COMMERCIAL REFRIGERATION UNIT COOLERS By: William J

Home , Cooler, Humidity, Thermal expansion valve

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

INTRODUCTION

Commercial refrigeration unit coolers, or blower coils as they are sometimes called, are today available in a wide variety of sizes and shapes for almost all current requirements. However, many service engineers today will remember the period, not so long ago, before standardized mass-produced unit coolers were generally available.

A unit cooler is defined as "A factory made encased assembly including means for forcing air circulation, and elements by which heat is transferred from air to refrigerant". The means for forcing air circulation is usually a propeller fan directly connected to a fractional horsepower motor. The element by which heat is transferred is usually a finned coil. Figure 17F21A shows a typical unit cooler of the open-face type.

Finned heat transfer surfaces were in extensive use in other industries many years before they were widely accepted in the refrigeration field. The early commercial refrigeration evaporator's were mostly gravity circulation bare pipe coils, often fabricated on the job. Ammonia, brine and carbon dioxide were the common refrigerants. In the 1920's, finned coils began to be more generally used, although gravity air circulation was still the standard.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

The introduction of sulphur dioxide and methyl chloride as refrigerant's made coil fabrication easier by permitting the use of non-ferrous tubing instead of steel. However, it was the introduction of the fluorocarbon refrigerants in the 1930's that gave real impetus to advancement in coil design and fabrication. From that time on, the use of the forced draft unit cooler has grown steadily, and today there are available a wide variety of standardized units to cover almost all of the applications of commercial refrigeration.

FUNDAMENTAL HEAT TRANSFER FORMULA

The design of a unit cooler evolves around the fundamental formula for heat transfer shown in Figure 17F21B.

Figure 17F21B

Surface is defined as the outside area of the tubes and both sides of the fins, that is exposed to the air flow.

"U" is called the over-all heat transfer coefficient and is defined as the amount of heat which will be transferred by one square foot of coil surface for each degree Fahrenheit mean temperature difference between air and refrigerant.

MTD is the mean, or average, temperature difference between the air and the refrigerant. Difference in temperature is the force which drives heat from one medium to another. Since both the air and refrigerant change in temperature as they pass through the coil, it is necessary to determine the mean difference to obtain the true driving force.

Let us investigate the possible ways in which we could increase the cooling capacity of a given design.

Considering each of the factors in the formula, it is clear that increasing the surface would increase the heat transferred. This, however means increasing the size and cost of the unit, and should only be done after all other possibilities are exhausted. Also, it is easy to see that increasing the temperature difference between air and refrigerant will increase the heat transferred. However, the temperature difference is

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan limited by other considerations, such as humidity control, which shall be discussed later. The only remaining item is, therefore, the heat transfer factor, "U".

Calculation of "U" is an important step in coil design, and careful consideration of all the factors which influence "U" will result in the most efficient unit.

"U" is determined from the following formula:

R = Ratio of outside coil surface to inside surface.

L/k = Resistance to heat transfer offered by the metal of the tubes and fins. hi=Refrigerant film transfer coefficient Btu/hr/sq.ft. inside surface per °F. ho=Outside air film transfer coefficient Btu/hr/sq.ft. outside surface per °F. This formula shows how the transfer factor is built up from three items:

a. The inside, or refrigerant coefficient, R/hi b. The metal resistance, L/k

c. The outside, or air coefficient 1/ho Surprisingly enough, of these three items, the metal resistance L/k is invariably the smallest and can usually be ignored, especially for aluminum or copper tubes and fins. The major resistance to heat transfer is in the thin films of refrigerant and air, which form on the inside and outside of the coil surface.

The temperature drop through the tube wall is small compared to the temperature drops in the air and refrigerant films. Thinner films result in less temperature drop, or in other words higher film coefficients,which results in better transfer rates. Thinner films can often be obtained with higher velocities of air and refrigerant because the faster flow scrubs the wall cleaner.

The temperature changes shown in Figure 17F22A should be distinguished from the temperature change which the air undergoes as it passes through the coil from inlet to outlet. Figure 17F22B is a simplified picture showing how the air changes in temperature as it passes through the coil.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

VALUE OF FINNED SURFACE

Inspection of the "U" formula will answer many of the questions which arise relative to the value of the fins on a coil. The fins increase the R factor to a value usually between 10 and 20. This makes R/hi the effective inside factor, about equal to 1/ho, the effective outside factor, because hi is usually 10 to 20 times as large as ho. Values for hi range from 100 to 500, depending on circuit loadings. Values for ho range from 4 to 8, depending on air velocity. Physically, it is simply a question of putting extra surface on the air side because air is more difficult to heat or cool than a refrigerant fluid.

Sometimes the question is raised as to how many square feet of finned surface is equivalent to one square foot of bare surface. The answer varies with the application, and depends on the relative transfer coefficients inside and outside the tube. With a very high inside coefficient, and a very low outside coefficient, the finned surface is almost equal to bare surface. With a low inside and high outside coefficient, the finned surface would be almost worthless. In the usual unit cooler coil, with R/hi about equal to 1/ho, finned surface on the outside is about equivalent to half as much bare surface. Economically, this can be justified very nicely because finned surface can be fabricated at a much lower cost per square foot than bare tubing. Since we can eliminate the metal resistance as a minor factor, the formula reduces to the following simplified form for a finned coil.

1/U = R/hi + 1/ho

In general, the way to get the highest possible hi, ho, and U factors, is to increase the refrigerant and air velocities to the maximum permissible. However, there are several factors which limit us in this direction and more detailed consideration of these may be of interest.

AIR SIDE HEAT TRANSFER

Turning first to the air side performance, we note that the cfm (cubic feet per minute) of air does not appear in the basic heat transfer formula. However, the quantity of air is a most important item, and its effect on capacity is felt through its influence on the mean temperature difference or MTD. The cooling capacity is related to the CFM through the following formula:

(Cooling Capacity) × (SHR) = cfm × (Air Temp. drop) × 1.08

(Btu/hr) × (Sensible heat ratio = cfm × (°F) × 1.08

"Sensible Heat Ratio" is defined as the ratio of the sensible cooling capacity of the coil to the total cooling capacity. The factor 1.08 is a conversion factor involving minutes/hr (60), cubic feet of air/lb (.075), and specific heat of air, (.24), so 60 × .075 × .24 = 1.08.

Take, for example, a unit cooler designed for one-ton capacity at 15° temperature difference between room and refrigerant. The cfm of air should usually be selected to give a temperature difference of 7° to 8° between the air leaving the coil and the refrigerant temperature. Less temperature difference results in a requirement for more air than is desired from the standpoints of both the cooling unit design and box operating conditions. More temperature difference results in a small quantity of air and a very deep coil. Note that the range over which the air is cooled is about half the theoretical maximum range of 15°, corresponding to the difference between the temperature of the air in the box and the refrigerant temperature. Assuming a sensible heat ratio of 85%, we can calculate the amount of air required for a one-ton coil, as follows:

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

The figure of 1,400 CFM per ton is a rough guide to use in judging the design of a unit cooler which is to be used in a 35° box at 15° TD.

For a freezer box, the unit cooler is more often selected at 10° TD instead of 15°. Under these conditions, the air temperature drop should be about 4° to 5°, which results in approximately 2,000 cfm per ton. Air conditioning coils are usually designed for about 400 cfm per ton, with wide TD's, and it is easy to see therefore, why the coils for a commercial application are larger and more expensive than for an air conditioning job of the same tonnage.

MEAN TEMPERATURE DIFFERENCE

Knowing the air temperature drop through the coil, the MTD is calculated as in the following example, assuming constant refrigerant temperature:

Box Temperature = 35°

Air off coil = 28°

Refrigerant Temp. = 20°/15°

Refrig.Temp.= 20°/8°

MTD should be distinguished from TD. TD is the difference between box temperature and refrigerant temperature, which, in this case, is 15°F. Coils are sometimes erroneously rated on the basis of so many degrees TD between refrigerant and average box temperature, the average box temperature being taken as the average of the air temperature entering the coil and the air temperature leaving the coil. This is misleading because the true average box temperature is very close to the temperature of the air entering the coil. The colder air off the coil is diffused into the warmer box very quickly, and it is only for a very short distance in front of the coil that the air temperature is lower than the bulk of the box.

The MTD of 11.5° calculated above is called the arithmetic average temperature difference and is slightly in error because of the curvature of the air temperature line plotted in Figure 17F22B. The curvature of the line is taken into account in what is called the logarithmic mean temperature difference, which is more strictly correct. In this case, the log mean temperature difference would be 11.1°. The arithmetic mean is therefore close enough for our purposes and will be used hereafter.

CATALOG RATINGS

Occasionally, catalog ratings are shown for coils under conditions which are almost physically impossible. To make a quick check, the catalog rating can be combined with the published cfm, and the air temperature drop through the coil can be easily calculated. On a low temperature coil, for instance, if the air temperature drop figures to be 8° or 9°, as is sometimes the case, then it is almost certain that the coil will never operate at 10°TD. In actual operation the compressor will balance off at a lower suction temperature and the result will be less capacity, lower humidity and higher electric costs.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

AIR FACE VELOCITIES

Air velocities across the unit cooler face are usually between 300 and 500 fpm (feet per minute). Higher velocities would give higher heat transfer coefficients, but would also result in higher fan horsepower. Also, above 500 fpm there is a tendency for moisture to be blown off the finned surface. fpm should be distinguished from cfm. Fpm is the velocity of the air over the face of the coil, measured in feet per minute. Cfm is the total quantity of air moved, measured in cubic feet per minute. Cfm and fpm are related as follows: cfm = fpm × FA

Where FA = coil face area in square feet.

HUMIDITY

Humidity is an important factor in refrigeration for food storage. Assuming that the air unit has been designed with the air volume in CFM in the right range, the major factor which affects the humidity is the TD for which the unit is selected. The best way to understand how a cooling coil removes heat and moisture from air is to plot the air path on the psychrometric chart. Figure 17F24 is a skeleton psychrometric chart showing the performance of a typical unit cooler coil.

The air enters the coil at point A, which corresponds to the assumed box conditions of 35° dry bulb temperature and 85% relative humidity (RH). It is a property of the chart that air passing over a cooling coil with constant refrigerant temperature travels in a straight line in the direction of the coil surface

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan temperature. The coil surface temperature is always somewhat higher than the refrigerant temperature, and in this case, let us assume the refrigerant temperature is 20°F, and the surface temperature 24°F (point B). Since the coil was designed to cool the air to a dry bulb of 28°F, (point C), we see that the air leaves the coil with a moisture content of 21 grains per pound and at a RH of about 96%.

Now let us consider the case where the refrigerant temperature, and correspondingly the coil surface temperature, was lower than in the example above. Each time the air passes through the coil, more moisture will be removed, and in due course the room humidity will be reduced until we balance off to a condition such as shown on the lower line on the chart. For both the upper line and the lower line, the room temperature is 35° and the air leaving the coil is 28°. However, for the higher suction temperature, the air leaving the coil has a moisture content of 21 grains per lb., but with the lower suction temperature, the moisture content drops to 20 grains per lb. As a result, then, the relative humidity in the box falls off from 85% to 81% as the TD goes up.

SELECTING THE PROPER TD

Other factors which affect the relative humidity include the system running time, moisture infiltration, condition and amount of exposed product surface, air motion, outside air condition and type of system control. However, the TD is the major factor, and for general guidance, the following method of selecting air coolers is suggested for temperatures in the range from 25°F to 45°F box temperatures:

TD between box temperature and refrigerant temperature:

Very High Relative Humidity (approximately 90% RH)

8° to 12°F TD for unpackaged cheese, butter, cut fresh meats and flowers.

High Relative Humidity (approximately 80% to 85%)

12° to 16°F TD for fresh carcass meat, fresh fruits and vegetables, fish and most general storage.

Medium Relative Humidity (approximately 75%) 16° to 22°F TD for dried meats, dried fruits, dried vegetables, beer, milk and cream.

For low temperature applications, 10°F or below, 10°F TD is generally used because of efficiency and economy, and frequency of defrosting considerations rather than for humidity control.

HOW TO CORRECT HUMIDITY THAT IS TOO HIGH OR TOO LOW

What can be done to change the humidity in an existing box which is too moist or too dry? If the box is too dry, it is usually the result of too much TD, and little can be done except to make the usual checks to insure that the coil is delivering its maximum capacity. Make certain that the coil is fully fed, that the expansion valve setting is correct, that the air flow is not obstructed or otherwise reduced, and that the control settings are right. We shall discuss trouble shooting of this nature in more detail later.

If the humidity in the box is too high it is usually caused by too close TD, as we have discussed, or by too much refrigerating capacity without resultant too little running time. Within limits, the TD can be increased on a job already installed by increasing the expansion valve superheat or by decreasing the air flow. Both of these will decrease the coil capacity and the result will be more TD and less humidity. A humidity that is too high throughout the entire box should be distinguished from a sliming condition, which might exist in only part of the box. Sliming is usually caused by poor air circulation and can often be cured by insuring effective and uniform air distribution to all parts of the box.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

BEST TYPES OF AIR FLOW

While it is difficult to vary the air over too wide a range, it is possible to minimize dehydration by directing the flow of air to best advantage. Unit coolers are designed to keep the air off the products as much as possible. Instead, the air is directed over the ceiling and walls where it will intercept the incoming heat as directly as possible. In this connection, it is desirable to mount the unit in such a position that air which enters the box as a result of door openings is drawn through the coil before it is allowed to contact the stored products.

HOW CAN OUTER FINS BE BEST UTILIZED?

What can be done along the lines of increasing the outside performance by varying the fin spacing, fin size and fin shape? With respect to fin spacing, the designer has little latitude. The temperature level has a strong practical effect on the fin spacing. For freezer applications, spacings closer than approximately four (4) fins per inch limit the operating time between defrosting. For medium temperature applications in the 35° to 40° range, it is not possible to use much more than six (6) fins per inch with a natural defrost cycle. Only on applications where the refrigerant temperature is above 30° can eight or more fins per inch usually be used.

The fin size is usually varied with the tube size, which is largely dependent on the tonnage of the unit. Units designed for the small capacity range require small tubes which we shall discuss further later. With small tubes the fins must be proportionately small.

Fin shape is a subject which has received much attention. It is possible to increase the air side coefficient by the use of spined fins or other arrangements which result in a high degree of air turbulence. The problem here is cleanliness, and it is usually not practical to use more than a few simple spacing fingers to induce air turbulence.

REFRIGERANT SIDE HEAT TRANSFER

Next, let's look at the inside performance of the coil. Here the basic problem is the same as the outside. A high refrigerant velocity is desirable to give a high inside coefficient. For any given tubing size, the inside coefficient increases as the refrigerant velocity increases and as the tons per circuit increase.

However, this is limited by refrigerant side pressure drop. It is desirable to keep the refrigerant pressure drop low, because pressure drop means temperature drop and this reduces the coil capacity. In our previous calculation of MTD, we had the following figures:

This was based on no refrigerant drop. If we had a pressure drop corresponding to 5°, the figures would have been quite different.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

We would have, therefore, lost about 20% in over-all coil capacity. It is easy to see how we have lost this 20% capacity by referring to the basic formula Q=U×A×MTD. In the example under discussion, since U and A are assumed constant, a 20% decrease in MTD from 11.5°to 9°results in a corresponding decrease in coil capacity of 20%.

COIL CIRCUITING

What is the best method of circuiting a coil in order to get the maximum capacity, but still control the pressure drop? A review of some older and some newer methods might be interesting here.

One of the earliest methods was the float control shown in Figure 17F27A.

This was useful in its day and is still used to advantage, especially in larger systems of the flooded ammonia type. The thermal expansion valve, however, is the standard control today because it is smaller, simpler and more economical.

The single series refrigerant circuit shown in Figure 17F27B is simple and effective up to its limit of pressure drop. With 3/4" OD tubing, 1/2 ton is about the maximum loading permissible on low temperature forced draft coils operating below 0°F suction temperature and 10°TD, and 1 ton is about maximum for coils operating at 20°F to 25°F suction and 15°TD.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

The coil which is headered top and bottom as shown in Figure 17F27C was used to a certain extent, principally before the importance of proper circuiting was fully understood. While the pressure drop is low, the arrangement results in too low velocities, and the inside coefficient is usually poor. Also, the loading on each circuit is different, resulting in part of the coil not being fully effective, particularly with the CFC refrigerants. However, ammonia has higher film coefficients than fluorocarbons, distribution is less of a problem, and headers on ammonia coils are seen more often today than on fluorocarbon coils.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

The split circuit in Figure 17F27D is an effective method of circuiting coils in the size range from 1/2 ton to 2 or 3 tons. The refrigerant flows first through a series circuit until the gas that is evaporated begins to build up the velocity to approximately 1,500 FPM. The circuit is then split into two parallel paths, and the result is maximum coefficient within the limits of the allowable pressure drop. The reason for the lower pressure drop with the split circuit lies in the fact that pressure drop increases appropriately as the square of the velocity. With a split circuit we have only half the velocity, and the pressure drop per foot is therefore only one quarter what we would have with a single circuit. Also the length of tube through each of the split circuits is only one-half of the length through the series circuit. For these reasons, the pressure drop through the split circuit portion of the coil is only one-eighth as much as we would have had with a straight series circuit.

For larger coils, the pressure-type distributor or TX valve and distributor combined is generally used. The coil designer must select the best circuit length and number of circuits for the operating conditions, so as to obtain the maximum transfer coefficients within the allowable pressure drop limits. The question is sometimes asked why pressure drop through the distributor does not affect the coil performance? The answer is that pressure drop is only harmful in the tubing of the coil where heat is being transferred. The refrigerant must be throttled from the high-side liquid pressure to the evaporating pressure anyway, and it makes no difference in the heat transfer if all of the pressure drop is taken across the expansion valve, or if some of it is also taken across the distributor. It should be noted, however, that the distributor pressure drop reduces the drop across the expansion valve, and consequently the valve capacity, so the valve should be sized accordingly.

OVER-ALL COIL DESIGN PROBLEM

Having now considered each of the major variables involved in coil design, it is easier to appreciate the over-all problem in coil design. Using the basic formulas, the cfm, the surface and the circuiting can all be selected to deliver a certain Btu/hr at a given TD. The designer must do a certain amount of juggling and must compromise here and there to come up with the best over-all result.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

HOW TO CHECK A COIL THAT IS ALREADY INSTALLED

Now, armed with the theoretical information we have just reviewed, you might ask how can we check a coil that is already installed to determine if it was properly designed? This would be difficult and would probably not be an entirely satisfactory calculation. It is, of course, best to purchase a coil in the first place only from a responsible manufacturer who has complete data in his catalog on the coil surface, amount of air, Btu/hr rating and other factors. However, let us assume you are confronted with a requirement to straighten out a job already installed which is not performing. Assume the blower coil is of unknown make.

First, let's assume that you have checked over the mechanical operation of the system and all is in good order. Also, let us assume you have made all necessary adjustments to the controls and that the cut-in and cut-out points are correct, and most important, that the TX valve is feeding properly and its super heat setting is right. If there is any question of valve performance, you have opened up the valve just enough to flood out of the unit and then closed it down a bit to insure that there is some superheat but not too much.

The performance of the condenser should be checked and the head pressure should be within reasonable limits. It is desirable to have a suction-liquid heat interchanger in the system and its size should be checked against the manufacturer's ratings. Suction side pressure drop through the heat interchanger must be below very low limits, rarely exceeding 1/2 lb. per square inch.

The next thing to do is to calculate the box heat load by the standard methods available and make sure that the compressor has the required capacity at the proper suction temperature. Now, knowing the Btu/hr capacity required from the blower coil, and knowing also the TD at which it should operate, you can easily calculate the cfm of air required. To check the actual cfm being delivered, a survey of the unit outlet velocities can be made if you have an air instrument, or a survey of the air outlet temperatures can be made and averaged.

Working backwards will of ten give you a clue as to whether or not the air flow is sufficient. If the air flow is low and the face velocity permits, a new fan and motor will sometimes provide the necessary increase.

Then check the refrigerant loading per circuit, and if there is any question in your mind, install a pressure tap between the TX valve outlet and the coil inlet, and one at the coil outlet as shown on Figure 17F28. Connect these taps to the same gauge through a manifold and check the coil pressure drop during normal operation. On a low temperature unit operating at -10°F suction, good design usually limits the pressure drop to 1 psi, which corresponds to about 2°F on R-12. On a unit operating at 20° suction, good design usually limits the pressure drop to 2 psi, which corresponds to about 3°F. In a particular case, of course, the designer may have allowed for a somewhat higher pressure drop, but if the drop is substantially more than the figures given above, it is a pretty good indication of an undersized unit or an improperly circuited unit.

Service Application Manual SAM Chapter 620-38 Section 5D

COMMERCIAL REFRIGERATION UNIT COOLERS By: William J. Donovan

If air cfm and refrigerant pressure drop are within reason, then there is little else you can do. The unit is probably deficient in surface, or the over-all transfer coefficient is low because of improper design. A few simple checks such as outlined above, however, will often confirm your conclusions on capacity trouble and will enable you to recommend with more confidence whatever removals or replacements are necessary.