HIGH-ENERGY GAS FRACTURING SUCCEEDS IN CENTRAL LAKE ERIE
Paul A. Druet, Sara J. O'Connor
Telesis Oil and Gas
London, Ont.
High-energy gas fracturing proved effective for stimulation of two wells producing gas from sandstone reservoirs in central Lake Erie.
A controlled propellant deflagration is used in the high-energy gas fracturing (HEGF) technique.
The two test wells had perforated gas pay with permeabilities in the range of 4-27 md.
Past stimulation programs in the area generally included hydraulic fracturing that often led to increased water production. The HEGF program was designed specifically to reduce well bore damage incurred by drilling and cementing, without the risk of fracturing out-of-zone into water.
TELESIS' PROGRAM
One of the major goals of Telesis' 1990 offshore drilling season was to test and evaluate HEGF for reducing well bore damage. The HEGF process, although not state-of-the-art, is a stimulation method that has been gaining popularity over the years, particularly when used with sandstone reservoirs.
HEGF has long been known for its usefulness in stimulating naturally fractured reservoirs and has recently become popular for perforation cleanup or skin reduction.1 2
With respect to the Thorold-Grimsby sandstone reservoirs of Lake Erie, the HEGF process was chosen to cleanup well bore damage without the fear of establishing conductivity into underlying water zones. The ease of modifying the HEGF procedures to the offshore environment and the low cost of HEGF when compared to hydraulic fracturing also supported its use.
The completion programs were designed for HEGF after perforating and a small matrix acidizing. Rigorous testing programs were setup for the first two Thorold-Grimsby successes drilled, wells 124-P-3 and 124-F-2. The two wells were both in the Clear Creek field of Central Lake Erie.
To properly quantify the benefits of HEGF, well tests were performed before and after the well was stimulated. These tests included an open hole drill stem test (DST) performed prior to casing the well.
This was followed by single rate flow and buildup tests after perforating and acidizing the well and then again after HEGF stimulation.
THEORY
High-energy gas fracturing uses a deflagration of solid propellants. The technology evolved both from the use of high-energy explosives for well bore stimulation and from the development of explosives for perforating.
Well bore stimulation using explosives has been around since the late 1800s. Recent experiments have shown that explosive stimulation creates a shock wave that is too fast. The result is compaction and disintegration of the near well bore porosity.1 3
However, use of a solid propellant will dramatically reduce the shock-wave velocity.
The chemical burning, or deflagration of the propellant creates a high-velocity gas pressure pulse that is transferred to the near well bore rock by the perforations.1 3 5 The slower action of the pressure pulse has been found to be effective in fracturing or extending existing fractures.1 3-5
The fracture mechanism is the dynamic pressure pulse. Extended channels or fractures are created when the pulse energizes existing perforations and ruptures through pore throats.1
As the pressure pulse dissipates, the fractures somewhat heal and leave a channel propped naturally by formation rock matrix.
In the oil industry, HEGF is used in a number of different ways. The most common is for reducing or removing well bore and perforation damage. Other applications include:
- Stimulation in naturally fractured reservoirs
- Prehydraulic fracturing treatment
- Preacid treatment
- Stimulation where underlying water is a problem.
Much research and development has been performed to design the proper propellant burn consistency for a variety of well bore and interval sizes. Each company that offers the HEGF service generally has strict procedures when running its HEGF tools.
Within the last few years, a pulse probe device has been developed to measure the pressure pulse created by the HEGF stimulation. The pulse probe records pressure and time.
Analysis of the pulse pressure-vs.-time profile can be used on the well site to provide a qualitative determination of formation breakdown. This is advantageous to the on site engineer because relevant data are available for quick decisions on whether or not to rerun the HEGF tool.
If the formation has not broken down, it is usually recommended to rerun the HEGF tool.
DESIGN AND RESULTS
In the Lake Erie wells, HEGF was selected to avoid the high cost of hydraulic fracturing and to reduce the risk of out-of-zone contact with the water zones in the Lower Grimsby sandstone.
A fracture height log was run prior to the HEGF treatment. The log confirmed that hydraulic fracture containment would be impossible within the Thorold-Grimsby zone on the test wells.
The HEGF program included the following steps that are recommended by Computalog, the company that provided the service:
- Perforate at 13 shots/m (4 shots/ft) or greater (test wells had 26 shots/m).
- Run cement bond log to ensure adequate bonding (i.e., large sections of free pipe are not recommended).
- Ensure a hydrostatic head of fluid of 3,450 kpag (500 psi) above the tool. Use some form of water for fluid head (i.e., oil may create a sludge after deflagration).
- Keep annulus line open.
- Allow a buffer of at least 30 m (100 ft) of air at surface to dampen the surge of fluid after deflagration.
- Run the HEGF twice over the interval.
To meet the recommended HEGF procedures for offshore, modifications were made to the existing riser system and an air compressor employed to provide the air buffer at surface.
The HEGF tool was run on wire line through a lubricator. After the first deflagration the tool was tripped out, and the pulse probe charts were checked.
The HEGF tool was run in and deflagrated a second time to complete the HEGF stimulation.
The actual pulse probe results for both test wells are provided in Fig. 1. Both wells showed pressure vs, time profiles which peaked at about 5 ms and fell off continuously afterwards. The peak at 45 ms is caused by a pressure pulse effect analogous to a water hammer effect.
These results indicate that the HEGF process achieved formation breakdown after the first deflagration in both wells. Table 1 gives the treatment steps used for each well.
TESTING PROGRAM
The testing programs incorporated pre-HEGF and post-HEGF well testing to evaluate the effectiveness of the high energy gas frac after removing well bore damage.
The pre-HEGF testing included a straddle-type DST, performed prior to running casing, and single-rate flow and buildup tests taken after the well was perforated and acidized.
The post-HEGF test was a single rate flow and buildup test run immediately after HEGF deflagration.
The first well, 124-P-3, was tested three times prior to and once after the high-energy gas frac.
Prior to the HEGF, the well was tested first in an open hole DST, then immediately after perforating, and finally after a small acid squeeze was performed to open the perforations.
The purpose of the two tests, both after perforating but before the HEGF, was to evaluate the effectiveness of the matrix acid squeeze.
Unfortunately, the test immediately after perforating appeared to be leaking by the packer on the shut-in and was deemed to be a misrun. Although leaking near the end of the buildup, the test did show extremely low permeability, indicating that few perforations were open.
The second well, 124-F-2, was tested only twice prior to the high-energy gas frac, including the DST and a test after the matrix acid squeeze. There was a difference in the completion programs of the two test wells.
The 124-P-3 well was perforated with a tubing-conveyed system at 26 shots/m (8 shots/ft), whereas the 124F-2 well was perforated with a casing gun also at 26 shots/meter. Both guns had similar charge sizes and perforation depths.
WELL 124-P-3
For Well 124-P-3, a Cyberlook log section of the Thorold-Grimsby sandstone is provided in Fig. 2. The matrix density was set to 2.65 gm/cc (sandstone) over the Thorold-Grimsby section. Based on a 10% porosity cutoff, the well encountered pay in the Thorold and Upper and Middle Grimsby sandstones as summarized in Table 2.
A DST was performed on the Upper Grimsby section 481-488 m (1,578-1,601 ft). The well was then perforated in the interval 482-487 m. Because the fluid content of the Thorold was in doubt, that zone was not perforated.
High water saturation was observed on logs in the Middle Grimsby. As such, a DST was run only over 5.5 m of the Upper Grimsby. This was the only section perforated.
The 5.5 m of net pay in the Upper Grimsby averaged about 16% porosity with several meters of pay near the top showing porosities of over 18%. A water contact was observed to exist at about 489.5 m in the Middle Grimsby, which was about 2.5 m below the bottom perforation.
The water zone appeared to be separated from the gas productive sand by about 1.5 m of shale.
Four well tests were performed on the 124-P-3 well, including one open hole DST and three buildup tests run after the well was cased. As mentioned earlier, the test immediately after perforating was a misrun.
The log-log and semilog plots for the three valid tests are shown in Fig. 3.
Progressive analyses of all four tests provided extremely useful results by showing how the HEGF treatment reduced well bore damage, induced by cementing and completion, to below that calculated from the original DST. All reservoir parameters were calculated using a semilog analysis.
The character of the reservoir as determined from the log (_P) vs. log (_Te) Plots exhibited remarkable similarity between plots implying that we were accessing a homogeneous reservoir modeled on storage and skin, each time we flowed the well.
The early time data from the DST was nonexistent and did not show any well bore storage, which is to be expected for a high-permeability reservoir with a shutoff valve downhole.
On the remaining three tests, performed with surface shut-in and bottom hole pressure recorders, the two poorer tests showed well bore storage in the early time, The post-HEGF test did not show any early time effects. This again is due to the high gas flow capacity, kh.
The kh, skin factor s, and absolute open flow potential AOF, changed chronologically through the completion as shown in Table 3.
It is evident from Table 3 that the high-energy gas frac increased by over four times the deliverability after acidizing. This increase is related directly to an increase in kh and a decrease in the skin factor.
The kh increase from 305 md-ft (pre-HEGF) to 435 md-ft (post-HEGF) most likely is the result of the HEGF treatment opening up plugged perforations. The decrease in skin factor from +16.4 to +2.0 probably was caused by removing near well bore damage and/or effectively increasing near well bore permeability by microfracturing the adjacent rock.
Noteworthy is that the flow capacity calculated from the DST is about 26% higher than the kh calculated after the HEGF. Alternatively, the skin factor from the DSTs is much higher than the skin factor calculated after the HEGF (+16.4 from DST, +2.0 from HEGF).
The combination of higher permeability and higher skin factor generates a lower deliverability for the DST production compared to the post-HEGF production. The higher deliverability observed after the HEGF is probably due to reducing well bore damage.
The fact that the DST kh is higher than the post-HEGF frac kh is probably a result of data quality relating to the short production time of the DST.
It is evident that the HEGF was successful at reducing skin damage, induced both by drilling operations and by cementing and completion operations. This resulted in maximized cased well deliverability.
WELL 124-F-2
For Well 124-F-2, a Cyberlook log section of the Thorold-Grimsby sandstone is shown in Fig. 4. The matrix density was set to 2.65 gm/cc (sandstone) over the Thorold-Grimsby. Based on a 10% porosity cutoff, the well encountered pay in the Thorold and Upper and Middle Grimsby as summarized in Table 2.
A DST was run over the entire Thorold-Grimsby section from 466 to 481 m. The Middle Grimsby was perforated from 473 to 480 m.
Because the content of the Thorold was in doubt it was not perforated. The 0.5 m of pay in the Upper Grimsby also was not perforated because it was considered to have low porosity.
The net pay in this well showed an average porosity of about 12%, well below that in the 124-P-3 well.
A water contact was not obvious in this well, however, an increase in water saturation was observed at 482 m in the Lower Grimsby that did correspond to known water contacts in the area.
Three well tests were performed on the 124-F-2 well, including one open hole DST and two buildup tests after the well was cased.
The log-log and semilog plots for the three tests are shown in Fig. 5. In all cases, reservoir parameters were calculated from the semilog plots.
The kh, s, and AOF are tabulated in Table 3.
The results from the high-energy gas frac were not as dramatic in Well 124-P-3. However, evident from Table 3 is that the HEGF did increase the cased well deliverability at pipeline pressure by about 75%.
The increase in deliverability appears to be related directly to an increase in kh and a decrease in S. The kh increase from 44 md-ft (pre-HEGF) to 79 md-ft (post-HEGF) probably resulted from the gas frac cleaning up plugged perforations.
The decrease in s from - 0.1 to - 0.7 is most likely a result of increasing near well bore permeability by microfracturing the adjacent rock. The HEGF was successful at reducing skin damage induced by cementing and completion operations.
Notable is that the kh from the DST of 114 md-ft is about 44% higher than the kh calculated after the HEGF. The deliverability at pipeline pressure is also greater than the post-HEGF deliverability.
The DST also shows a higher extrapolated pressure, p', than calculated from the post-HEGF test. These data are indicating that not all of the productive pay was perforated in the Thorold-Grimsby section as defined by the DST.
Evident from the attached log sections is that the 1 m of Thorold pay and 0.5 m of Upper Grimsby pay were not perforated. This pay may have supplied the 44% additional kh shown on the DST. This pay may also be responsible for the higher pressure and should be considered in any future workovers.
Both the semilog and log-log plots show late-time effects on all tests performed on the 124-F-2 well. These effects indicate that a boundary, reasonably close to the well bore, is influencing the pressure transients.
The boundary may be a pinchout, a fault, or even a major reduction in permeability in one or more directions. From the semilog plots, for both the DST and post-HEGF tests, the approximate location of the boundary was about 46-47 ft away from the well bore.
Note that the DST with higher kh shows the boundary to be approximately the same distance from the well bore as that calculated from the post-HEGF test. This obviously increases the accuracy of the interpretations.
The character of the log-log plot shows no early time effects for the DST and some early time effects for both cased hole tests.
These early time effects, shown on the cased hole tests, appear to be well bore storage followed immediately by transition effects and late-time effects.
It is unlikely that any of the tests had a good middle-time region because of the early onset of boundary effects. This sort of test data usually makes accurate predictions of well bore potential difficult. However, in this case the log (_P) vs. log (_Te) plots could be used to better quantify the reservoir parameters and provide more reliable data.
Disregarding the fact that some pay may have been missed in Well 124-F-2, the high energy gas frac appears to have been successful at increasing well deliverability and reducing the skin factor.
The final test, performed after the HEGF, shows a well in a relatively nondamaged state.
This is all one could hope for with this type of completion method. Any further reduction in skin-factor could only be achieved with hydraulic fracturing.
REFERENCES
- Schmidt, R.A., "In Situ Evaluation of Several Tailored-Pulse Well-Shooting Concepts," Paper No. 8934, Proceedings Symposium on Unconventional Gas Recovery, 1980, pp. 105-10.
- Schatz, J.F., and Zeigler, B.J., "Laboratory, Computer Modeling, and Field Studies of the Pulse Fracturing Process," Paper No. 18866, Proceedings SPE Symposium Production Operations, 1989, pp. 375-79.
- Swift, R.P., and Kusobov, A.S., "Multiple Fracturing of Boreholes by Using Tailored-Pulse Loading," SPE Journal, Vol. 23, 1982, pp. 923-32.
- Cuderman, J.F., and Northrop, D.A., "A Propellant-Based Technology for multiple Fracturing Wellbores to Enhance Gas Recovery: Application and Results in Devonian Shale," Paper No. 12838, Proceedings SPE/DOE/GRI Symposium on Unconventional Gas Recovery, 1984, pp. 77-82.
- Simmons, A.L., "Stressfrac-A New Dimension in Well Treatment," Paper No. 11827, Proceedings SPE Rocky Mountain Regional Meeting, pp. 219-22.
Copyright 1991 Oil & Gas Journal. All Rights Reserved.