Spe 133268 MS PDF

SPE 133268
Successful Application of CO2 Tracer Technology for Surveillance of Gas

Lifted Wells
Fathi Shnaib, SPE, Production Engineering Team Leader, Dubai Petroleum, Manickam S. Nadar, SPE, Consultant
Production Engineer, Smart Zone Solutions, Sreekumar M.P. Lead Production Optimization Engineer, SPE, Dubai
Petroleum, Ponnuvel K., SPE, Production Optimization Engineer, Dubai Petroleum and Larry Peacock, SPE,
AppSmiths
Copyright 2010, Society of Petroleum Engineers
This paper was prepared for presentation at the SPE Production and Operations Conference and Exhibition held in Tunis, Tunisia, 8–10 June 2010.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed
by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or
members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is
restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
Abstract
Well surveillance is a key component of production optimization cycle of the gas lifted assets. Even though monitoring of the
surface well parameters can give some clues on the downhole gas lift system performance of oil wells, Flowing (pressure and
temperature) Gradient Surveys (FGS) in most cases is still required to check the health of the downhole gas lift system and
troubleshoot the downhole inefficiencies of well gas lift operations. The FGS has certain level of risk as it involves running
electronic memory gauges on wireline in the well at flowing conditions.
An emerging technology using CO2 tracer was implemented as a pilot project in eleven gas lifted wells offshore fields Dubai. The
results show that this method is very effective for the quick and reliable determination of lift gas entry points in the well. The
method can detect the operating lift depth, detect multiple points of injection and even detect tubing leaks.
It is true that the downhole pressure and temperature data cannot be obtained from tracer surveys. Nevertheless, the technology can
be used as an alternate tool to FGS for the following advantages:
1. Minimum equipment hook-up and no need to shut-in or choke the well, means no production loss.
2. Less equipment required for the survey, reduces logistics issues especially in offshore locations.
3. No wireline tools introduced into the wellbore, eliminate well intervention risk
4. Suitable for wells where a FGS is not possible due to well slugging, significantly deviated or with downhole obstruction.
5. Possible to run tracer survey on 3-4 wells on a day if the wells are located at the same platform.
This paper will describe the objectives of the pilot project, the well candidate selection criteria, details of the execution of the
survey and interpretation of the results. Practical suggestions for getting the best results from such surveys will also be provided.
Introduction
Gas lift is one of the most common methods of artificial lift used for improving field production. “It is an excellent application for
formations that have a strong water drive, or in water flood reservoirs with good productivity indices (PI’s) and high gas-oil ratios
(GOR’s) (ref: 1). Compared with other artificial lift methods, continuous flow gas lift requires less major well intervention activities
and therefore considered particularly suitable for offshore installations where well intervention could be expensive.
2 SPE 133268
Continuous flow gas lift involves continuous injection of high pressure lift gas into the production conduit of the well. Figure 1
shows the schematic of a continuous flow gas lifted well with gas lift injection down the annulus and the production flow up in the
tubing. The injected gas mixes with and travels along with the reservoir fluids to surface. The gas lift injection reduces the density
of the vertical fluid column in the production conduit. As the fluids move towards the surface, the pressure continues to decrease
and more associated gas comes out of the oil. This combined with the expansion of gas increases the velocity of the fluid stream.
The reduction of density helps decrease the bottom-hole pressure, thus creating a reservoir drawdown.
Gas lifted well completions often require more than one gas lift mandrel to be run as a part of the well completion. The gas lift
valves are set inside the gas lift mandrel pockets. Majority of the gas lifted wells use the production casing x tubing annulus for gas
lift injection and the tubing for production. Gas lift unloading valves help remove the completion fluid from the annulus into
production tubing during the well unloading process. This allows the high pressure gas to reach the designed injection point in the
string. A deeper injection point results in a longer column of reduced density, thus enhancing the reservoir drawdown. The gas lift
valve installed at the designed operating depth is called ‘operating valve’.
The unloading valves have a moving stem and a fixed seat so that the valve can either close or open. While the well pressures offer
the opening force, a compressed spring or nitrogen charged bellows offer the closing force in these valves. Once the well
unloading is completed, the unloading valves are in closed position. The operating valve is usually an orifice or a venturi valve and
does not incorporate a closing mechanism. The operating valve will facilitate gas lift injection once the gas lift mandrel at the
opering depth is uncovered and a positive pressure differential is established between the annulus and tubing at the depth of the
mandrel. Both the types of gas lift valves feature a check valve mechanism to prevent back flow of fluids from tubing to annulus.
A ‘dummy valve’ is used block the communication path between the casing and tubing through a gas lift mandrel.
A dummy valve may be installed if the gas lift mandrel may be required in future. A dummy valve may be replaced with active gas
lift valve using wireline methods when required.
Gas Lift Efficiency
An efficient gas lift operation requires the following key requirements:

• Single point gas lift injection (all unloading valves should remain closed)
• Gas lift injection as deep as possible (to maximize the drawdown)
• Stable gas lift supply pressure sufficient to sustain the operating depth
• Optimum amount of gas lift injection rate
• Low flowing wellhead pressure (as low as practically possible)
Gas Lift injection through multiple or shallow depths means reduction of gas lift efficiency, hence loss of production.
The objective of gas lift surveillance scheme is to ensure that the conditions required to maximize gas lift efficiency are maintained
in the field. The field conditions are dynamic and therefore a continuous surveillance is important to ensure that gas lift operates
with maximum efficiency at all times.
Gas Lift Surveillance Process
Gas lift surveillance can be broadly classified into two types:

• Surface surveillance
• Sub-surface (downhole) surveillance
Surface Data: Advanced real-time gas lift surveillance systems are now capable of capturing the well surface data from the field
and displaying the trends of the captured data in the screens of engineers located far away from the field (ref: 2). Typically the
parameters captured are: gas lift injection pressure, gas lift injection rate and flowing wellhead pressure (see Figure 2). This may
also include flowing wellhead temperature. By looking at the values and trend patterns it is sometimes possible to identify if the
well has any gas lift related issue. However it is not possible to frame conclusions by reviewing surface data alone.
Sub-surface Data: A continuous measurement of flowing bottom-hole pressure (FBHP) and production rate may be ideal to
continuously assess and monitor the performance of a gas lifted well. However, many wells do not have permanent gauges
installed to measure FBHP. Continuous testing of wells is not feasible in most cases. Therefore, in order to gather downhole data,
SPE 133268 3
flowing gradient surveys (FGS) are run in gas lifted wells at periodic intervals. Downhole pressures and temperatures are
measured during the FGS at various depths in the production string with the well on production. A production well test is also
carried out simultaneously. This set of data is used to analyze gas lift performance of the well.
“In gas lift systems, downhole equipment and surface facilities are closely related. Because well parameters and conditions like
reservoir pressures are dynamic, producing operations change over time” (ref:3). Therefore in order to perform a proper analysis of a
gas lifted well, the surface and sub-surface data should be considered together during the analysis process.
FGS Operations
In order to carry out an FGS, electronic gauges are lowered into the well using wireline while the well is on production. The
gauges will be run to the end of tubing, kept at this depth till the well stabilizes and then pulled out of hole, making gradient stops
at predetermined depths. These depths are mostly 30-50 feet above and below each gas lift mandrel. An electric line will allow a
surface read-out of the P and T data. Use of slick line with electronic memory gauges is more common. The data will be
downloaded from the memory gauges after completion of the survey. The gradient stop timings will be correlated to the stop
depths thus creating a depth versus P and T data table from the gauge data. The use of a wireline depth recorder assists in creating
an accurate depth vs. data file, eliminating the possibility of keying in the wrong depth data. The P and T data will be plotted
against the true vertical depth (TVD) of the well and then analyzed.
FGS Advantages
“Subsurface pressure surveys are the best and most widely used method of properly analyzing gas lift installations” (ref: 4). Flowing
gradient survey is a very efficient tool to analyze and optimize gas lift. A properly run FGS will help the engineer ascertain
whether gas lift is functioning as designed. Following are the specific advantages of FGS as a well surveillance tool:
• The FGS data gives the FBHP of the well
• The flowing pressure and temperature gradient in the production conduit can be calculated from the FGS data.
• The FGS can identify the lift gas injection depth in the tubing string. This is indicated by a reduction in the flowing
pressure gradient, usually supported by a decrease in temperature across the gas injection mandrel.
• As the FGS gives a direct FBHP measurement for the gas lifted well, tracking the historic trend of FBHP of a well can be
used to analyze the reservoir and well performance. Furthermore, if the reservoir pressure is known the productivity
index (PI) of the well can be calculated.
• The flowing gradient measured during the survey can be used to model the hydraulics for similar wells. After verifying
well test data is reasonably accurate, the multi-phase flow correlation that gives the closest match with the measured data
can be identified. The matched correlation can be used to model similar wells that do not have FGS data. This is often
used for designing development wells in the field.
A carefully acquired FGS data gives a lot of insight into the overall well performance and its optimization potential. Thus the FGS
is much more valuable than a gas lift surveillance tool.
FGS Limitations
The biggest limitation of FGS is in the execution of the wireline work that is required to acquire the data from the live gas lifted
well. The limitations are:
• The data acquisition involves running a wireline tool string with pressure and temperature gauges to the end of tubing
string in a live gas lifted well against the direction of the flow. It is possible that the tool string may get blown up hole,
resulting in a fishing job and loss of production. If the well is slugging (experiencing flow rate fluctuations) the risk
associated with the wireline work is very high.
• As the gradient stops are made above and below gas lift mandrel stations during the FGS, the survey can identify the
mandrel(s) through which gas lift injection is occurring. FGS may not be able to identify the minor gas leak through
tubing joints or any hole in the tubing at other depths.
• The wireline equipment mobilization, rig up, rig down and operations have implications with regard to the HSE exposure
of the personnel and equipment.
• There could be well production downtime associated with rig up/rig down operations. Most of the operations require a
dummy run in the well prior to RIH with the electronic gauges. This will increase the time of wireline operations.
• The stops at gradient stop stations may have to be extended for longer periods in surging wells. This increases the
duration of the survey.
• Sometimes it becomes impossible to run the FGS tool string in wells without reducing the flow rate. Reducing the flow
rate for the purpose of running the gauges would alter the well conditions during the survey.
4 SPE 133268
• In a small sized production tubing the bypass area around the tool string would be small, causing additional back
pressure on the formation during the survey.
• An FGS gives a picture of the well conditions at the time of the survey. Analysis is required to confirm if the survey data
can be taken as a representative data for normal flowing conditions.
Alternate gas lift surveillance tool

The CO2 Tracer Technology was proposed as a low cost effective method to quickly and reliably determine lift gas entry points in
the well. This method has been described in the literature as early as 1993 (ref: 5). The results of a 20-well pilot test with the CO2
Tracer Technology carried out by Petroleum Development Oman (PDO) have been reported in the literature (ref: 6). It is claimed
that the CO2 tracer survey would:
• Detect operating lift depth
• Detect multiple points of injection
• Detect tubing leaks
The application requires injecting liquid CO2 into the annulus along with the lift gas and capturing the CO2 concentration in the
produced gas. As the liquid CO2 moves down in the annulus, it moves into the tubing along with lift gas wherever there is a path of
communication. Once the CO2 enters the tubing, it becomes gas and moves up along with the produced fluids. The exact depth of
entry can be calculated by measuring the travel time of CO2 in the annulus and tubing. The gas lift injection rate, surface pressure
and temperature should be recorded along with flowing wellhead pressure. The well should be tested during the tracer survey and
the production rate measured. The tubing / casing dimensions and depths should be accurately known from as-built completion
records.
The CO2 Tracer System consists of three main elements: a portable separator, a CO2 Analyzer and a Data Acquisition Module.
The Separator enclosure contains a blow-down type separator and pressure/flow/filter controls for supplying clean regulated
process gas to the CO2 Analyzer. The Data Acquisition Module collects sensor information and sends this information to a laptop
computer for real-time display and storage. A small quantity of CO2 sufficient for detection is injected into the well’s injection
line as a single “pulse” at the start of the survey. The test is terminated once all CO2 returns have been observed at surface and/or
the test has run sufficiently long to allow for the return of all the injected CO2. Figure 3 shows the schematic of a typical set up for
the tracer application.
CO2 Tracer Accuracy
The well tracer measures concentration of CO2 over time and times are related back to depths using a well model. Prior to
performing the survey it is required to build a well model using a nodal analysis package incorporating the well completion
mechanical data, reservoir data, PVT data, production well test and surface parameters. The velocity of gas in the annulus and the
produced fluids in tubing string are used to determine the depth of the CO2 injection. Therefore the accuracy of the results is
directly related to the accuracy of well test data and other well information input into the model. While it is relatively easy to get
an accurate measurement of the surface gas lift data, the accuracy of well test data is always a challenge.
CO2 Tracer Advantages
Following are the major advantages which led us to conduct the pilot study:
• The most important benefit of using this technology is the non-usage of well intervention tools. Lift depth can be
determined directly from surface data.
• Well is not shut-in for the survey; no production down-time
• Minimum equipment hook-up and simple operation
• Suitable for cases where pressure surveys are not feasible such as surging wells and highly deviated wells.
• Some smaller points of gas injection can be missed in FGS due to lack of pressure gradient change or temperature
cooling effect.
• Takes less time than a FGS. Possible to run tracer survey on 3-4 wells on a day if wells are at the same location.
• Can detect tubing leaks.
CO2 Tracer Limitations

• Even though the gas entry point can be determined, the downhole pressure and temperature data are not obtained. The
downhole data is required to confirm well parameters to ascertain the optimization potential of the well. This is critical to
SPE 133268 5
determine and propose changes in gas lift depth or gas lift rates in order to maximize production from the well.
• The FBHP and the flowing gradient are estimated using multi-phase flow correlations and the accuracy of this estimation
will depend on the suitability of the correlation that has been selected. FGS will be required at some point in time to
validate these predictions.
• Accuracy of results depends on the accuracy of well test data and surface gas lift data.
Pilot Study Results Summary
The main objectives of the CO2 tracer pilot project was to field test and confirm if the technology can be used effectively for the
offshore gas lift operations of Dubai Petroleum. Eleven wells were identified for conducting the pilot study with the CO2 tracer
technology. These wells were located in three satellite platforms in one of the offshore fields.
Following objectives were considered while identifying the wells for the pilot project:
• Run CO2 tracer survey on wells that have recent flowing gradient surveys carried out. This would re-confirm the lifting
depth as observed from the FGS.
• Run CO2 tracer survey on wells where a recent gas lift valve change has been carried out. Use the survey to verify if the
well has transferred to the new proposed gas lift mandrel.
The results summary of the CO2 tracer pilot project is shown in Figure 4. Details of the results are presented in Table 1. The break
down of the results from these 11 surveys is as follows:
• Results from seven (7) surveys showed possible point of gas lift injection; survey results fairly matched well with the
FGS data.
• Results from three (3) wells were inconclusive; most likely due to insufficient quantity of CO2 injected during the test.
• One test failed as the separator of the analyzer got plugged. No data available.
The test duration varied with lift depth and annulus volumes. In most of the wells, the test time was less than two hours, with the
longest duration being 3-1/2 hours.
The details of the pilot tests from Well 1 through 4 are detailed in the following sections.
Pilot Study Details – Well No. 1
Well No.1 has been completed with combination tubing string, with a 4-1/2” x 3-1/2” tubing crossover at 4991 ft MDRKB,
installed in 9-5/8” casing. Details of the gas lift string and gas lift mandrel depths are shown in Table 2.
A FGS had been conducted in this well in June 2007. The objective was to compare the finding of the CO2 Tracer survey with that
of the FGS. The FGS had clearly indicated single point lifting from the bottom most mandrel (GLM-9) without any significant
leak through any of the unloading valves in the string.
CO2 Tracer survey was performed in this well by injecting 23.5 pounds of CO2 and monitored the return for about 90 minutes.
Only one spike of CO2 concentration was observed and that was after about 70 minutes. The time versus CO2 concentration plot
during the tracer survey is shown in Figure 5. The single spike confirms the single point gas lift injection in this well.
Table 3 shows the calculated velocities in the annulus and tubing and the time that would be required for the CO2 to reach specific
downhole depths in the well. The time required for the CO2 travel was estimated in the annulus and tubing using the well model.
The production test taken during the tracer survey was input in the model. Calculations indicate that it would require about 70
minutes for the CO2 tracer to travel to GLM-9 and return back to surface.
Based on the elapsed time of the CO2 detection, the depth of gas lift injection was calculated as 7,354 ft. MDRKB (see Table 4).
The actual depth of the mandrel (GLM-9) is 7454 ft. Thus the calculated depth of gas lift injection was different from the actual
depth of mandrel 9 by only 1.36%.
The conclusion for Well No. 1 was that the well was operating normally and lifting through the orifice in gas lift mandrel no. 9
which was the bottom-most mandrel. The tracer survey confirmed the single-point lift as indicated by the FGS. No follow-up
action was required.
6 SPE 133268
Pilot Study – Well No. 2
Well No. 2 has 3-1/2” OD tubing installed in 7-5/8” OD casing. Details of the gas lift string and mandrel depths are shown in
Table 5.
Based on a FGS recommendation, gas lift valves (GLV) had been changed in this well in August 2007. This redesign was installed
to deepen the gas lift operating point from gas lift mandrel no. 8 to no. 9. The objective of the CO2 Tracer survey was to evaluate
the gas lift string performance after this GLV change.
CO2 Tracer survey was performed in this well by injecting 30 pounds of CO2 and monitored the return for about two hours. The
time versus CO2 concentration plot during the tracer survey is shown in Figure 6. There were two spikes observed. The first spike
of CO2 concentration was observed after about 52 minutes followed by the second spike after another 10 minutes (i.e., after 62
minutes of injection). As per calculations shown in Table 6 and Table 7, gas lift injection depths were estimated as 6,196 and 7,293
ft. MDRKB respectively. These calculated depths are deferred to the actual depth of mandrels 7 and 9 (6148 and 7326 ft.
MDRKB) by only 0.07% and 0.63% respectively.
The conclusion from this survey was that the well was multi-point lifting. A large amount of lift gas was being injected through the
orifice at the mandrel no.9. A fairly large amount of gas was being injected into the mandrel no.7. Nodal analysis indicated that
under the same conditions, single point injection at mandrel no. 10 could increase fluid production by 4%. The recommendation
was to redesign gas lift in this well to have single point lift from the orifice valve.
No FGS was performed after the Tracer survey to confirm the finding especially to confirm the leaky unloading valve in GLM-7.
Following an unloading failure in Oct 2007, the gas lift string was subsequently re-designed to lift the well from GLM-10.
The objective of the Well Tracer survey was to evaluate the gas lift valve string status and lift point after changing the gas lift
valves in early September 2007. The lift point had been redesigned to GLM-10. As per previous FGS there was a possible leak
through a defective gas lift valve in GLM-4 which could not be replaced due to wirelining difficulties. This well has 4-1/2” x 3-
1/2” combination tubing installed in 9-5/8” casing. The gas lift string details and mandrel depths are shown in Table 8.
CO2 Tracer survey was performed in this well by injecting 9.0 pounds of CO2 and monitored the return for about 105 minutes. The
time versus CO2 concentration plot during the tracer survey is shown in Figure 7. Three spikes were observed. The first spike of
CO2 concentration was observed after about 70 minutes followed by the second spike after another 6 minutes. A third spike was
observed after about 105 minutes of the CO2 injection. As per calculations shown in Table 9 and Table 10 the calculated depths of
injection were 5,237 ft., 5,679 ft. and 8,102 ft. MDRKB. The nearest mandrel depths in the well are 5319 ft (GLM-4), 5679 ft
(GLM-5) and 8084 ft (GLM-9). The calculated depths are deferred to the actual depth of mandrels 4, 5 and 9 by 1.56%, 6.33% and
0.22% respectively.
Analysis of this tracer survey did not provide any conclusive result. The survey indicates possible leak at the fourth mandrel and
below. However, the shift in CO2 concentration is only 0.4 percent and the CO2 returns are inconclusive. Actually the valve in
GLM-4 is a dummy valve. Unloading valves in GLM-6 through 9 had just been replaced prior to the tracer survey. The
recommendation was to run another Tracer survey with more quantity of CO2 which can detect the leak more clearly.
This well had some history about the well completion integrity prior to performing this tracer survey. This well has a 3-1/2”
completion installed in 7-5/8” casing. The gas lift string details and mandrel depths are shown in Table 11. The designed lift point
is the bottom gas lift mandrel at 9988 ft (GLM-12), equipped with the operating valve. A 2004 FGS confirmed single point lift at
GLM-12. Prior to the CO2 tracer survey, slick line tools had identified possible tubing damage based on metal marks obtained on a
lead impression block at GLM-7 located at 7138 ft. Suspecting a parted tubing, gas lift rates were changed to verify the response in
the gas lift injection pressure. This test confired that the gas lift injection is going through a controlled restriction (GLV). Further
trouble shooting performed with Echometer indicated annulus fluid level GLM-12, suggesting gas lift functioning as designed.
CO2 Tracer survey was performed in this well by injecting 13.0 pounds of CO2 and monitoring the return for about 135 minutes.
SPE 133268 7
The time versus CO2 concentration plot during the tracer survey is shown in Figure 8. As per calculations from Table 12 and Table
13, the survey indicates a clear single point gas lift injection at GLM-5 at 6078 ft. This result could not be explained at all. It was
suspected that the well surface parameters and the well test readings may be incorrect thus causing this discrepancy. A follow-up
FGS performed in 2008 showed the well being gas lifted through the bottom mandrel.
Lessons Learned
1. For the velocity calculations to be correct, the well parameters and well test data have to be fairly accurate. If these are
not accurate the conclusions from the survey may not be reliable.
2. The CO2 tracer survey showed reasonable match with the flowing gradient survey results in most cases. However, in
wells where the gas lift mandrels are spaced close to each other and measurement issues exist, the CO2 tracer survey may
not be able to indicate the exact gas lift mandrel of injection. In wells using production pressure operated (PPO) valves
the mandrel spacing will be closer and such wells would be affected by this issue.
3. The weight of CO2 injection should be more than 20 pounds per well and should maintain more than 500 psi differential
pressure between the CO2 source and gas injection pressure. The gas lift injection pressure in the wells is in the range of
900 – 1400 psig. Hence, 45 Kg CO2 with 3000 psi N2 cylinders (special category) should be used to increase the CO2
pressures.
4. Well should be tested while performing CO2 Tracer survey. The well test data and the well parameters input for anlaysis
should be accurate.
5. The well modeling software package has to have the provisions to feed the results from the CO2 Tracer survey.
6. The following operational issues will need to be addressed:
• The equipment is not intrinsically safe and must be located in a safe zone
• The equipment is resistant to environment but not built as completely waterproof nor dustproof
• Equipment requires constant connection to a power source with 110 VAC – 240 VAC @ 50/60 Hz.
Conclusions
1. The biggest advantage of the CO2 tracer technology is that it is not invasive – there is no need to run a wireline toolstring
in the well to gather data to ascertain the gas lift efficiency of the well. This eliminates the risk of losing wire or toolstring
in the well.
2. Just like any new technology this requires several precautions to be taken before running a survey. The weight of CO2
injected is an important consideration for reliable results.
3. If used with all precautuions the tracer technology can be used as a valid downhole gas lift surveillance tool. However it
should not be considered as a substitute for flowing gradient surveys.
4. By combining the outcome of the tracer survey in a nodal analysis model, a well performance analysis may be possible.
However the inflow and outflow models will have to be assumed as there is no downhole pressure measurement made.
5. The CO2 tracer technology can supplement the FGS for the gas lift modeling, analysis and trouble-shooting of a well. The
FGS will still be required to measure the downhole pressures and temperatures, but may be less frequently.
Nomenclature
BOPD Barrels of oil/day
BFPD Barrels of fluid/day
STbopd Stock tank barrels of oil/day
FWHP Flowing wellhead pressure
FBHP Flowing bottom-hole pressure
GL Gas Lift
GLR Gas-liquid ratio
GOR Gas-oil ratio
HSE Health, Safety and Environment
IGLR Injected gas-liquid ratio
8 SPE 133268
IGOR Injected gas-oil ratio

MDRKB Measured Depth from Rig Kelly Bushing
PI Productivity Index
PTRO Test Rack Opening Pressure
PTOV Pressure to open the valve
TGLR Total gas-liquid ratio
TVD True Vertical Depth
References
1. Lea, James F., Texas Tech University: Artificial Lift Selection, SPE Production Handbook.
2. Shnaib, Fathi; Nadar, Manickam S. and Chacon-Fonseca, Joel: Enhanced Field Surveillance through Real-Time Data
Visualization in Dubai Offshore Fields, SPE 126682, presented at the SPE North Africa Technical Conference and
Exhibition held in Cairo, Egypt, 14–17 February 2010.
3. Fleshman, Roy and Obren Lekic, Harryson: Artficial Lift for High Volume Production, Oilfield Review, Spring 1999,
Schlumberger Journal.
4. Brown, K. E., et al.: The Technology of Artificial Lift Methods – Volume 2a, PenWell Publishing Co., Tulsa, Oklahoma,
1980.
5. S. L. Wellington, J. F. Simmons and E.A. Richardson: An on-line method for trouble-shooting gas lifted wells without
wireline tools, SPE 26593.
6. Peringod, Chandran; Clark, James; Barboussat, Didier (PDO); Peacock, Larry (Appsmiths) and Holloway, Jeff (Oryx):
Gaslift diagnosis using Well Tracer Survey- successful trials in 20 wells of Petroleum Development of Oman, paper
presented at the 2009 Middle East Artificial Lift Forum held in Bahrain, 17-18 February, 2009.
SPE 133268 9
Produced fluids
+ GL gas
GL Injection
GLM with gas lift

unloading valves
GLM with gas lift

operating valve
Annulus Fluid Level
GLM with dummy valve
Production Packer
Perforations
Figure 1: Simplified Schematic of Tubing Flow GL Well
Gas Lift Manifold Pressure
Gas Lift Injection Pressure
Wellhead Flowing pressure
Gas Lift Injection rate set point
Actual GL Injection rate
Figure 2: Gas Lift Surveillance Chart

10 SPE 133268
Figure 3: Schematic showing CO2 Tracer Survey Set up

SPE 133268 11
CO2 Tracer Survey Results
Successful results
Questionable results
Failed survey
Figure 4: CO2 Tracer Results Overall Summary
Figure 5: CO2 Tracer Survey Results for Well-1

12 SPE 133268

SPE 133268 13
8.5
8.0
Concentration (%)
7.5
7.0
6.5
6.0
0:00:00 0:16:40 0:33:20 0:50:00 1:06:40 1:23:20 1:40:00 1:56:40 2:13:20
Time (hh:mm:ss)

14 SPE 133268
Well No. Designed Weight of CO2 Injected, Result / Comments

Lifting pounds
Mandrel
1 9 23.5 Lifting from 9th mandrel without any leak in unloading valves.
2 9 30.0 Multi-lifting from 7th and 9th mandrels
3 13 10.5 Inconclusive result due to less quantity CO2 injection
4 6 9.0 Lifting from 6th mandrel without any leak in unloading valves.
6 9 9.0 Lifting from 9th mandrel without any leak in unloading valves
11 Separator of the analyzer got plugged. No data. Test failed
Table 1: CO2 Tracer Survey Pilot Test Summary Results
Mandrel Depth Ft. Depth Ft. Temperature Valve Port Size PTRO Pcsg (psia) PTOV
No. MDRKB TVD Degree F Type (64th in) (psia) (psia)
1 2353 2311 172 Dummy - - 835 -
2 3698 3616 188 Dummy - - 868 -
3 4818 4704 200 PPO 20 961 896 1158
4 5542 5400 205 PPO 20 1022 914 1250
5 5975 5818 208 Dummy - - 925 -
6 6376 6206 210 PPO 20 1089 935 1351
7 6780 6597 211 PPO 20 1123 945 1401
8 7182 6990 213 PPO 20 1156 956 1449
9 7454 7254 213 Orifice 28 - 963 -
Table 2: Gas Lift String Details for Well-1
WellTracer Gas Lift Valve Analysis

Total Casing Tubing Total Casing Tubing
Mandrel Mandrel Travel Travel Travel Gas Gas Gas
Man MD TVD Tim e Tim e Tim e Velocity Velocity Velocity
No. feet feet hh:mm:ss hh:mm:ss hh:mm:ss feet/sec feet/sec feet/sec
1 2,353 2,311 00:20:38 00:19:10 00:01:28 3.80181 2.04637 26.74227
2 3,698 3,616 00:33:06 00:30:32 00:02:34 3.72440 2.01874 24.01513
3 4,818 4,704 00:43:35 00:39:58 00:03:37 3.68476 2.00910 22.20198
4 5,542 5,400 00:50:51 00:46:44 00:04:07 3.63290 1.97646 22.43721
5 5,975 5,818 00:55:17 00:50:54 00:04:23 3.60236 1.95629 22.71677
6 6,376 6,206 00:59:25 00:54:46 00:04:39 3.57675 1.94022 22.85147
7 6,780 6,597 01:03:35 00:58:39 00:04:56 3.55443 1.92670 22.90564
8 7,182 6,990 01:07:44 01:02:31 00:05:13 3.53445 1.91469 22.94569
9 7,454 7,254 01:10:33 01:05:08 00:05:25 3.52189 1.90739 22.93563
Table 3: CO2 Tracer Traveling Time Analysis for Well-1
SPE 133268 15
WellTracer
Operator Injection Injection CO2 Casing Casing Tubing Tubing
Nam e Tim e Duration Weight Pressure Tem perature Pressure Tem perature
Time Stamp hh:mm:ss lbs psig dg.F psig dg.F
Don/Alan 09/18/2007 12:30:00 00:01:30 23.5 775 96.0 231 92.0
WellTracer Return Data

Return Return Return Calc Casing Tubing Total Mand Calcd Pct Man
# Tim e Duration Elapsed Trav Tim e Trav Tim e Trav Tim e Gas Vel MD Inj MD Error No.
hh:mm:ss hh:mm:ss hh:mm:ss hh:mm:ss hh:mm:ss feet/sec feet feet %
1 09/18/2007 13:39:36 00:08:35 01:09:36 01:10:33 01:05:08 00:05:25 3.52189 7,454 7,354 1.36% 9
Table 4: Well Tracer Injection and Return Data for Well-1
1 2032 2032 172 Dummy - - 1067 -
2 3089 3071 182 Dummy - - 1103 -
3 3858 3792 189 Dummy - - 1127 -
4 4400 4267 194 PPO 20 815 1144 928
5 4974 4728 198 PPO 20 840 1160 967
6 5548 5173 202 PPO 20 860 1175 1000
7 6148 5633 206 PPO 20 880 1191 1032
8 6721 6074 209 PPO 20 900 1206 1064
9 7326 6541 212 Orifice 48 - 1223 -
10 7865 6976 214 Orifice 36 - 1238 -
11 8405 7418 215 Orifice 28 - 1254 -
12 8945 7875 216 Orifice 32 - 1271 -

1 2,032 2,032 00:15:35 00:14:46 00:00:49 4.34652 2.29345 41.46939
2 3,089 3,078 00:24:07 00:22:42 00:01:25 4.26952 2.26799 36.34118
3 3,858 3,800 00:30:20 00:28:25 00:01:55 4.23956 2.26276 33.54782
4 4,400 4,272 00:34:45 00:32:26 00:02:19 4.22062 2.26105 31.65468
5 4,974 4,733 00:39:24 00:36:39 00:02:45 4.20812 2.26194 30.14545
6 5,548 5,174 00:44:06 00:40:52 00:03:14 4.19350 2.26264 28.59794
7 6,148 5,633 00:49:02 00:45:15 00:03:47 4.17947 2.26446 27.08370
8 6,721 6,074 00:53:45 00:49:25 00:04:20 4.16806 2.26678 25.85000
9 7,326 6,543 00:58:45 00:53:48 00:04:57 4.15660 2.26952 24.66667
10 7,865 6,973 01:04:55 00:57:42 00:07:13 4.03851 2.27181 18.16397
11 8,405 7,417 01:11:09 01:01:36 00:09:33 3.93769 2.27408 14.66841
12 8,945 7,867 01:17:26 01:05:29 00:11:57 3.85062 2.27666 12.47559
16 SPE 133268
WellTracer
Don/Alan 09/17/2007 11:43:30 00:02:20 30.0 984 98.0 250 93.0

1 09/17/2007 12:32:55 00:03:52 00:49:25 00:49:02 00:45:15 00:03:47 4.17947 6,148 6,196 0.78% 7
2 09/17/2007 12:41:59 00:05:11 00:58:29 00:58:45 00:53:48 00:04:57 4.15660 7,326 7,293 0.46% 9
1 2385 2328 165 Dummy - - 1115 -
2 3543 3324 172 Dummy - - 1152 -
3 4570 4111 177 PPO 20 950 1182 1118
4 5326 4699 181 Dummy - - - -
5 5988 5216 183 PPO 20 1035 1224 1240
6 6556 5658 185 PPO 20 1065 1241 1283
7 7091 6069 187 PPO 20 1150 1257 1345
8 7625 6470 188 PPO 20 1180 1273 1386
9 8093 6812 189 PPO 20 1205 1287 1419
10 8562 7157 190 PPO 24 1220 1301 1425
11 8972 7467 190 PPO 24 1225 1314 1446
12 9284 7712 190 Orifice 32 - 1324 -

1 2,365 2,328 00:29:07 00:27:40 00:01:27 2.70750 1.42470 27.18391
2 3,528 3,324 00:44:59 00:42:27 00:02:32 2.61430 1.38516 23.21053
3 4,565 4,111 00:59:35 00:55:52 00:03:43 2.55385 1.36187 20.47085
4 5,319 4,699 01:10:29 01:05:46 00:04:43 2.51549 1.34795 18.79505
5 5,979 5,216 01:20:13 01:14:32 00:05:41 2.48452 1.33699 17.53372
6 6,546 5,658 01:28:42 01:22:07 00:06:35 2.45998 1.32860 16.57215
7 7,080 6,069 01:36:56 01:29:29 00:07:27 2.43466 1.31868 15.83893
8 7,613 6,470 01:45:50 01:37:47 00:08:03 2.39780 1.29760 15.76190
9 8,084 6,812 01:53:47 01:45:11 00:08:36 2.36824 1.28094 15.66667
10 8,555 7,157 02:02:48 01:52:37 00:10:11 2.32220 1.26609 14.00164
11 8,964 7,467 02:10:42 01:59:07 00:11:35 2.28615 1.25423 12.89784
12 9,278 7,712 02:16:48 02:04:08 00:12:40 2.26072 1.24570 12.20789
SPE 133268 17
WellTracer
Don/Alan 09/14/2007 13:59:50 00:08:00 9.0 1,044 103.5 780 183.0
1 09/14/2007 15:09:14 00:03:04 01:09:24 01:10:29 01:05:46 00:04:43 2.51549 5,319 5,237 1.56% 4
2 09/14/2007 15:15:05 00:03:21 01:15:15 01:10:29 01:05:46 00:04:43 2.51549 5,319 5,679 6.33% 4
3 09/14/2007 15:53:52 00:07:15 01:54:02 01:53:47 01:45:11 00:08:36 2.36824 8,084 8,102 0.22% 9
1 2085 2036 180 PPO 16 765 891 905
2 3338 3075 183 PPO 12 740 923 905
3 4432 3924 185 PPO 16 845 950 1012
4 5352 4607 187 PPO 16 945 970 1143
5 6078 5149 189 PPO 16 985 987 1196
6 6616 5582 193 PPO 16 975 1000 1191
7 7187 5982 197 PPO 20 1055 1012 1208
8 7791 6382 201 PPO 20 965 1024 1147
9 8395 6782 204 PPO 20 995 1036 1193
10 8988 7182 208 PPO 20 1005 1048 1213
11 9477 7582 212 PPO 20 1015 1061 1234
12 9988 7975 215 Orifice 32 - 1073 -

1 2,085 2,036 00:19:29 00:18:35 00:00:54 3.56748 1.87013 38.61463
2 3,338 3,075 00:31:55 00:30:19 00:01:36 3.48666 1.83534 34.77583
3 4,432 3,924 00:42:50 00:40:32 00:02:18 3.44890 1.82230 32.11478
4 5,352 4,607 00:52:05 00:49:07 00:02:58 3.42549 1.81620 30.06927
5 6,078 5,149 00:59:25 00:55:52 00:03:33 3.40959 1.81312 28.53329
6 6,616 5,582 01:05:29 01:00:52 00:04:37 3.36794 1.81169 23.88560
7 7,187 5,982 01:12:01 01:06:11 00:05:50 3.32646 1.80982 20.53374
8 7,791 6,382 01:18:59 01:11:48 00:07:11 3.28804 1.80850 18.07661
9 8,395 6,782 01:26:01 01:17:24 00:08:37 3.25330 1.80774 16.23818
10 8,966 7,182 01:32:45 01:22:42 00:10:03 3.22234 1.80696 14.86925
11 9,477 7,582 01:38:49 01:27:26 00:11:23 3.19673 1.80647 13.87514
12 9,988 7,975 01:44:59 01:32:11 00:12:48 3.17132 1.80583 13.00529
18 SPE 133268
WellTracer
Don/Alan 09/16/2007 09:05:00 00:01:00 13.0 829 108.0 260 94.4

1 09/16/2007 10:04:25 00:02:37 00:59:25 00:59:25 00:55:52 00:03:33 3.40959 6,078 6,078 0.00% 5

Spe 133268 MS PDF

Uploaded by

Copyright:

Available Formats

Spe 133268 MS PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spe 133268 MS PDF

Uploaded by

Copyright:

Available Formats

SPE 133268

Successful Application of CO2 Tracer Technology for Surveillance of Gas

Copyright 2010, Society of Petroleum Engineers

Gas Lift Efficiency

An efficient gas lift operation requires the following key requirements:

Gas Lift Surveillance Process

Gas lift surveillance can be broadly classified into two types:

Alternate gas lift surveillance tool

CO2 Tracer Accuracy

CO2 Tracer Advantages

CO2 Tracer Limitations

Pilot Study Results Summary

Pilot Study Details – Well No. 1

Pilot Study – Well No. 2

Pilot Study – Well No. 3

Pilot Study – Well No. 4

IGOR Injected gas-oil ratio

GLM with gas lift

GLM with gas lift

Figure 1: Simplified Schematic of Tubing Flow GL Well

Gas Lift Manifold Pressure

Gas Lift Injection Pressure

Wellhead Flowing pressure

Gas Lift Injection rate set point

Actual GL Injection rate

Figure 2: Gas Lift Surveillance Chart

Figure 3: Schematic showing CO2 Tracer Survey Set up

CO2 Tracer Survey Results

Figure 4: CO2 Tracer Results Overall Summary

Figure 5: CO2 Tracer Survey Results for Well-1

Figure 6: CO2 Tracer Survey Results for Well-2

Figure 7: CO2 Tracer Survey Results for Well-3

Figure 8: CO2 Tracer Survey Results for Well-4

Well No. Designed Weight of CO2 Injected, Result / Comments

WellTracer Gas Lift Valve Analysis

WellTracer Return Data

WellTracer Gas Lift Valve Analysis

WellTracer Return Data

WellTracer Gas Lift Valve Analysis

WellTracer Gas Lift Valve Analysis

WellTracer Return Data

You might also like