NOTICE CONCERNING COPYRIGHT RESTRICTIONS

This document may contain copyrighted materials. These materials have been made available for use in research, teaching, and private study, but may not be used for any commercial purpose. Users may not otherwise copy, reproduce, retransmit, distribute, publish, commercially exploit or otherwise transfer any material.

The copyright law of the United States (Title 17, United States Code) governs the making of photocopies or other reproductions of copyrighted material.

Under certain conditions specified in the law, libraries and archives are authorized to furnish a photocopy or other reproduction. One of these specific conditions is that the photocopy or reproduction is not to be "used for any purpose other than private study, scholarship, or research." If a user makes a request for, or later uses, a photocopy or reproduction for purposes in excess of "fair use," that user may be liable for copyright infringement.

This institution reserves the right to refuse to accept a copying order if, in its judgment, fulfillment of the order would involve violation of copyright law.

Geothermal Resources Council TRANSACTIONS, Vol. 14, Part I, August 1990

WELL LOGGING RESULTS FROM THREE GEOTHERMAL FIELDS IN

Bert Dennis and Ed VanEeckhout Los Almos National Laboratory, Los Alamos, New Mexico 87545 USA

ABSTRACT Los Alamos National Laboratory has participated in a prm gram of geothermal resource and field development in cen- tral America (in the countries of , El Salvador, Guatemala, Honduras, and Panama) since 1985. One por- tion of this involvement included utilizing high-temperature well logging tools that had been developed at Los Alamos for the Hot Dry Rock Geothermal Project. These tools were success- fully used at three different geothermal field locations: Mi- ravalles in Costa Rica, Ahuachapb in El Salvador, and Zunil in Guatemala.

INTRODUCTION FIGURE 1 The work described herein was funded by the U.S. Agency WELL LOGGING EQUIPMENT IN AHUACHAPAN for International Development (USAID), and was administered through its Regional Office of Central American Programs (RO- CAP), in Guatemala City. Due credit for assistance on much of this work must go to our counterparts in Central America: in Costa Rica, the Instituto Costarricence de Electricidad (ICE), Well logging operations performed in the Miravalles in El Salvador, the Comision Hidroelectrica del Rio Lempa Geothermal Field in Costa Rica were conducted during two sep- (CEL), and the Instituto Nacional de Electrificacion (INDE) in arate field trips. The Phase I program provided the deployment Guatemala. Many personnel from these organizations greatly of a suite of high-temperature borehole instruments, including assisted in our efforts. the temperature/rabbit, fluid sampler, and three-arm caliper in Well PGM-3. These same tools were deployed in Well PGM- One the Los focuses in Central America was to of Alamos 10 along with an additional survey run with a combination fluid assist in the development of identified geothermal fields in the re- gion. This included Miravalles in Costa Rica, Ahuachaph in El velocity/ temperature/pressure instrument used to measure t her- modynamic properties under flowing well conditions. The Phase Salvador and Zunil in Guatemala. Efforts included well logging I1 program complemented Phase I with the suite of tools de- of geothermal wells, geochemical brine sampling and analysis, ployed in Wells PGM-15, PGM-ll, and PGM-12. reservoir engineering, and geophysics. This paper focuses on the well logging. Well logging operations performed in the AhuachapAn Geothermal Field in. El Salvador concentrated on eight of the Equipment patterned after that developed by Los Alamos wells there. High-temperature downhole instruments including for the Hot Dry Rock Project in the United States wkab furnished a temperature/rabbit , caliper, fluid velocity spinner/ temper* to the region, Figure 1. This allowed the monitoring of downhole ture/pressure (STP), and fluid sampler were deployed in each temperatures, pressures, flows, and well diameters, plus the col- well. The caliper tool was used primarily to determine if chemi- lection of brine samples at depth. These tools are calibrated for cal deposits were present in well casings or liners and to investi- use at temperatures up to 3OOOC in very caustic environments. gate a suspected break in the casing in one well. STP logs were This equipment will be operated and maintained by a consor- obtained from six of the eight wells at various flow rates ranging tium involving CEL, ICE, INDE, Los Alamos, and USAID in from 30 to 80 kg/s. A static STP log was also run with the the future. wells shut-in to provide data to be used in the thermodynamic Training has also been an important part of this program. analysis of several production wells. Los:Alamos and its contractors have made major efforts to trans- Field operations in the Zunil geothermal wells in Guatemala fer the ability to perform and analyze each of the aspects men- included the deployment of the "standard" suite of logging tools tioned above to Central American counterparts. Several training described above in each of four wells ZCQ-3,ZCQ-4,ZCQ-5 and sessions have been held, and certain personnel within the organi- ZCQ-6. Downhole fluid samples were taken after flowing a least zations of CEL, ICE, INDE, have been trained in the techniques one wellbore volume. The STP tool was run in the potential used. production wells at two flow rates.

373 Dennis aiitl VanEcckhout The well logging strategy in each country was to log not the known !gradients, a least-square curve fit to the STI' pressure only the good production wells but to also investigate damaged data in thr: liquid regions gavc the required calibratioa. and non-production wells. The results of measurements in those wells where the thermodynamic analysis could be performed will be presented. WELL PGM-10 MIRAVALLES COSTA RICA The Republic of Costa Rica is pursuing the development of a geothermal power plant located on the southern flank of the Miravalles in the Guanacaste Volcanic Range. This geothermal development would complement the hydroelectric re- sources of the country and iidk eliminate the use of fossil-fueled power plants for generating electricity. The Instituto Costar- ricense de Electricidad (ICE) drilled eight production wells in the Miravalles Geothermal Field that have penetrated a high- temperature geothermal brine reservoir at a depth of less than 2 km. Fluid temperatures exceed 240° and flow rates range from 40 to 100 kg/s. Flow testing of the initial wells also reveal that

calcite scaling could be a potential problem in the wellbores. ....Id Well PGM-10 was a producing well and was logged under 100 I200 OEPTH (m) both static and flowing conditions. A thermodynamic analysis FIGURE 3 of the flowing data was performed using results of the STP logs. STP SPINNER OUTPUT MIRAVALLES WELL During the initial temperature rabbit survey the tool set down PNM-10 at 1160 m because of calcite buildup in the liner as confirmed by The saturation pressures and calculated pressures are shown the caliper survey, Figure 2. Thc water level with the well shut in Figure 5 for both the static and flowing conditions. The in was given to be around 300 m. It was detected at a depth pressures in the two-phase fluid from 0 to 850 m in the flow- of 290 m. Calcite buildup is clearly shown in the liner below ing fluid were lower that the saturation pressures, meaning that 1000 m. the vapor phase was in the superheat region. Assuming no to- tal (enthalpy) loss from the two-phase fluid and knowing both temperatures and pressures, the vapor mass ratio (quality) dis- tribution was calculated for the region between 0 and 850 m. These results are show in Figure 6. These calculations revealed that the vapor enthalpy in the entire two-phase region was es- sentially constant. 5 1127.0 mm 250E

DEPTH (m) STATIC WELL FIGURE 2 THREE-ARM CALIPER SURVEY MIRAVALLES WELL PGM-10

Using the STP instrument, data were recorded in well PGM-10 with the well shut in (static) and again with the well I Obo 150~ki~~--1~~~i~~~~FIGUkEOEPTH (m) 4 flowing at a reported rate of 21 lig/s. Figure 3 shows the spinner iiii..ll data for the flowing condition. Figure 4 shows the data from the STP TEMPERATURE SUXVEY MIRAVALLES temperature surveys for both the static and the flowing condi- WELL PCM-10 tions. For the flowing condition, the spinner and temperature data The spinner data from the liquid region of the flowing well show the boiling region to be between 0 and 850m. At a liquid were used to calibrate the spinner. The spinner output on the surface, where boiling starts, the absolute pressure is the satura- log-in is proportional to the fluid velocity plus the tool speed. tion pressure based on the temperature at that point, and in the On the log-out, the spinner output is proportional to the fluid liquid region, the pressure gradient is the hydrostatic gradient. velocity minus the tool speed. (Tool speeds are obtained from Geochemical data taken on liquid samples from the well gave a time/depth data.) Using these relationships, the proportionality value of 1.00+ for the specific gravity. At an average tempera- constant for the spinner was calculated to be 0.017m/s/Hz. This ture of 236OC, the pressure gradients in the liquid regions would value was in turn used to calculate the fluid velocity distribution be 0.00803 MPa/m. Using the two known values of pressure and for the entire flowing well from the spinner data.

374 1)ennis imd VaiiEeckhout 207-

DEPTH (m) FIGURE 7 MEASURED AND CALCULATED E'LUID VELOCITIES MIRAVALLES WELL PGM-10

r,..,....l....l...~j The caliper data taken in Well PGM-10 showed a mini- 127.0 tam mum radius of 82 mm (3.22 in.) at a depth of approximately 1150 m. The maximum calculated fluid velocity in this region was 1.34 m/s. Using these values and the appropriate fluid den- sity, the mass flow rate from the well calculated to be 23 kg/s. Using this mass flow rate, the vapor qualities and the thermo- dynamic properties of the fluid, the velocity distribution was 3k76.2 mm calculated for the two-phase region. These results and those ob- tained separately from the spinner data are shown in Figure 7. 2 I50.8 mm

FlxURE 8 EFFECTIVE FLOW RADII MIRAVALLES WELL PGM- 10

WELL AH-1 AHUACHAPAN, EL SALVADOR The Ahuachapin geothermal field is located in the north- west section of El Salvador about 120 km from . The first geothermal electric generating plant in Central Amer- ica came on-line in 1975 at Ahuachapin. Its generating capacity DEF'TH (m) grew from an initial 30 MW to 95 MW in a 6-year period. Over FIGURE 6 40 wells have been drilled in the Ahuachapdn field. The high- CALCULATED VAPOR QUALITY MIRAVALLES WELL PGM-10 quality geothermal resource at one time produced 42% of the electrical power used in El Salvador. In recent years, however, power plant output has dropped primarily because well produc- tion has decreased. With a constant mass flow rate, the velocities in a liquid The Los Alamos well logging and geochemistry team, in col- are inversely proportional to the flow areas. using the calculated laboration with La Comision Ejecutiva Hidroelectrica del Rio flow rate of 23 kg/s and the results from the spinner data, flow Lempa (CEL), the agency of the Salvadoran government re- radii were calculated for the well below 850 m. These results are sponsible for Ahuachapin geothermal operations, focused on shown in Figure 8. The dotted line is the inside radius of the the problem of declining well production. Using the high- slotted liner. The calculated radii show the amount of calcite temperature logging tools developed at Los Alamos, the team deposits on the inside of the slotted liner below 1000 m; more made downhole measurements to determine whether production important, they show calcite deposits behind the liner. declines could be attributed to individual well damage, reduc-

375 Dennis and VanEeckhout tion in the central reservoir pressure, or a combination of these and other factors. The well logging data was also to be used to improve existing reservoir models of the Ahuachapin field and used for an integrated reservoir engineering investigation that will culminate in the design and implementation of an effective spent brine reinjection program. Well AH-1 had a static water level depth at 525 m, the main production zone at 554 m, and a maximum temperature of 225OC. The maximum flow rate was 54 kg/s with a flowing wellhead pressure of 40.68 kPa (5.9 psi). The shut-in pressure could build up to 1.71 MPa (248 psi). Temperature surveys and fluid samples where completed as precursors to the iiuid velocity/temperature/pressure (STP) logs. The STP tool was run in AH-1 starting at a depth of 100 m DEPTH (11 and logging the well to 1100 m. The first survey was run with FIGURE 11 the well shut in (static) to provide for in situ calibration in- STP SPINNER SURVEY 30 Kg/s AH-1 formation for the pressure and spinner transducers. A second This plot does indicate a major production zone around 500 to log was run with the well flowing at approximately 30 kg/s. 580 m. The temperature data from this survey show that the vapor-to- The fluid velocity is obtained by subtracting the tool ve- liquid interface was at 600 m (Figure 9), which agrees with the locity or logging rate from the indicated spinner values. The measured borehole pressure and calculated saturation pressure spinner or velocity transducer is normally calibrated for each (based on temperature) intersection plotted in Figure 10. The log. This calibration is determined from the spinner data and spinner output in hertz versus depth is plotted in Figure 11. the tool velocity (time versus depth data) in the liquid region of the well during the static survey. The proportionality constant used for the spinner for this series of logs was 0.0318 m/s/Hz. The mass flow rates were calculated as a function of depth, and the results are shown in Figure 12. The average mass flow in the casing is approximately 30 kg/s. Vapor quality was also cal- culated using the value of enthalpy at the liquid interface and is plotted in Figure 13.

I*

DEPTH (n) FIGURE 9 STP TEMPERATURE SURVEY 30 Kg/s AH-1

. . - '. ..i:o*. '. . ' -.".'do.. 1 ' IdOO' ' $00 DEPTH (m) FIGURE 12 CALCULATED FLOW RATE 30 Kg/s AH-1 WELL ZCQ-4 ZUNIL, GUATEMALA Geothermal exploration began in Guatemala during 1972. Initial studies were performed at the and Zunil geother- mal fields. The volcanic belt that hosts the geothermal area lies in a strip nearly 40 km wide and contains 35 volcanoes (three of which are active). The Zunil geothermal field is located 120 miles northwest of Guatemala City in western Guatemala's volcanic province, near the Cerro Quemado and Santa Maria volcanoes. Preliminary exploration at Zunil began in 1973 and continued through 1977. Deep drilling began in 1977 by the National Elec- &TU (at trification Institute (INDE) as a prelude to a power plant feasi- FIGURE 10 bility study. The drilling program was successful in discovering PRESSURE SURVEY AND SATURATION a high- temperature reservoir encountered at m. PRESSURE 30 Kg/s AH-1 (287OC) 1,130 A total of six exploratory wells were drilled, with four eventually producing steam in commercial quantities.

376 Iknnis n.ud VanEeckhout

n STATIC WELL

(L 5 W a 3 e-- v) ## v) c-- aW

-_ DCPTW fn) FIGURE 13 DEPTH Cm) VAPOR QUALITY 30 Kg/s AH-1 FIGURE 15 PRESSURE SURVEY ZUNIL WELL ZCQ-4 Activities were focused on the immediate need for reser- voir data for additional drilling and reservoir evaluation. The program involved making detailed downhole measurements in production wells using Los Alamos-developed high- temperature logging tools. These measurements were made in the four exist- ing production wells. Data was obtained with the wells static and flowing and included pressures, temperature, fluid velocities, .C casing calipers and downhole fluid sampling. 3 Well ZCQ-4 was a potential production well that was logged with the STP tool under both static and flowing conditions. The well had been flowing for several hours and then shut in when the STP static log was run. A small temperature anomaly oc- curred at about 480 to 500 m where the liquids interface was en- countered, Figure 14, as confirmed by the static pressure record shown in Figure 15. Three flowing logs were performed over a period of about 3 hours and during this time the liquid to vapor interface continued to move down the well. The spinner survey DEPTH Cm) taken during the second flowing log shows the liquid interface FIGURE 16 at 520 m, Figure 16. This survey ale0 indicates a fluid inflow . SPINNER SURVEY ZUNIL WELL ZCQ-4 in the well at 910 m. Pressure and spinner data from the third flowing log, Figure 17 and 18, confirms that the liquid interface had reseeded in the well to a depth of 910 to 920 m. As clearly shown by the pressure curve the well did not reach a steady state thermal condition. It was therefore not possible to calculate the thermodynamics of this well for long term flow.

n STP LOG 1 e FLOWING WELL W a a b MEASURED 3 A I-c LI: id a STP LOG r STATIC WELL I-W

2

loox200 400 600 800 1000 I '0

DEPTH Cm) mCURE 14 DEPTH Cm> TEMPERATURE SURVEY ZUNIL WELL ZCQ-4 FIGURE 17 PRESSURE SURVEY ZUNIL WELL ZCQ-4

377 Dennis aid VanEeckhout

' ' " " ' ' * ' ' ' ' " ' ' 'I ' ' ' ' " ' ' " 'I " " ' nN 400rSTP LOG 4 YI FLDWINC WELL

I- cR3 03

L1: FOCUSIN0 MAQNq Wz c. i a ln 1 POF

DEPTH Cm) FIGURE 18 SPINNER SURVEY "NIL WELL ZCQ-4

CONCLUSION SPINNER-P~ERATURe-PRESSU~FIGURE 19 STP TOOL To determine the thermodynamic state of a flowing well- bore, simultaneous measurements of temperature, pressure, and fluid velocity are necessary. The high-temperature (3OOOC) STP (Figure 19) was developed specifically to measure these param- eters in the production wells in Central America. The fluid velocity transducer (spinner) incorporates a rotating impeller BIBLIOGRAPHY with hardened steel pivot bearings. The rotation shaft operates a reed switch that transforms the rotational speed of the im- B. R. Dennis, S. P. Koczan, and E. L. Stephani, "High- at peller to pulses recorded as frequency in hertz. The rotational Temperature Borehole Instrument ion ," Los Alamos X* speed in hertz is proportional to the velocity of the fluid relative tional Laboratory report LA-10558-IIDR (October 1985). to the logging tool. A proportionally constant is determined by B. R. Dennis, "Symposium on High-Temperature Well-Logging logging in the liquid-filled region of the borehole. Instrumentation;" Los Alamos National Laboratory report The temperature sensor is a thermistor that has been cal- LA-107454 (November 1985). ibrated to an accuracy of 0.10' up to 300OC. The thermistor B. R. Dennis, "Vapor Mass Ratio Measurements in Geothermal provides very fast response and exceptional resolution. The pres- Production Wells," Geothermal Resources Council Transac- sure transducer provides accurate pressure measurement in the tions, Vol. 11 (October 1987). geothermal fluids when meticulous calibration procedures are P. Chavez, B. Garcia, and J. Kolar, "Passive Electronics Sys- used. Pressure measurements are made in the STP tool using tems for Logging and Characterization of Geothermal Wells a 0 to 34.5 MPa (5000 psi) potentiometer gauge pressure trans- in the Miravalles Geothermal Field, Costa Rica," Los Alamos ducer. The constant current excitation is sensed at the pressure Laboratory document LA-UR-86-1569. gauge so that the power supply at the surface can compensate B. R. Dennis, R. Lawton, J. Kolar and A. Alvarado, 198923, for line losses over the 3000 m armored cable. "Results of Investigation at the Miravalles Geothermal Field, The analysis of the STP data demonstrates the value of Costa Rica, Part 1: Well Logging," Los Alamos National Lab- this newly developed logging tool in analyzing the conditions of oratory report LA-11510-MS, Part 1, 18 pp. a flowing natural geothermal well. Simultaneous measurements B. R. Dennis, F. Goff, E. VanEeckhout and R. Hanold, 1989a, of temperature and pressure are needed to determine the ther- "Results of Investigation at the AhuachapBn Geothermal modynamic state of the fluid, and the additional fluid velocity Field, El Salvador-Part 1: Well Logging and Brine Geochem- measurement allows mass flow rates to be calculated. istry," Los Alamos National Laboratory Report, to be pub- Also, using the data taken in the liquid region of the well, an lished. in situ calibration can be made on the fluid velocity transducer. This calibration can then be used to calculate the fluid velocities in the entire well. This is demonstrated by the comparison of the velocity distribution calculated using the spinner data and that calculated using the thermodynamic fluid properties obtained from the temperature and pressure data.

378