have the same poz of 3.5 psia. After any failure an evaluation as to the probability that the total of this type, the amount of time available wiil, system can be developed within the known limita- of course, depend on the volume of cabin per man. tions of time and budget. For example, the design As was discussed previously, transient exer- of some components, such as heat exchangers cise loads in the cabin can condition the selection and ducts, is so well advanced that they may be of cabin and wall temperatures and thus influence assumed to perform as required with little or no gas-specific power penalties. development. On the other hand, development of helium and neon tankage or regenerative chemi- POWER SYSTEM FACTORS cal systems for space application will require The power-weight penalties for several systems extensive development, together with some risk have been discussed. In general, the larger the that it will not be possible to reach the goals at system, the lower the power-weight penalty. the desired time. A closely related problem is the For missions of long duration, solar-cell and adaptability of the system to different mission nuclear-power weight penalties are about 200 profiles, a requirement which is becoming ever to 300 lb/kW. Smaller vehicles and shorter flight more essential. durations require 400 or more lb/kW. Since little or no flight experience has been gained with Uses of Existing Hardware and Equipment larger solar cells or nuclear reactors, these esti- Uses of existing hardware and equipment are mated penalties must be used with much reser- closely related to development time. Whenever vation. There is a recent review on the availability possible, off-the-shelf hardware should be con- and reliability of several different power-genera- sidered in evaluating system weight, reliability, tion systems in spacecraft.1w and costs in gas-specific tradeoffs. The most Fuel-cell penalties are closely dependent on obvious lack of this type of equipment is in small the cryogenic systems. Integration of fuel cell flightworthy helium and neon tankage and mixed and life support systems is a complex factor gas controls. which has received much study.13,579339453*,80,48 Integration of the total power system with life Maintenance and Convertibility support equipment has also been reported.37.118,48 The maintenance and convertibility factor A review of this problem is beyond the scope of relates primarily to the control systems involved. the present study. The role of cryogenic inte- There is a big gap between the simple aneroid gration in possibly reducing the power penalty controls for the 5-psia oxygen system and the ENGINEERING CONSIDERATIONS 101 complex controls of the mixed gas systems, but environment should least influence this require- except for diluent sensors, very little difference ment. The 5-psia oxygen, oxygen-helium, and exists between the individual mixed gas systems. oxygen-neon environments will exert a slightly greater effect. In any case, where the gaseous Crew Acceptance environment will interfere with interpretation of In view of the discussions regarding the physio- results, appropriate simulated ground controls logical factors in parts I, 11, and I11 of this will be required for any of these mixtures. In a seriesw,91,92 and in chapter 1 of this present practical sense, this factor should in no way report, there is little to suggest that any of the discriminate between the gas mixtures in gas mixtures will cause problems as far as crew question. acceptance is concerned. There may be problems Complexity of Design and Operation of aural and pulmonary atelectasis in longer missions with 5-psia oxygen, but they have not Aside from the partial pressure control system appeared as yet in the limited flight experience of all the mixed gases and the design of small, up to 14 days. The more subtle metabolic prob- flyable cryogenic tankage for helium and neon, lems hypothesized for 5 psia oxygen or for the there appears to be little major difference be- lack of nitrogen in the helium or neon mixtures tween the different gas systems in complexity also require further experimentation and flight of design. There are second-order design and testing before evaluation of crew acceptance operational factors which may arise in systems can be made. integration or mission analysis which would require gas-specific orientation. All of these Contaminant Buildup must ultimately be taken into consideration in There is a definite interaction between 5 psia arriving at conclusions relative to the advantages oxygen atmospheres and the toxic contaminant and disadvantages of competing gas systems. problem.g0* Current experimentation by the Interfaces also may be considered as placing Aerojet-General Corp. and the Toxic Hazards restraints or requirements on the system. Many Branch of the 6570th Aerospace Medical Re- of the interfaces have already been covered. search Laboratories of Wright-Patterson Air Typical of the often unconsidered interfaces are Force Base, Ohio, should answer the many the following: 95 questions along these lines. There is no reason (1) Thermal loads to and from other vehicle to suspect that the slightly reduced pN2in the systems, including vehicle structure. oxygen-nitrogen mixtures or presence of helium (2) Power requirements (including quality, type, and neon in the other mixtures will significantly amount, and variation of rate of secondary alter the toxic contaminant hazard.92 systems). (3) Metabolic inputs from occupants, carbon Qualification Testing dioxide, water vapor, and odors in normal and in off-design modes. This factor will extend development programs (4) Vibration and shock loads, including those for the mixed gas systems, especially helium and generated within the system and those received neon beyond those required for 5-psia oxygen. from outside. (5) Noise generated by operation of the system. Environment for Inflight Experiments (6) Control linkages for operation of the atmos- There is always a problem of comparing re- pheric control system itself. sults of inflight physical and physiological experi- (7) Space and relative location requirements ments with the controls performed on the ground within the vehicle. The resolution of this item in sea-level air environments. The need for usually requires the use of mockups and tradeoff simulation chambers for ground controls does studies with other spacecraft systems. add an expense and nuisance factor. From an (8) The ground checkout system. overall point of view, the 7-psia oxygen-nitrogen (9) Onboard display instrumentation. 102 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS (10) Instrumentation providing information to (15) Mechanical support of system components be telemetered. on vehicle load-bearing points. (11) Provision for supplying an atmosphere for use in a backpack to provide atmospheric control Cost for a pressure suit used for extravehicular The cost factor may loom large in a tradeoff operations. analysis, although it often is not as large as in a (12) Provision for the use of an airlock to enable basic decision regarding launch-vehicle function. occupants in pressure suits to leave and reenter The least expensive is no doubt the 5-psia oxygen the space vehicle and resulting repressurization system. Availability of hardware and simplicity needs. of controls dictate this factor. Development costs (13) Prompt detection of malfunctions within of oxygen-nitrogen systems will no doubt be less the system and the transmission of the informa- expensive than corresponding helium and neon tion to the astronauts. systems. Because of the overall weight savings (14) Interaction with operator; manual control which may be possible with the helium or neon, required; extent and scheduling of operator’s the long-range, operational and total program time; special skills required. costs may more than make up the difference. CHAPTER 3 Comparative Analysis of Atmosphere Tradeoffs of the Environmental

Control SvstemJ

IN THE FIRST TWO CHAPTERS, basic data and ana- lock compartment or whole vehicle was used for lytic techniques for performing tradeoff analysis extravehicular operations. To these expendable of atmosphere-related environmental control sub- requirements was added a 10-percent reserve. systems were presented. An attempt was made to Table 31 represented the tabulation of these reduce the data to readily available tabular or estimates. graphic form and to outline the pitfalls of analysis It can be seen that for oxygen, metabolic in each subsystem. The absolute weight penalties utilization in the spacecraft cabin was the single determined by the Boeing Co. for a 30-day two- greatest usage factor, followed by the Gemini man orbiting laboratory with a volume of about leakage and extravehicular trips. For diluent, 1200 ft3 (divided into 750 and 450 ft3 compart- spacecraft cabin leakage was the greatest factor, ments) were presented in detail. The diluent gas followed by extravehicular trips and laboratory storage tradeoffs were presented in table 16, and re pressurization. the overall air-conditioning tradeoffs were pre- The approximate tankage penalty for oxygen sented in table 30. To arrive at a total system may be determined by entering the graph for weight penalty, an analysis of the usage rates and supercritical oxygen (fig. 34(b)) and noting that total weight of oxygen and diluents must be in the range of 315 and 344 pounds of oxygen the determined. To these are added the pertinent tankage penalty is 1.13 pounds of total weight storage weight penalties to arrive at the total per pound of oxygen stored. This is lower than gas-storage subsystem penalty. By adding the the older equivalent penalty figure of about 1.5 total air-conditioning subsystem power-weight pounds per pound of fill seen in figure 3qu). It penalties to the total storage subsystem weight also includes no accessory weights. For the heli- penalties and adding control weight differences, um diluent tankage penalty of table 31, the data one can arrive at the gas-specific penalties for for low-temperature gaseous storage in table 16 the total environmental control system. were used. The values in parentheses were cal- The total expendable gas weight was deter- culated using the supercritical nonvented storage mined by the Boeing CO.,'~who analyzed the penalty of 3.8 lb/lb determined by Mason and normal requirements for metabolic consumption Potter for pressure variant delivery.a For the and leakage as described in chapters 1 and 2. To nitrogen diluent tankage penalty, the values for this was added the estimated oxygen consumed in supercritical vented storage in table 16 were extravehicular trips, and consumed in the Gemini used. The values in parentheses are the tank trips including leakage. An additional weight of penalties calculated from the advanced state- gas was needed to cover gas lost in the carbon of-the-art value of supercritical nitrogen from the dioxide removal process and in the catalytic data of Rousseau et a1.95 in figure 39(a) where burner. It was also required to account for the the penalties for 86 and 32 pounds of fill are 1.18 gas lost in cabin repressurizations when the air- and 1.20, respectively. The two different sets of

103 104 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS , 31. -Expendable Fluid Requirements and Total Environmental Control System Weight Penalties TABLE I for a 2-Man Orbiting-Laboratory Mission [CALCULATED FROM THE DATA OF BOEING,'* ROUSSEAU ET AL.,95 AND MASON AND POTTER "3 . [ALL VALUES IN POUNDS]

7 psia 5 psia Function 5.0-psia 02 3.5-psia 02 3.5-psia 02 3.5-psia 02 3.5-psia O2 3.5-psia N2 3.5-psia He 1.5-psia N, 1.5-psia He

Oxygen use: Extravehicular trips ..... 32.4 32.4 32.4 46.3 Gemini trips ...... 2.4 2.4 ...... Laboratory repressurization...... 18.0 18.0 18.0 18.0 25.7 Metabolic...... 120.0 120.0 120.0 120.0 120.0 Leakage...... 32.1 29.0 22.9 21.6 31.5 CO, removal...... 13.6 13.6 13.6 13.6 19.5 ...... 5.0 5.0 5.0 5.0 5.0 ...... 75.0 75.0 75.0 75.0 75.0 29.5 28.7 28.6 32.3

Total...... 328.4 324.9 315.6 314.2 355.3

Diluent use: Extravehicular trips ......

......

Tankage:

Diluent ......

Gas storage: Total 02 system ...... Total diluent system ......

Fan power:

Ventilation......

Tot d......

Controls......

otal ECS penalty ...... ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM 105 diluent tankage weights and total weights seen in gaseous storage cooled in liquid hydrogen with all table 31 represent a difference in approach often the accessories added to the weight penalty was found in tradeoffs. The tankage penalty not in 36 pounds heavier at 495 pounds. The maximum parentheses, all of which were taken from table difference of 265 pounds is between 7-psia 16, include, along with the dry tank weight, the oxygen-nitrogen mixture and 5-psia oxygen- weight of vent fluid, unavailable fluid, fill toler- helium mixture. The lightest system was 76 ance fluid, valves, heaters, and heater power pounds lighter than the 5-psia oxygen system weight. The values in parentheses include only currently being employed. dry tank weight with none of the accessories and What is the projected weight penalty for an controls. The latter are most often used in trade- oxygen-neon system? Liquid neon with an ac- off studies. Differences in total storage weight cessory-free tankage penalty of 1.85 pounds per and in total weight penalty reflect these differ- pound of useful fill as opposed to 3.85 for liquid ences in basic assumptions regarding gas storage. helium, (both supercritical and pressure variant At the bottom of table 31, the differences in mode) should reduce the diluent tankage below total gas-specific weights for the environmental that of helium by a factor of only about 12.4 - 1.85 control system are tabulated with both 7-psia x4.48=3.9 pounds. This is calculated by as- oxygen-nitrogen mixtures and 5-psia oxygen as suming the mass leak rates are similar (table 10). baselines. The lightest weight penalty for the The dehumidifying system is assumed to be environmental control system was 459 pounds of constant water removing capacity with respect for the 5-psia oxygen-helium mixture. This was to power penalty. The power ratios of the dehu- determined for optimum supercritical helium midifying fans for oxygen-neonloxygen-helium operating in the nonvented, pressure variant can be estimated by the ratios in equation (67). mode with an accessory-free penalty of 3.8 pounds total weight per pound of useful fill. The total HPOz-Ne - wcJOz-Ne POz-He APOz-Ne x-x- (68) penalty for an environmental control system using HPOz-He wg02-He POz-Ne UOI-He From equation (66),

Since

and

therefore,

The ventilating system is a constant Q and AT to table 23 the ventilating power ratios of the system with respect to power penalty. According 02-Ne/02-He should be:

261-559 0-67-0 106 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS of the total fan-power weights are similar, with the only exception being the oxygen-helium mixtures. In table 32, the 7 psia has a higher weight than the 5 psia and in table 31, a slightly Since the heat capacity of helium and neon are lower weight. The total environmental control the same, the C, factor drops out of the ventila- system weights are quite similar; those of table tion ratio. The relative power for ventilation 32 run slightly higher. The weight differences becomes: between the gases are quite similar, with the oxygen-helium mixture at 5 psia having the most favorable and oxygen-nitrogen mixture at 7 psia, the least favorable penalty. A rough estimate of the total weight of a 5-psia Table 33 is far less complete than the others. oxygen-neon system may be obtained from table Leakage was very much lighter, and repressur- 31: ization gas usage was much heavier than in table 31. This may be a result of differences in inter- 355 + 1.85 (4.48) + 1.1 (53) pretation and optimization of mission tasks. + 0.83 (19)+ 15= 452 pounds. Unfortunately, no tankage penalties were recorded. There is, therefore, much weight un- This estimate suggests that the total weight for a accounted for when the total vehicle penalties 5-psia oxygen-neon system of 452 pounds is are balanced against recorded data. Not all of slightly less than that for an oxygen-helium sys- this could possibly be tankage. The power weights tem (459 lb) under the given design assumptions. were calculated by assuming a penalty of 2.4 It must be emphasized again that the estimated lb/W. This is twice as high as that used in table penalties of table 31 and the extrapolation to 31 and, probably, in table 32 as well. The total neon hold only for the specific 30-day, two-man weight penalties for the environmental control orbiting laboratory under study. subsystem are generally higher than those Now these weight penalites are compared with recorded in either tables 31 or 32. This may be those calculated independently by two other due to the higher power weight penalty and groups: the Lockheed Aircraft Corp.G1 and the possibly a higher tankage penalty. Aerospace Corp.46 Unfortunately, no textual Table 33 is optimized for a circulation velocity material was available to explain the final tabular of 50 ft/min. Figure 88 represents the pressure tradeoffs which are presented in tables 32 and sensitivity of the total weight penalty for the 33. Both of these tables were for a 30-day, two- vehicle, including tankage and power at the man orbiting laboratory of the same general size optimum velocity for each gas. The alveolar pol and mission as that of table 31. It is clear, how- is constant at 102 mm Hg in the mixed gas ever, that different assumptions were made re- systems and increases with pressure in pure garding several critical parameters. oxygen. A comparison of tables 31 and 32 shows that Once again, as in tables 31 and 32, the 5-psia there was a difference in major oxygen-leakage oxygen-helium system had the lowest total parameters assumed for the mission. There is a environmental control subsystem penalty, and difference in diluent leakage and the tankage 7-psia oxygen-nitrogen system had the highest. penalties for both oxygen and diluents. Fortui- The 7-psia oxygen-nitrogen system was slightly tously, the total gas-storage penalties are in the higher than the 5-psia oxygen-nitrogen system. same general range, and the ratios between the Although the absolute values are not equal, all different gas mixtures are quite similar. three tables appear to show similar weight ratios. In table 32, there was no partition between the The 5-psia oxygen system seems to have an inter- dehumidification and ventilation functions. Also, mediate penalty between 7-psia oxygen-helium fan weights of 11 pounds are included. Total fan system and 5-psia oxygen-nitrogen system. The power weights are higher in table 31. The ratios significance of the similarity of results in all three ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM 107 TABLE32. -Expendable Fluid Requirements and Total Environmental Control System Weight Penalties for a 30-Day, 2-Man Orbiting-Laboratory Mission [AFTER LOCKHEED e”]

FALL v. UES IN POUNDS]

Function 5.0-psia 02

Oxygen use: Metabolic ...... 120 120 120 Leakage...... 45 55 41 49 60 Molecular sieve loss ...... 10 10 10 10 15 Repressurization of laboratory...... 84 84 84 84 120 Repressurization of Gemini B ...... 12 12 12 12 12 Reserve, 10 percent ...... 27 28 27 28 33

Total fluid stored ...... 298 I 309 1 294 I 303 360

Diluent use: Leakage ...... Molecular sieve loss...... Repressurization of laborato ry...... Reserve, 10 percent...... Total fluid stored...... 4136 Tankage: Oxygen tank penalty ...... 89 93 88 91 108 Diluent tank penalty...... 74 87 31 36

Total tankage ...... 163 180 119 127 108

Gas storage: Total 02system ...... 387 402 382 394 468 Total diluent system...... 210 108 87 45 ......

Total ...... 597 510 469 439 468

Fan system: Fan weight ...... 11 Fan power ...... 180

Total ...... 191

Total ECS penalty ...... r5 663153 660191 659 AW ...... - 191 1 -188 1 -2981: - 195 AW ...... + 195 +4 +1 -103 0 studies will be more meaningful when the inter- system penalties was demonstrated. It was made mediary calculations are made available. clear that as the duration of mission increases, the size and weight efficiency of many systems EFFECT OF MISSION LENGTH ON OVERALL increase. Long- standby times do penalize lisuid CONTROL SUBSYSTEM storage. It is of interest now to compare the TRADEOFFS effects of mission duration on the total vehicle At many points during the discussion of weight imposed by cabin factors such as struc- chapter 2, the effect of mission duration on sub- tural weight, oxygen storage, diluent storage, 108 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS TABLE33. -Expendable Fluid Requirements and Total Environmental Control System Weight Penalties for a 30-Day, 2-Man Orbiting-Laboratory Mission [AFTER JOHNSON 461 [ALL VALUES IN POUNDS]

3 Function 5.0-psia O2

Oxygen use: Metabolic...... 120 42.0 87.2

249.2

Diluent use: Leakage......

Total ......

Gas storage: Tankage penalty ...... Additional gas controls......

Total......

Power (optimized at T= 50 ftlmin): Heat transfer...... 169.8 Circulation...... 169.8 I I I I Total...... 290.6 173.2 302.9 222.1 339.6

(Unaccounted weight)...... 325 313 231 223 220

Total vehicle penalty:...... 926 760 793 694 809 AW ...... 0 - 166 - 133 - 232 - 117 AW ...... + 117 - 49 - 16 - 115 0

and fan and expulsion power. Figure 89 repre- higher than those used for the corresponding sents these interactions for cabins of pressures 30-day vehicle of table 31. varying from 5 to 7 psia.12 Basic vehicle weights It can be seen that even for the 60-day mission, of loo00 and 13000 lbs were used for mission the maximum total environmental control sub- durations of 30 and 60 days, respectively. An system penalty is only 17000 pounds, or about oxygen-helium mixture was used, with the helium 13 percent of the total vehicle weight. The maxi- being stored in gaseous form within the hydrogen mum weight differences between designs are tank (table 16). The weight effects of the in- in the order of only 3 percent of the vehicle weight. creased hydrogen tank size are accounted for In the longer missions there are some savings in the power penalties for atmosphere circulation. in environmental control subsystem weight from Because the study was performed at an earlier the more efficient storage of cryogenic fluids. stage of design, the penalties are somewhat The increase in structural weight conditioned ~~

ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM 109 900

I 800

3’ \ 3 5I psi0 i I -$ 700 E CL c L .-0) 0 - 600 I ._0 02system I I -W 0 + - Q-He system 500

0 2 00 250 300 350 400 450 500 550 Total atmospheric pressure, P,, rnm Hg

FIGURE88.-Total vehicle weight penalty for differentgas systems with gas velocity optimized for each System. (AFTER JOHNSON.‘B) by the higher pressure is minimal. The weight At all levels of interest, there are mission-specific penalty of the diluent is almost equal to that of variables which strongly condition the optimi- the fan and cryogenic expulsion power and is zation process. only slightly more sensitive to total pressure In tables 34, 35, and 36 an attempt has been changes. Increased leakage at higher pressure made to summarize the comparative values of probably accounts for the increase in diluent each system. References to figures and tables penalty. in this study and to other parts of this series are It thus appears that the relative effect of dif- included to aid the reader in reviewing the ferent atmospheres on total vehicle weight is nature of the interaction involved. In the columns minimal. For a 30-day, two-man orbital mission, headed “Selection,” an attempt has been made several hundred pounds are at stake. These few to place the five gas mixtures in a descending hundred pounds, however, may be quite signifi- order of desirability. Those mixtures of equal cant in appreciation of total mission success. desirability are placed in parentheses. The remarks covering oxygen-helium systems also apply to oxygen-neon mixtures. The only SUMMARY OF TRADEOFFS IN THE SELECTION major differences are the theoretical superiority OF SPACE-CABIN ATMOSPHERES of neon over helium in reducing bends and In the several parts of this series, an attempt neurocirculatory collapse and increasing survival has been made to develop scientific criteria for time after ebullism. Dehumidification power the selection of space cabin atmospheres. The weight is slightly greater for oxygen-neon mix- many interacting variables preclude a bold state- tures, while ventilation power weight is slightly ment regarding a single optimum atmosphere. less than that for oxygen-helium mixtures. 110 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

h a 7 Y c4 N a 4 ? v N 0 v) r: .- .$ n v) h !! '6 nm 3 t ? a a .-- Y, n h O .4 V Y v) ...g 4

N v) 3 0 3 ? a m In m Y h 0, PI X v P -v)

uII) e a c m c n e 3 0 4 E c m !! e -0 c U -.-a 0 2 D 0" v)

N I 0

m ~~

ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM 111 112 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS ~

ATMOSPHERE TRADEOFFS OF THE ENVIRONMENTAL CONTROL SYSTEM 113

I, -

iii 6 .-s k ,.v)

5 v) .- 114 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS Tankage penalties for liquid neon are about weights to give the oxygen-neon mixture a mini- half those for liquid helium. For the 30-day, two- mal advantage over the oxygen-helium mixture in , man orbiting laboratory, the increased power total environmental control system weight weights almost match the decreased storage penalty. References

1. American Society of Heating, Refrigerating and Air Conditioning Engineers. ASHRAE Guide and Data Book. Fundamentals and Equipment for 1965 and 1966. ASHRAE, New York, 1%5. 2. BACK,K. C.: Toxic Hazards Branch, 6570th Aerospace Medical Research Laboratories, Wright- Patterson AFB, Ohio. Biochemical effects of chronic exposure of animals at 100 percent oxygen environments. Unpublished data presented at the NASA Workshop Conference on Criteria for Selection and Evaluation of Space Mission Atmospheres, 1965. 3. BACK,K. C.; THOMAS,A. A,; and PINKERTON,M. K.: Toxicology, in Bioastronautics Data Book, P. Webb. (ed.), NASA-SP-3006, 1964, pp. 17-29. 4. BEARD,S. E.; ALLEN,T. H.; MCIVER,R. G.; et al.: Comparison of Helium and Nitrogen in Pro- duction of Bends in Simulated Orbital Flights, in Preprints of Scientific Program, 37th Annual Scientific Meeting, Aerospace Medical Association, Apr. 1966 (Las Vegas, Nev.), 1966, pp. 43-44. 5. Beckman Instruments, Inc.: Proposal for Gas Analyzer. Rept. CS-62-227, Fullerton, Calif., 1962. 6. BENZINGER,T. H.: The diminution of thermoregulatory sweating during cold-reception at the skin. Proc. Nat. Acad. Sci., 47: 1683-1688, 1%1. 7. BERENSON,P. J.: General Analysis of Human Thermal Comfort. Rept. No. SS-3245, AiResearch Mfg. Co., Los Angeles, Calif., 1965. 8. BERENSON,P. J.: Prediction of Human Thermal Comfort in Oxygen-Nitrogen Atmospheres, in Physiological and Performance Determinants in Manned Space Systems, Vol. 5., P. Horowitz (ed.), American Astronautical SOC., Baltimore, 1%5, pp. 1-29. 9. BERRY,C. A,; COONS, D. 0.;CATTERSON, A. D.; et al.: Man's Response to LongDuration Flight in the Gemini Spacecraft, Gemini Midprogram Conference Including Experiment Results. NASA-SP-121. 1966, pp. 235-261. 10. BINDER,H. J.; HERTING,D. C.; HURST, V.; et al.: Tocopherol deficiency in man. New Eng. J. Med., 273: 1288-1297, l%5. 11. BLOCKLEY,W. V.: Temperature, in Bioastronautics Data Book, P. Webb (ed.). NASA-SP-3006, 1964, pp. 103-131. 12. Boeing Company, ECS Flight Technology Group: Engineering Tradeoffs of Different Gas Systems Pertaining to the Selection of Space Cabin Atmospheres. Seattle, Wash., 1965 (unpublished data). 13. Boeing Company: Manned Environmental System Assessment. NASA-CR-134, 1964. 14. BOLLES,T. V.; SPEZIALI,V. A.; and THOMSON,G. W.: Microcontractor Utilizing High-Density Metallic Superoxide. NASA-CR-57021, General Dynamics Co., Electric Boat Div., Groton, Conn. (Rept. No. U41364-164), 1964. 15. BOSTICK, C. R.: AiResearch Mfg. Co., Department 93-3, Los Angeles, Calif., personal communication, 1965. 16. BRINK.N. 0.. Research on an Expandable Airlock Utilizing the Elastic Recovery Principle. NASA-CR-351, 1965. 17. BURTON,D. R.: Performance of water conditioned suits. Aerospace Med., 37: 500-504, 1966. 18. CHRISTIAN,J. L.; and HURLICH,A.: Physical and Mechanical Properties of Pressure Vessel Materials for Application in a Cryogenic Environment -Part 11. ASD-TDR42-258, Pt. 2, General Dynamics/Astronautics, San Diego, Calif., Apr. 1963. 19. CLIFFORD,J.; KERSLAKE,D. McK.; and WADDELL,J. L.: The effect of wind speed on maximum evaporative capacity in man. J. Physiol., 147: 253-259,1959. 20. COE, C. S.; ROUSSEAU,J.; and SHAFFER,A.: Analytical Methods for Space Vehicle Atmospheric Control Processes. ASD-TR-61-162, Pt. 11, AiResearch Mfg. Co., Los Angeles, Calif., Nov. 1962. 21. COOK, G. A. (ed.): Argon, Helium, and the Rare Gases: The Elements of the Helium Group. Vol. I: History, Occurrence, and Properties. Vol. 11: Production, Analytical Determination, and Uses. Interscience Publishers, 1961. 115 116 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 22. DOBRYNINA,T. A.: Lithium Peroxide, Moscow, 1964. 23. DOHNER,C. V., Supervising Engineer, HTT-Thermodynamics Technology Component, Missile and Space Division, General Electric Co., Philadelphia, Pa., personal communication, 1966. 24. DRYDEN,C. E.; HAN,L. S.; HITCHCOCK,F. A.; et al.: Artificial Cabin Atmosphere Systems for High Altitude Aircraft. WADC-TR-55-353, Ohio State Univ. Research Foundation, Columbus, Ohio, Nov. 1956. 25. DUBOIS, A. B.; TURAIDS,T.; MAMMEN,R. E.; et al.: Pulmonary atelectasis in subjects breathing oxygen at sea level or at simulated altitude. J. Appl. Physiol., 21: 828-836, 1%. 26. DUCROS,H.; and BERANGER,S.: Use of Lithium Peroxide as an Agent for Purification of Tainted Air. FTD-TT-65-1792, 1 + 2 + 4, Foreign ,Technology Division, Wright-Patterson AFB, Ohio, Jan. 1966. 27. DUNBAR,W.: Corona Onset Voltage of Insulated and Bare Electrodes in Rarefied and Other Gases. AF-APL-TR-65-122, Boeing Co., Seattle, Wash., 1966. 28. ECKERT,E. R. G.; and DRAKE,R. M., JR.: Heat and Mass Transfer, McGraw-Hill Book Co., Inc., 1959. 29. EPPERSON,W. L.; QUIGLEY,D. G.; ROBERTSON,W. G.; et al.: Observations on Man in an Oxygen- Helium Environment at 380 mm Hg Total Pressure: 111. Heat Exchange. SAM-TR66-233, USAF School of Aerospace Medicine, Brooks AFB, Tex., May 1966. 30. FAULCONER,A.; and RIDLEY,R. W.: Continuous quantitative analysis of mixtures of oxygen and ether with and without nitrogen. I. Acoustic gas analyzer for mixtures of first three gases. Anesthes., 11: 265, 1950. 31. FISCHER,C.: Manned Space Center, National Aeronautics and Space Administration, Houston, Tex., personal communication, 1966. 32. GREEN,F. H.: Psychrometric Data. ARMC-66-537, AiResearch Mfg. Co., a division of the Gar- rett Corp., Los Angeles, Calif., May 15, 1966. 33. HALL, G. M.; and TREIRAT,E.: Study of Integrated Cryogenic Fueled Power Generating and Environmental Control Systems. Vol. 111: Power Generating Equipment Study. ASWTR- 61-327, Vol. 3, Kidde (Walter) & Co., Belleville, N.J., Nov. 1961. 34. HALL,J.: Copper Manikin Regional Loss and Cooling Constants. Memo Rept. MCREXD-696- 105P, Wright-Patterson AFB, Oct. 1950. 35. HALL,J. F.; BUEHRING,WILLI J.; and STROBL,W. W.: Effects of Various Gases on Handgear Insulation. AMRL-TRd5-4, Aerospace Medical Research Labs., Wright-Patterson AFB, Ohio, Dec. 1965. 36. Hamilton Standard Div. of United Aircraft Corp.: Manned Orbiting Space Station. Vol. 4: En- vironmental Control and Life Support System Study, Final Report, NASA-CR-65088, May 1964. 37. HANSON,K. L.: Thermal Integration of Electrical Power and Life Support Systems for Manned Space Stations. NASA-CR-316, Nov. 1965. 38. HARDY,J. D. (ed.): Temperature. Its Measurement and Control in Science and Industry. Pt. 3: Biology and Medicine, Reinhold Pub. Corp., 1963. 39. HONMA,H.; and CROSBY,H. J. (eds.): A Symposium on Toxicity in the Closed Ecological System. Research Laboratories, Lockheed Missiles & Space Co., Palo Alto, Calif., 1%3. 40. HUGGETT,C.; VON ELBE, G.; and HAGGERTY,W.: The Combustibility of Materials in Oxygen- Helium and Oxygen-Nitrogen Atmospheres. SAM-TR46-85, USAF School of Aerospace Medicine, Brooks AFB, Tex., 1966. 41. HUGGETT,C.; VON ELBE,G.; HAGGERTY,W.; et al.: The Effects of 100 percent Oxygen at Re- duced Pressure of the Ignitibility and Combustibility of Materials. SAM-TR-65-78, USAF School of Aerospace Medicine, Brooks AFB, Tex., 1965. 42. HULL,W.: Crew Systems Division, Manned Spacecraft Center, National Aeronautics and Space Administration, Houston, Tex., personal communication, 1965. 43. JACOBSON,S. L.: Engineering of the Sealed Cabin Atmosphere Control System. Presented at the 30th Annual Meeting of the Aeromedical Association, Apr. 27-29, 1959. 44. JENKINS, A. C; and COOK,G. A.: Gas Phase Properties, in Argon, Helium, and the Rare Gases: The Elements of the Helium Group. Vol. I: History, Occurrence, and Properties. G. A. Cook (ed.), Interscience Publishers, 1961, pp. 173-243. 45. JENNINGS,A. J.; and HALL,G. M.: Study of Integrated Cryogenic Fueled Power Generating and Environmental Control Systems. Vol. V: Integration and Control Studies. ASD-TR-61-327, Vol. 5, Kidde (Walter) & Co., Belleville, N.J., Nov. 1%1. 46. JOHNSON,A. L.: Aerospace Corp., Manager, Life Support Section, Los Angeles, Calif., unpublished data. To be published in the Diluent Selection Study for the MOL Program, 1966. REFERENCES 117 47. JOHNSON.L. F., JR.: NEVILLE.J. R.; BANCROFT.R. W.: et al.: Physical Transducers for Sensing Oxygen. Review 8-63. USAF School of Aerospace Medicine. Brooks AFB. Tex., Aug. 1963. 48. JOHNSTON,R. S.: The Douglas G-189 Program for Calculating the Performance for Environmental Control in Life Support Systems. Chief. Crew Systems Division. Manned Spacecraft Center, NASA, Houston, Tex., personal communication. 1966. 49. KEATING,D. A.: Application of Gas Expansion to Fluid Circulation Devices in Manned Space Assemblies. AMRL-TR-65-26, Air Force Systems Command, Aerospace Medical Division, Wiigh-Pa::erson AFB, Ohio, Ap:. 1965. 50. KEATINL D. A.: Design Study of High Prrssurr. Oxygen Vessels. WADC-TR-50-767, Wright Air Development Center. Aerospace Medical Division, Wright-Patterson AFB, Ohio. Feb. 1960. 51. KEATING,D. A.; WEISWURM,K.; MEYER,C. M.: et al.: Manned Testing of a Semipassive Po- tassium Superoxide Atmosphere Control System. AMRL-TR-65-194. Aerospace Medical Division, Wright-Patterson AFB, Ohio. Nov. 1965. 52. KEATING. D. A.: and WEISWURM.K.: Potassium Superoxide Passive Air Regeneration Studies for Manned Sealed Environments. WADD-TRe707. Aerospace Medical Division. Wright- Patterson AFB, Ohio. 1960. 53. KERSLAKE, D. McK.: Errors Arising From the Use of Mean Heat Exchange Coefficients in the Calculation of the Heat Exchanges of a Cylindrical Body in a Transverse Wind, in Temperature. Pt. 3: Biology and Medicine, J. E. Hardy (ed.), Reinhold Pub. Corp., 1%3, pp. 183-190. 54. KIMZEY,J. H.; DOWNS,W. R.; ELDRED,C. H.; et al.: Flammability in Zero-Gravity Environment. NASA-TR-K-246.1066. 55. KING, B. G.: High concentration-short time exposures and toxicity. J. Indust. Hyg. & Toxicol., 31: 365-375. 1949. 56. KINSLOW,M.; and MAJOR, B. M.: Systems of Units and Conversion Tables. AEDC-TDR-624. von Karman Gas Dynamics Facility, ARO Inc.. Arnold Air Force Station, Tenn., Feb. 1962. 57. KNIGHTS.A.: JENNINGS.A.: and FORTE.M.: Study of Integrated Cryogenic Fueled Power Gen- erating and Environmental Control Systems. Vol. IV: Environmental Control and Attitude Con- trol Studies. ASDTR-61-327. vol. 4, Kidde (Walter) & Co.. Belleville. N.J., Nov. 1%1. 58. KRANTZ, P.: Calculating human comfort. ASHHAE J., 1964. pp. 68-77. 59. KREITH,F.: Principles of Heat Transfer. International Textbook Co., 1%2. 60. LAWTON,R. W.: General Electric Co., P.O. Box 8555, Philadelphia, Pa.. personal communication, 1965. 61. Lockheed Missiles 81 Space Co., Bioastronautics Div.. Sunnyvale, Calif., personal communication, 1%5. 62. LUFT, U. C.: Data on Oxygen Pressure Effects Compiled for the Garrett Corp. The Lovelace Foun- dation, Albuquerque, N. Mex., 1962. 63. MCADAMS,W. H.: Heat Transmission. McGraw-Hill Book Co., Inc., 1954. 64. MCGOFF, M. J.: Potassium Superoxide Atmosphere Control Unit. AMRL-TR-65-44. MSA Re- search Corp., Callery, Pa., 1965. 65. MARKOWITZ,M. M.: and DEZMELYK,E. W.: A Study of the Application of Lithium Chemicals to Air Regeneration Techniques in Manned, Sealed Environments. AMRL-TDR-64-1, Foote Mineral Co., Exton, Pa., Feb. 1964. 66. MASON,J.; and POTTER,J.: AiResearch Mfg. Co., Los Angeles, Calif., personal communication of unpublished data, Mar. 1966. 67. MASON,J. L.; WAGGONER,J. N.; and RUDER,J.: The Two-Gas Spacecraft Cabin Atmosphere Engineering Considerations. Presented at the International Astronautics Federation Meeting (Athens, Greece), Sept. 13-19, 1965. 68. MATSCH,L. C.: Associate Technical Director, Linde Division, Union Carbide Corp., Tonawanda, N.Y.. personal communication, Mar. 1966. 69. MAUSTELLAR.J. W.: MIXOFF, M. J.: KEATING.D. A.: and WEISWURM,K.: Superoxide Atmosphere Control System for Manned Space Assemblies. Presented at the 36th Annual Meeting of the Aerospace Medical Association (New York), Apr. 1965. 70. MELFI, L. T., JR.; and WOOD, G. M., JR.: The Use of an Ionization Gage as a Quantitative Ana- lyzer for Bi-Gaseous Mixtures. NASA-TN-D-1597, Dec. 1%2. 71. MENEELY,G. R.; and KALTREIDER,N. L.: Volume of lung determined by helium dilution. J. Clin. Invest., 28: 129, 1949. 118 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS 72. MENGEL.C. E.: KANN.H. E.: HEYMAN,A.: et al.: Effects of in vivo hyperoxia on erythrocytes. 11: Hemolysis in a human after exposure tu oxygen under high pressure. Blood, 25: 822-829, 1965. 73. MICKELSON,W. F.; and HYPES, W. D.: Oxygen-Nitrogen Indicator-Controller and Space Cabin Outboard Leak Simulator. AMRL-TDR-64-25, Aerospace Medical Division, Wright-Patterson AFB, Ohio, June 1964. 74. MILLER,R. R.: and PIATT. V. R.: The Present Status of Chemical Research in Atmosphere Purification and Control on Nuclear-powered Submarines. NRL-5465, US. Naval Research Lab., Washington, D.C., Apr. 1960. 75. National Aeronautics and Space Administration, Manned Spacecraft Center, Houston, Tex.: Mercury Project Summary Including Results of the Fourth Manned Orbital Flight, May 15 and 16,1%3. NASA-SP-45, May 1%3. 76. NELSON,N.; EICHNA,L. W.; HORVATH,S. M.; et al.: Thermal exchanges of man at high temperatures. Amer. J. Physiol., 151: 626452,1947. 77. NEVISON,T. O., JR.: Letter Report to the Garrett Corp. The Lovelace Foundation, Albuquerque, N. Mex., Jan. 1%2. 78. PARKER,F. A.; EKBERG,D. A.; WITHEY, D. J.; et al.: Atmosphere Selection and Control for Manned Space Stations. Presented at the International Symposium for Manned Space Sta- tions (Munich, Germany), 1%5. 79. PARKER,F. A.; EKBERG,D. R.; and WITHEY, D. J.: Atmosphere Selection and Environmental Control for Manned Space Stations. General Electric Co., Philadelphia, Pa., July 23, 1964. 80. PASELK,R. A.: Integration and Optimization of Space Vehicle Environmental Control Systems. ASD-TR41-175, pt. 11, North American Aviation, Inc., Los Angeles, Calif., Apr. 1963. 81. PETROCELLI,A. W.: Progress in the Development of Active Chemicals for Use as Air Revitaliza- tion Materials. Presented at the International Astronautics Federation Meeting (Athens, Greece), Sept. 1965. 82. PETROCELLI,A. W.; and CAPOTOSTO,A., JR.: The Synthesis and Utilization of Low-Molecular Weight Ozonides for Air Revitalization Purposes. NASA-CR-135, Nov. 1964. 83. PETROCELLI,A. W.; and CHIARENZELLI,R. V.: The inorganic ozonides. J. Chem. Ed., 39: 557- 560,1962. 84. PRESTI,J. B.: Semi-passive Potassium Superoxide Air Revitalization System, Annual Report. Rept. No. U-413-64-202, General Dynamics Corp., Electric Boat Div., Groton, Conn., Dec. 1964. 85. RADNOFSKY,M. I.: Crew Systems Division, National Aeronautics and Space Administration, Houston, Tex., personal communication, Feb. 1966. 86. REED, A.: Evaluation of Gas Composition Detection Methods in Manned Space Vehicles. Rept. ADR-04-06-62.1, Grumman Aircraft Engineering Corp., Bethpage, N.Y., Dec. 1962. 87. ROBERTS, D. K.: Brief Outline Summarizing Electron Microscopic Studies on “Oxygen Tox- icity.” Unpublished data: NASA Workshop Conference on Criteria for Selection and Evalua- tion of Space Mission Atmospheres, Nov. 1965. 88. ROSS, M. D.: Space/Defense Corp., Birmingham, Mich., personal communication, Feb. 1965. 89. ROTH, E. M.: Bioenergetics of Space Suits for Lunar Exploration. NASA-SP-84, 1966. 90. ROTH, E. M.: Space-Cabin Atmospheres. Part I: Oxygen Toxicity. NASA-SP-47, 1964. 91. ROTH. E. M.: Space-Cabin Atmospheres. Part 11: Fire and Blast Hazards. NASA-SP-48, 1%4. 92. ROTH, E. M.: Space-Cabin Atmospheres. Part 111: Physiological Factors of Inert Gases. NASA- SP-117, 1967. 93. ROTH, E. M.: Supplementary Bibliography on Fire and Blast. A: Combustion Studies. B: Sec- ondary Effects. Contract NASr-115, Lovelace Foundation for Medical Education and Research, Albuquerque, N. Mex., Feb. 1966. 94. ROTH, E. M.; and BILLINGS,C. E., JR.: Atmosphere, in Bioastronautics Data Book, P. Webb, (ed.), NASA-SP-3006, 1964, p. 5. 95. ROUSSEAU,J.; BURRISS,W. L.; COE, C. S.; et al.: Atmospheric Control Systems forspace Vehicles. ASD-TDR-62-527, Part I, AiResearch Mfg. Co., Los Angeles, Calif., Mar. 1963. 96. ROUSSEAU,J.: OLSON,R. L.: COE. C. S.: et al.: Atmospheric Control Systems for Space Vehicles. ASD-TDR-62-527, Pt. 11, AiResearch Mfg. Co., Los Angeles, Calif., Feb. 1964. 97. ROWLETT.B. H.: and LEE, R. H.: Environmental control. Spacc/Acronautics, 42: 106-109. Sept. 1964. 98. R~ss,E. J.: Atmospheric Stores System Evaluation for Space Flights of One Year Duration. Report 64-26213, General Dynamics/Astronautics, San Diego, Calif., Oct. 1%3, rev. Sept. 1964. REFERENCES 119 99. SHAEFER,K. E.: A concept of triple tolerance limits based on rhronic carbon dioxide toxicity studies. Aerospace Med., 32: 197-204,1961. 100. SCHREIHANS,F. A.; and DRYSOL,D. E.: Flammability Characteristics of Some Organic Space- craft Materials in Zero Gravity. NAA-SIWS40, North American Aviation, Inc., Downey, Calif., May 1965. 101. %CORD, T. C.: Chief, Life and Environmental Systems Branch, Douglas Aircraft Corp., Santa Monica, Calif., personal communication, Feb. 1%. 102. SECORD,T. C.; and BONUM, M. S.: Life Support Systems Data from 62 Days of Testing in a Manned Space Laboratory Simulator. Douglas Paper No. 3397, Douglas Missile & Space Sys- tems Division, Santa Monica, Calif. (Presented to AIAA Fourth Manned Space Flight Meeting, St. Louis, Mo., Oct. 11-13,1965.) 103. SNOW,R. H.: Thermodynamic Evaluation of the Possibility of Lithium Superoxide Production. WADD-AMRL-TR-65-126, Illinois Inst. of Technology, Chicago, 1%5. 104. SPIETH,C. W.; BELL,J. E.; HUNTER,B. J.; et al.: Power Generating and Environmental Control Systems. Vol. 11: Cryogenic Tankage Investigation. ASD-TR-61-327, Vol. 2, Beechcraft Re- search & Development, Inc., Boulder, Colo., Nov. 1961. 105. STERNBERGH,S. A.; and HICKAM,W. M.: Flight Type Oxygen Partial Pressure Sensor. NASA- CR-534, Aug. 1966. 106. TONELLI,A. D.; and SECORD,T. C.: Auxiliary Power Generating System for a Large Space Lab- oratory. DAC-P-1993, Douglas Missile & Space Systems Division, Douglas Aircraft Co., Inc., Huntington Beach, Calif., Sept. 1964. 107. TROUT,0. F., JR.: Sealing manned spackcraft. Astronautics and Aerosp. Engr., 1: 444,1964. 108. VANCE, R. W. (ed.): Cryogenic Technology. John Wiley & Sons, 1963. 109. VORONIN, G. I.; GENIN,A. M.; and FOMIN,A. G.: Physiological and Hygienic Evaluation of the Life Support Systems Used on Vostok and Voskhod Spacecraft. (Translation from paper presented at the 2d International Symposium on Basic Environmental Problems of Man in Space, Paris, June 14-18, 1965.) NASA-TT-F-9424, July 1965. 110. WARNER,H.: Oxygen partial pressure sensor. MP~.Electron. Biol. Engng., I: 79-84, 1963. 111. WEBB. P.: Human Water Exchange in Space Suits and Capsules. Study in preparation under NASA Contract NASr-115, Lovelace Foundation for Medical Education and Research, Albu- querque, New Mexico, 1966. (To be published as a NASA CR.) 112. WELCH, B. E.: Chief, Environmental Systems Branch, USAF School of Aerospace Medicine, Aerospace Medical Division, Brooks AFB, Tex., personal communication, Mar. 1966. 113. WELCH,B. E.; MORGAN,T. E., JR.; and CLAMANN,H. G.: Time concentration effects in relation to oxygen toxicity in man. Fed. Proc. 22: 1053-1065, 1963. 114. WILHITE,W. F.: The Development of the Surveyor Gas Chromatograph. TR-32-425, California Institute of Technology, Jet Propulsion Lab., Pasadena, Calif., May 1963. 115. WILHITE,W. F.; and BURNELL,M. R.: Lunar Gas Chromatograph. ISA Journal, 10: 53-59, 1963. 116. WINSLOW,C.-E. A,; GAGGE,A. P.; and HERRINGTON,L. P.: The influence of air movement upon heat losses from the clothed human body. Amer. J. Physiol., 127: 50.5-518, 1939. 117. WINSLOW,C.-E. A,; HERRINGTON,L. P.; and GAGGE,A. P.: Relations between atmospheric conditions, physiological reactions and sensation of pleasantness. Amer. J. Hyg., 26: 103-115, 1936. 118. WOODS,R. W ; and ERLANSON,E. P.: Thermal Integration of Electric Power and Life Support Systems for Manned Space Stations. NASA-CR-543, Sept. 1966. 119. WORTMAN,J. J.: A Feasibility Study of a Thin Film Oxygen Partial Pressure Sensor. NASA- CR-66084,1%6. 120. WORTZ,E. C.; DIM, R. A.; GREEN,F. H.; et al.: Reduced Barometric Pressure and Respiratory Water Loss. SAM-TR-66-4, USAF School of Aerospace Medicine, Brooks AFB, Tex., Feb. 1966. APPENDIX A Nomenclature

A area, sq ft C fraction of maximum evaporative capacity cd coefficient of discharge CP molar heat capacity at constant pressure, Btulmole-"R or Btulmole-"F C" molar heat capacity at constant volume, Btulmol-"R of Btulmole-"F CFM V, ft3/min Clo clothing heat transfer resistance, 1 Clo = 0.88" F-ftz-hr/Btu D diameter d diffusion coefficient e base of natural logarithms F mole rate of flow, moles/day or moleslsec f fanning friction factor or gray-body view factor .fi fraction of wetted surface of skin

G mass velocity, lblsec-ftz g acceleration due to gravity, g = 32.2 ft/sec2 H humidity h heat transfer coefficient, Btuihr-ft2-"F hD mass transfer coefficient, ft/hr hfiJ heat of vaporization, Btu/lb k thermal conductivity, Btu/hr-ft-"F L clothing thickness, or length, ft M metabolism MAC maximum allowable concentration m molecular weight, lb/mole NLe Lewis number NPr Prandtl number, Cpp/k,dimensionless NRe Reynolds number n number of moles of gas P pressure, psia PL power loss, watts PP vehicle power penalty P partial pressure, psia Q heating rate, BTU/hr R gas constant, ft-lb/lb-OR, or universal gas constant, R= 1545 lb-ft/mole-"R R' Rlm R, Specific humidity coefficient RH Relative humidity 121

261-559 0-67-9 122 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS RP Vehicle heat rejection penalty, lb/watt or lb/(Btu/hr) RQ Respiratory quotient (moles CO2 produced/moles 02consumed) NSC Schmidt number, p/pD, dimensionless T temperature, OF or OR -V volume of compartment, ft3 -V velocity, ftlmin VC critical gas velocity, ft/sec W weight, lb w mass flow rate, lb/day or lb/min Wf weight of fluid at fill, lb WT total weight of pressure vessel and stored fluid, lb WU useful fluid weight, lb X system penalty Y mole fraction of a gas CdA a=233 -~I((Y) -Vm

E emissivity rl efficiency I.L viscosity, lb/sec-ft P density, fluid and insulation, lb/ft3 0- Stefan-Boltzmann constant (0.1713 x 10-8 Btu/hr-ftZ-OR4) 7 time, day or sec 7 time constant, day or sec 4 specific humidity, lb HpO/lb gas Subscripts: a atmosphere acc accessory coz carbon dioxide C critical or clothing or convection can canister crit critical E equipment e system equivalent F feed f fill state H hardware He helium I inert gas or inspired in inner L leakage Ls lines I latent lk lock M metabolism MAT materials -

NOMENCLATURE 123 N nitrogen Ne neon 0 oxygen 0 initial value P power Q of heat R residue r radiation S support S skin or sensible STD standard conditions T total mass t total number U useful V vent W water vapor or work W Wall * steady-state value Superscripts: rate of change with time I per person APPENDIX B Conversion Tables

THE MULTIPLE SYSTEMS of units and measures across the top of the page by multiplying by the used in the basic biological, physical, and engi- given factor. The exponent to the numbers indi- neering sciences is often confusing to those with cates the power of 10 by which the factor is to interdisciplinary interests. The following tables be multiplied. For example, 8.68977-2 is equiv - have been selected from a report by Kinslow and lent to 0.0868977. Underscores indicate exac5 Maj0r.~6Tables of electrical and thermodynamic values. units not covered in this extract may be found Slight numerical differences will be noticed in the report. between the values herein and those in most The reader may convert from the measure of current texts because of the redefinition of the quantities in the units listed on the left to those foot in 1960.

125 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

3 E

Y P

...... icz;::::::.... ,..__.._ . .- ...... :za i; I; :i j .--as:. ... _.... 127

cma ft' gal (U.S.) in' liter yd' (US.)

.ma ...... 1.00000 3.53146-' 2.64171-' 6.10236-' 9.99972-4 1.00000-. 1.30794-* ft' ...... 2.031604 -1.00000 7.48052 1.7- 2.03161' 2.03160-' 3.70370-* gd (U.S.)...... 3.78513' 1.33680-' 1.00000 23.3 3.70533 3.70543-5 4.951 13-3 in2 ...... 1.63071' 5.70704-' 4.32900- 1.00000 1.63866-2 1.63071 -' 2.14335-5 liter ...... 1.oooO3' 3.53156-' 2.64170-1 6.10253' -1.00000 1.00003-~ 1.30790-' ma ...... 1.000001 3,53146' 2.64171' 6.10234 9.99972' -1.00000 1.30794 y&(U.S.)...... 7.w 2.700001 2.01974. 4.6656(r 7.6453w 7.645544 -1.00000

Mass I grain Ib. ton (short)

grain ...... 1.00000 6.47900-* 6.47988-' 2.20571-* 1.42857-4 4.44012- 7.1420%' &...... 1.54323' 1.00000 1.00000-' 3.52739-' 2.2@%22-' 6.05216-s 1.10231-S kg...... 1.543234 1.00000, -1.00000 3.52739' 2.20462 6.05216-* 1.10231-3 oz (avdp)...... 2.03495' 2.03495-1 -1.00000 6.25000-* 1.94256-3 3.12500- Ib...... -7.OOODO' 4.5359F 4.53592-1 -1.60000' -1.00000 3.10809-* 5.00000-' slug ...... 2.25216 1.459394 1.45939' 5.147053 3.21740' 1.00000 1.60870-' ton (short)...... 1.40000' 9.071W 9.07104' 2.00000. 6.21618' 1.00000

day hour Wet miec min seC

day ...... -1.00000 2.400001 8.64000'0 8.64ooo' 1.m 8.61oo(r hour ...... 4.16661-1 -1.00000 3E 3.6woo" 6.owoo1 E @see...... 1.15741-" 2.77770-" -1.00000 1.00000-5 1.66667-1 1.00000- msec ...... 1.15741-8 2.77778-1 -1.00000, -1.00000 1.66667-1 1.00000-' min ...... 6.W-6 1.66667-1 6.00000' 6- -1.00000 6.00000' sec ...... 1.15741-1 2.77778-4 - 1.- -1.- 1.66666-' -1.00000

Angle

min *quadrant revolutions (right de)

deg...... -1.00000 E 1.11111-' 1.74533- 2.77770-a min ...... 1.66667-' 1.00000 1.85105-' 2.90889-' 4.62963-1 quadrants (right ~gk)...... 9.000001 5.400001 1.00000 1.57080 2.50000-' radians ...... 5.72958' 3.43775. 6.36620-' -1.ow00 1.59155-' revolutiona...... 3.60000r 4.00000 6.20320 1.00000 see ...... 2.77777-' 1.66667-1 3.08612-' 4.04015-* 7.71605-7 128 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS Velocity I cmlacc ftlmin ft/sec mlmin mlsec milelhi

cmlaec ...... 1.00000 1.96850 3.28084.' 3.60000-1 1.94384-* 6.00000-1 1.00000-' 2.236%-' ftlmin ...... 5.08000-' -1.00000 1.66667-a 1.82880- 9.87473-3 3.04800 - ' 5.08000-5 1.13636-2 ft/sec ...... 3.04800' -6.00000' -1.00000 -1.09728 5.92484 - -1.82880' 3.04800-1 6.81818-3 km/hr ...... 2.77778' 5.46807' 9.11341-1 -1.00000 5.39957 - ' 1.66667' 2.77778-' 6.21371 - knot ...... 5.14444' 1.01268' 1.68781 -1.85200 -1.00000 3.08667' 5.14444-i 1.15078 m/min ...... 1.66667 3.28084 5.46807 - 2 6.00000 - 3.23974-1 -1.00000 1.66667-a 3.72823-1 mlsec...... 1.00000' 1.96850' 3.28064 1.94384 6.00000-' 1.00000 2.236% milelhr ...... 4.47040' 8.80000' 1.46667 1.60934 8.68976- I 2.682241 4.47040.' __1.00000

Force

dyne a ka newton poundal b

dyne ...... 1.00000 1.01972-3 1.01972-8 1.00000-5 7.2330-' 2.248oy-8 ...... 9.806651 1.00000 1.00000-3 9.80665-3 7.09316-2 2.20462-5 a ~ ...... 9.806655 -1.m 1.00000 9.80665 7.09316' 2.20462 newton ...... 1.WMO" 1,01972' 1,01972-' 1.00000 7.23301 2.24809-' poundal ...... 1.38255' 1.40981' 1.40981-* 1.38255-' 1.00000 3.10809-* Ib, ...... 4.4482ZS 4.535w 4.53594-1 4.44822 3.21740' 1.00000 ~ CONVERSION TABLES 129 130 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS

4 :

W -'E -0 P

-E

PEwc s

3 d

...... CONVERSION TABLES 131

...... , . . , . . 132 ENGINEERING TRADEOFFS OF ONE- VERSUS TWO-GAS SYSTEMS Mass Flow Rate

~~ gmdsec khec Ib./hr !Ih,/rnin Ib./sec sluglsec L gm./sec ...... 1.00000 1.00000-3 7.93664 1.32277-' 2.20462-a 6.85218-5 kgdsec ...... 1.000005 I.oM)o 7.936615 1.32277' 2.20462 6.85218.: b Ib./hr ...... 1.25998.' 1.25998-' LpMMe 1.66667-' 2.77778.' 8.6336V Ib./rnin ...... 7.55987 7.55987-3 6.09000' 1.wooo 1.66667.' 5.18016-* IbJsec ...... 4.53592' 4.53592.' 3.600005 6.00000' ~1.00000 3.1M)o9-' slug/sec.,...... 1.45939' 1.45939' 1. 158265 1.930445 3.21740' 1.00000 I

Pumping Speed or Volume Flow

Wlsec gallmin literlmin literlsec mYhr m"/min

cm'lsec ...... 1.00000 2.11888-5 3.53147-* 1.58503-* 5.99983-* 9.99972-' 3.60000-5 6.00000-s ft'lmin ...... 4.71947* 1.00000 1.66667-' 7.48052 2.83160' 4.71934-1 1.69901 2.83168-' ~ flalsec...... 2.83168' 6.00000' -1.00000 4.488312 1.698%* 2.83160' 1.01941' 1.69901 gallmin ...... 6.30902' 1.33680-1 2.22801-3 1.00000 3.78530 6.30884-' 2.27125-' 3.78541-' literlmin...... 1.66671' 3.53156-' 5.88594-4 2.64179.' 1.00000 1.66667-2 6.00017-: 1.m3-5 literlsec...... 1.oooo35 2.11894 3.53156-' 1.585081 6.wooo' 1.00000 3.60010 6.00017-* mVhr ...... 2.777781 5.88578-1 9.80%3-3 4.40287 1.66662' 2.77770-' -1.00000 1.66667- mJImin...... 1.66667' 5.88578-1 2.64172' 9.99972' 1.66662' 6.00000' 1.00000 i 3.53147'

Temperature [Values are based on the thermodynamic temperature scale as defined by the 10th General Conference on Weights and Measures meeting at Pans in October 1954. Temperature of triple point of water=273.16" K=491.688" R=32.018" F=0.01" C; temperature of ice point of water=273.15" K=491.67" R=O" C=32" F]

To convert from the units below to those on the right, perform 'C "F 'R the indicated Operations in order. I OK "C I x1 x 915 + 32 + 273.15 X9/5+491.67 'F - 32 X 5/9 Xl X 5/9 + 255.372 U 459.67 'K -273.15 X 9/5-459.67 Xl X 915 "R X 5/9 -273.15 - 459.67 x 519 XI

U.S. GOVERUYLUT PRIUTIUG OWICE : 1967 OL-Z~l-S59