Caliper-behind-casing from pulsed neutron apparatus

Abstract

A caliper-behind casing apparatus and method determines a location and size of a washout (i.e., a void) behind casing in a borehole for a salt cavern used for storing hydrocarbons. The cavern and tubing in the casing are filled with brine. Gaseous nitrogen is used to fill an annulus between the tubing and the casing above and below a casing shoe to obtain image responses from operating a pulsed neutron logging tool in the tubing along the borehole. Analysis of burst ratios of near and far detectors on the pulsed neutron logging tool from these passes is used to detect and estimate a void or washout in the formation behind the casing.

Claims

1. A method of logging a borehole in a formation, the borehole having casing installed therein to a casing shoe, the borehole having tubing installed in the casing and extending beyond the casing shoe, the method comprising: filling the tubing and an annulus between the tubing and the casing with a liquid; injecting a gas into the annulus between the tubing and the casing to a first point above the casing shoe of the casing in the borehole; obtaining a first image response of a portion of the borehole with the annulus filled with the injected gas above the casing shoe by operating a logging tool in the tubing along the borehole to image the borehole in the formation behind the casing; injecting the gas into the annulus between the tubing and the casing to a second point below the casing shoe of the casing in the borehole; obtaining a second image response of the portion of the borehole with the annulus filled with the injected gas below the casing shoe by operating the logging tool in the borehole to image the borehole in the formation behind the casing; comparing the second image response to the first image response; and detecting a void between the casing and the borehole in the formation behind the casing based on the comparison.

2. The method of claim 1, wherein filling with the liquid comprises filing with brine as the liquid; and wherein injecting the gas comprise injecting gaseous nitrogen as the gas.

3. The method of claim 1, comprising obtaining, before the first image response, an initial image response of the portion of the borehole with the tubing and the annulus filled with the liquid by operating the logging tool in the tubing along the borehole to image the borehole in the formation behind the casing.

4. The method of claim 3, further comprising calibrating the first image response based on the initial image response.

5. The method of claim 1, further comprising estimating a volume of the detected void between the casing and the borehole in the formation behind the casing.

6. The method of claim 5, further comprising estimating an amount of resin to fill the estimated volume of the detected void.

7. The method of claim 6, further comprising pumping the estimated amount of resin to fill the detected void.

8. The method of claim 1, wherein obtaining the second image response of the portion of the borehole with the annulus filled with the injected gas below the casing shoe by operating the logging tool in the tubing along the borehole to image the borehole in the formation behind the casing comprises: making a first pass of the portion of the borehole with the logging tool; making at least one second pass of the portion of the borehole with the logging tool after a period of time; and comparing the second image response of the first pass to that of the at least one second pass.

9. The method of claim 8, further comprising determining, from the comparison, that an interface between the liquid and the injected gas has not shifted.

10. The method of claim 8, further comprising verifying, from the comparison, the second image response.

11. The method of claim 1, wherein obtaining the first and second image responses by operating the logging tool in the tubing along the borehole to image the borehole in the formation behind the casing comprises operating a pulse neutron logging tool as the logging tool.

12. The method of claim 11, wherein operating the pulse neutron logging tool comprises: counting first bursts as a function of depth at a first detector of the pulsed neutron logging tool being a first distance the neutron source; counting second bursts as a function of depth at a second detector of the pulsed neutron logging tool being a second distance from the neutron source, the second distance greater than the first distance; and calculating a burst ratio of the first burst count relative to the second burst count as a function of depth.

13. The method of claim 12, wherein comparing the second image response to the first image response comprises subtracting the calculated burst ratio of the first image response as a function of depth from the calculated burst ratio of the second image response as a function of depth; and extrapolating caliper of the borehole as a function of depth from the difference based at least on a porosity of the formation in which the borehole is disposed.

14. The method of claim 1, wherein injecting the gas into the annulus between the tubing and the casing to the first point above the casing shoe of the casing in the borehole comprises monitoring pressure of the injected gas for a predetermined amount of time; and calculating a leak rate of the injected gas as a function of the monitored pressure relative to the predetermined amount of time.

15. The method of claim 1, wherein injecting the gas into the annulus between the tubing and the casing to the second point below the casing shoe of the casing in the borehole comprises monitoring pressure of the injected gas for a predetermined amount of time; and calculating a leak rate of the injected gas as a function of the monitored pressure relative to the predetermined amount of time.

16. The method of claim 1, wherein detecting the void in the formation behind the casing based on the comparison comprises basing the detection on a porosity of the formation.

17. A method of logging a borehole in a formation, the borehole having casing installed therein to a casing shoe, the borehole having tubing installed in the casing and extending beyond the casing shoe, the method comprising: filling the tubing and an annulus between the tubing and the casing with a liquid; obtaining a first image response of a portion of the borehole with the annulus filled with the liquid by operating a logging tool in the tubing along the borehole; injecting a gas into the annulus between the tubing and the casing to a point below the casing shoe of the casing in the borehole by injecting the gas directly at an open hole portion of the borehole below the casing shoe, thereby allowing the gas to enter a potential void in the borehole behind the casing; removing any of the injected gas from the annulus while the gas is allowed to remain in the potential void; obtaining a second image response of the portion of the borehole while the annulus is filled with the fluid and the potential void is filled with the gas by operating the logging tool in the tubing along the borehole; comparing the second image response to the first image response; and detecting the potential void in the formation behind the casing based on the comparison.

18. The method of claim 14, wherein calculating the leak rate comprises predicting a washout in the formation behind the casing based on the calculated leak rate; and wherein detecting the void comprises estimating a volume of the washout in the formation behind the casing based on the comparison of the second image response to the first image response.

19. A method of logging a borehole in a formation, the borehole having casing installed therein to a casing shoe, the borehole having tubing installed in the casing and extending beyond the casing shoe, the method comprising: filling the tubing and an annulus between the tubing and the casing with a liquid; injecting a gas into the annulus between the tubing and the casing to a first point above the casing shoe of the casing in the borehole; obtaining a first image response of a portion of the borehole with the annulus filled with the injected gas above the casing shoe by operating a logging tool in the tubing along the borehole; injecting the gas into the annulus between the tubing and the casing to a second point below the casing shoe of the casing in the borehole; obtaining a second image response of the portion of the borehole with the annulus filled with the injected gas below the casing shoe by operating the logging tool in the tubing along the borehole; comparing the second image response to the first image response; detecting a void in the formation behind the casing based on the comparison; estimating a volume of the detected void in the formation; and estimating an amount of resin to fill the estimated volume of the detected void.

20. The method of claim 19, further comprising pumping the estimated amount of resin to fill the detected void.

21. A method of logging a borehole in a formation, the borehole having casing installed therein to a casing shoe, the borehole having tubing installed in the casing and extending beyond the casing shoe, the method comprising: filling the tubing and an annulus between the tubing and the casing with a liquid; obtaining a first image response of a portion of the borehole with the annulus filled with the liquid by operating a logging tool in the tubing along the borehole; injecting a gas into the annulus between the tubing and the casing to a point below the casing shoe of the casing in the borehole, thereby allowing the gas to enter a potential void in the borehole behind the casing; extracting the injected gas from the annulus by displacing the gas in the annulus with the liquid supplied through the tubing to remove any of the injected gas from the annulus while the gas is allowed to remain in the potential void; obtaining a second image response of the portion of the borehole while the annulus is filled with the fluid and the potential void is filled with the gas by operating the logging tool in the tubing along the borehole; comparing the second image response to the first image response; and detecting the potential void in the formation behind the casing based on the comparison.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 illustrates a system for detecting a void or a washout in a cased borehole extending from surface to a subterranean salt cavern.

(2) FIGS. 2A-2B illustrate schematic cross-sections of a logging tool disposed in tubing and casing of a borehole for modeling a void or a washout in a surrounding formation.

(3) FIG. 3 illustrates several graphs where MCNP radiation transport simulations are used to model the GB15 burst ratio (burst counts in the nearest detector, DP, ratioed to the farthest detector, DL) as a function of porosity for different fluid types.

(4) FIG. 4 illustrates a graph showing the numerical correlation between radii of washout behind casing verses the simulated ratio of two GB15 measurements (SW in annular region vs N.sub.2 gas in annular region).

(5) FIG. 5 illustrates a process for detecting a void or a washout behind casing with a logging tool deployed in tubing disposed in casing.

(6) FIG. 6A illustrates an image response of the logging tool in a base pass of the detection process in which the tubing is filled with brine and in which gas is disposed in the annulus between the tubing and casing.

(7) FIG. 6B illustrates an image response of the logging tool in a subsequent pass of the detection process in which the tubing is filled with brine, gas is disposed in the annulus, and gas has potentially seeped behind the casing.

(8) FIG. 6C illustrates an image response of the logging tool after processing to determine radius of a washout behind the casing.

(9) FIG. 7 illustrates another process for detecting a void or a washout behind casing with a logging tool deployed in tubing disposed in casing.

DETAILED DESCRIPTION OF THE DISCLOSURE

(10) A caliper-behind casing apparatus and method disclosed herein determines a location and size of a washout (i.e., a void) behind casing in a borehole. The apparatus and method can be generally applied to any cased hole environment. A particular application of the disclosed subject matter is in salt caverns used for storing hydrocarbons.

(11) A. Well Arrangement for Salt Cavern Storage

(12) FIG. 1 illustrates a system for detecting a void or a washout in a cased borehole 12 extending from surface to a subterranean formation 10. As shown in schematic form in FIG. 1, the cased borehole 12 penetrates the formation, which here includes a salt formation. The borehole 12 typically would be formed through a drilling operation and may be several hundred or thousand meters in length. The borehole 12 extends from the surface to the salt cavern 10, which has been created to store hydrocarbons. A typical nominal diameter (caliper) of the borehole 12 may be 12-in, although other diameters can be used. The borehole 12 is nominally circular in cross-section, but this is unlikely to be consistent from place to place along the borehole 12.

(13) In the configuration for the salt cavern 10, the borehole 12 extending from surface to the cavern 10 has a casing 20 cemented to a depth therein. A tubing string 30 extends from a wellhead 28 at surface and is disposed in the salt cavern 10. As noted, the salt cavern 10 may be created deep in the salt formation using solution mining in which fresh water is injected into the borehole 12 drilled into the salt formation. The cavern 10 can be used to store natural gas, in which case gas can be pressurized by compressors 40 and injected into the cavern 10. The built-up pressure can then be used to deliver the gas from the cavern 10 when needed.

(14) The cavern 10 can also be used to store liquid hydrocarbons by using a liquid (namely, brine) as a displacement fluid. To remove the hydrocarbons from the cavern 10, the brine is pumped down the tubing string 30, which extends into the cavern 10 near the bottom. In the cavern 10, the brine displaces the hydrocarbons, which can then be drawn through the annulus 22 between the casing 20 and the tubing 30 toward the top of the cavern 10. To inject the hydrocarbons into the cavern 10, a reverse process is used with the brine being removed from the bottom of the cavern 10 and with the hydrocarbon injected through the annulus 22 between the casing 20 and the tubing

(15) A conductor pipe 24 is typically used toward the surface, and the casing 20 extends down the borehole 12 and ends with a casing shoe 26 toward the top of the salt cavern 10. The conductor pipe 24 may be 16-in. The bit sized used to drill the borehole 12 may have been 14-¾-in, and the casing 20 cemented in the borehole 12 can be 11-¾-in. Under normal circumstances, the casing 20 would be of essentially consistent cross-section along its length, and in a typical situation, the casing 20 is made of steel or another rigid metal alloy. In some implementations, an inner liner (not shown) may be cemented inside the casing 20 at least partially along the depth of the casing 20. For example, the cased liner (not shown) may extend just short by about 20-ft (6-m) or so above the casing shoe 26 on the end of the cemented casing 20.

(16) For its part, the tubing 30 extends through the casing 20 and any internal cased liner 24 and extends well into the cavern 10, typically further than depicted here in FIG. 1. As an example, the tubing 30 may have a diameter of 8-⅝-in and may typically be decentralized in the casing 20, contrary to the way it is depicted here in FIG. 1.

(17) As is typical, the casing 20, the liner 24, and the uppermost part of the borehole 12 near the wellhead 28 are sealed with cement. This can be achieved using a series of nested casing sections cemented in the formation near the wellhead 28 at the surface.

(18) When working normally, the borehole 12 would appear approximately as depicted in FIG. 1. However, as described herein various, washouts and leaks can occur. This is represented in FIG. 1 by an enlarged caliper borehole section or void 15, delineated by dotted lines, which can result from a washout of the formation outside the casing 20.

(19) 1. Logging Arrangement with Logging Tool

(20) Logging the caliper of the borehole 12 in such situations is desirable (i) to ensure the integrity of the salt cavern 10, the borehole 12, and the casing 20; and (ii) to establish the volume of any void 15 lying outside the casing 20 so that a determination can be made of the cost for remediating the borehole 12. Unfortunately, any conventional techniques for logging to determine the caliper of the borehole 12 behind the casing 20 are complicated in the present arrangement due to the presence of the additional elements of the tubing 30, the brine in the tubing 30, the gas in the annulus 22 between the casing 20 and the tubing 30, the casing 20 itself, and any cement fixing the casing 20 in the borehole 12.

(21) To log the caliper of the borehole 12 according to the present system, FIG. 1 shows a logging tool 50 suspended on a conveyance 55, such as wireline, inside the tubing 30 extending through the casing 20 (and liner 24 if present). As is well known, the wireline 55 permits the transmission at least of electrical power and commands for the logging tool 50 from surface equipment 60 to the logging tool 50 and permits the transmission of electrical signals representing log data from the logging tool 50 to the surface equipment 60.

(22) In a typical deployment of the logging tool 50, the wireline 55 is paid out from a surface drum (not shown) as the logging tool 50 is conveyed—e.g., under gravity or by being pumped as is known in the art, to a depth at which logging is to commence. The logging tool 50 is then withdrawn back towards the surface by winding the wireline 55 back on to its dispensing drum. During this uphole movement, the logging tool 50 records information about the borehole 12.

(23) 2. Type of Logging Tool

(24) Various types of logging tools 50 could be used. For example, the logging tool 50 can be a pulse neutron logging tool having a neutron source 52 and a number of gamma ray detectors 54. Alternatively, the logging tool 50 can be a density logging tool having a source 52 of gamma radiation and a number of gamma ray detectors 54. Although shown embodied in a wireline logging tool, the source 52 and detectors 54 can also be embodied in other borehole instruments. These instruments include pump-down (“memory”) instruments conveyed by fluid flow, instruments conveyed by coiled tubing, instruments conveyed by a drill string, and instruments conveyed by a “slick line.”

(25) As shown in FIG. 1, for example, the logging tool 50 can be a multipurpose pulsed neutron logging tool 50, including a pulsed neutron generator or source 52, an array of detectors 54, an instrument assembly 56, and a telemetry assembly 58. In the arrangement shown, four of the detectors 54 are referenced as a proximal detector (DP), a near detector (DN), a far detector (DF), and a long detector (DL). These four detectors 54 can include spectroscopic Lanthanum Bromide (LaBr.sub.3) detectors having a LaBr.sub.3 detector crystal and a digital spectrometer for filtering and pulse inspection. These four detectors 54 are disposed at increasing axial spacings from the neutron source 52, as their names imply. An additional detector 54 can be a fast neutron detector (DFS) disposed between the near detector (DN) and the far detector (DF).

(26) The instrument assembly 56 houses control circuits and power circuits to operate and control the elements of the tool 50. The telemetry assembly 58 is operationally connected to the instrument assembly 56. A suitable connector connects the logging tool 50 to a lower end of a preferably multiconductor logging cable 55, and the upper end of the logging cable 55 terminates at a draw works (not shown).

(27) Detector response data is telemetered from the tool 50 to surface where the data is received by an uphole telemetry unit (not shown) preferably disposed within the surface equipment 60. The data is processed in a surface processor (not shown) within the surface equipment 60 to yield logs 62 of one or more parameters of interest. Alternately, data can be partially or completely processed in a downhole processor within the instrument assembly 56 and telemetered via the telemetry assembly 58 to the surface equipment 60. Control parameters can also be telemetered from the surface equipment 60 to the tool 50 via the telemetry system and wireline cable 55.

(28) 3. Mechanical Integrity Test of Well Arrangement

(29) On a regular basis, a Mechanical Integrity Test (MIT) is performed to test whether the salt cavern 10 and the casing 20 (and the liner 24) are robust and not leaking. For the test, the tubing string 30, the cemented casing string 20, and the salt cavern 10 are filled with a liquid (most preferably brine). At surface, pumping or compressing equipment 40 is operable to inject gas, such as gaseous nitrogen (N.sub.2), from a source 42, through the wellhead 28, and into the annulus 22 between the casing 20 and the tubing string 30. Instead of nitrogen, the MIT can be performed using a liquid, such as oil, condensate, diesel, etc. However, gaseous nitrogen (N.sub.2) may be preferred according to the MCNP model of the present disclosure because the tool sensitivity to the gaseous nitrogen versus water is good. Overall, any fluid (gas or liquid) can be used according to the present techniques as long as there is a sufficient difference in density between the two fluids.

(30) For the test, the logging tool 50 is deployed into the tubing string 30. Rather than just determining the interface between the brine and the nitrogen downhole for the purposes of performing the Mechanical Integrity Test, the neutron logging tool 50 of the present disclosure is operated to make specific caliper measurements as discussed in more detail below. In turn, the surface processor 60 obtains readings from the logging tool 50 through storage, telemetry, or the like and produces output in the form of logs 62 or the like to detect and estimate the volume of any void or washout 15 behind the casing 20.

(31) In general, the logging tool 50 is operated in the tubing 30, and the source 52 produces bursts of high-energy neutrons. After emission from the source 52 and passage through the surrounding environment between the source 52 and the detectors 54, the detected radiation at the detectors 54 implies information about the surrounding environment from which the caliper can be determined behind the casing 20.

(32) As is well known, the detectors 54 produces electrical signals that can be transmitted and/or processed as required. In the wireline-based embodiment of FIG. 1, the output signals of the detectors 54 can be transmitted via the wireline 55 to the surface equipment 60 represented schematically in the figure as a desktop computer. Other forms of processing device can be used within the scope of the present disclosure, including processors located far from the surface location and processors that are carried on-board on the logging tool 50.

(33) In more detail, the source 52 emits neutrons in a plurality of bursts into the surrounding environment at an energy level sufficient to induce inelastic scatting gamma rays. Each burst occurs for a defined duration and is followed by a wait time having another defined duration. The neutrons scatter and are eventually captured by atomic nuclei in the environment at a rate proportional to the population of the neutrons. When capture occurs, gamma rays are produced, some of which can be detected by the detectors 54. Chlorine is a strong neutron absorber so the response of the logging tool 50 may be based primarily by the chlorine present (as sodium chloride) in the surrounding environment.

(34) During operation, the detectors 54 count the resulting gamma rays produced by interactions of the emitted neutrons with the surrounding environment, such as the formation, brine, etc. The number of gamma rays detected decays as the population of the neutrons eventually decays and the neutrons are absorbed by the surrounding environment. In this way, the surface equipment 60 and/or the logging tool 50 can determine the rate of decay of thermal neutrons in the environment based on the measured decay in the gamma rays produced as these neutrons are absorbed.

(35) The gamma rays are detected at the spaced nearest and farthest (e.g., DP, DL) detectors 54 during each of the bursts and during each of the wait times. Therefore, a “burst ratio” of the counts of gamma rays detected by the nearest (e.g., DP) detector 54 relative to the counts of gamma rays detected by the farthest (e.g., DL) detector 54 during the bursts is determined. Likewise, a “capture ratio” of the counts of gamma rays detected by the nearest (e.g., DP) detector 54 relative to the counts of gamma rays detected by the farthest (e.g., DL) detector 54 during the wait times can be determined.

(36) Using the detector responses, for example, the surface equipment 60 can obtain a nearest-to-farthest burst ratio during the neutron burst and can obtain a nearest-to-farthest capture ratio during the wait times. Both of these values are derived from the count rates of the nearest and farthest detectors (e.g., DP and DL) 54. In this way and as noted previously, the ratio of the nearest and farthest count rates during the neutron burst refers to the burst ratio, whereas the ratio of the nearest and farthest count rates during a time interval after the burst refers to the capture ratio. The burst ratio contains inelastic gamma ray events induced by the fast neutrons.

(37) As discussed in more detail below, the processing equipment 60 uses these detector responses to detect the presence of any washouts or voids 15 behind the casing 20. Additionally, the processing equipment 60 can estimate the volume of the washout 15 so that a resin can be pumped to fill the washout 15. Use of a resin is the most common approach and may be the most effective solution. However, other options can be used to such as squeezing cement or using some other type of sealant.

(38) Because the logging tool 50 must be deployed in the tubing 30 that extends in the casing 20 (and the cemented liner 24 if present), conventional techniques of determining the caliper behind the casing 20 cannot be used. Instead, the processing equipment 60 of the present disclosure uses a modelling and analysis technique and a logging process that detects voids or washouts in an arrangement as shown here in which tubing 30 that extends in the casing 20 cemented in the borehole 12 of the formation.

(39) Given an understanding of the well arrangement and components of the system for logging the well arrangement, discussion turns to details for the modelling and analysis technique to detect voids or washouts.

(40) B. Modelling Washout Behind Casing

(41) FIGS. 2A-2B illustrates in schematic cross-sections the logging tool 50 disposed in the tubing 30, which in turn is disposed in the cemented liner 24 and the casing 20 of a borehole. In FIGS. 2A-2B, the tubing string 30 is shown decentralized from the cemented liner 24 and the casing 20, and the logging tool 50 is shown deployed in the tubing 30, which is filled with brine. The annulus 22 between the tubing 30 and the liner 24/casing 20 is filled in the modelling alternatively with brine and then gas. Various hole sizes 17 indicative of possible void or washout diameters are depicted outside the cemented casing 20.

(42) As an example, the tubing 30 may have 8-⅝-in diameter, and the casing 20 may have a 11-¾-in diameter. The various hole sizes 17 for modeling may have diameters of about 12, 15, 18, 21, 24, 27-in, etc. representative of increasing amounts of standoff from a potential void or washout behind the casing 20 that could be filled with gas (i.e., the gaseous nitrogen) during analysis.

(43) As noted above, the arrangement may not pass mechanical integrity testing due to leaking, and operators may suspect a washout behind the casing 20. The remedy would be to pump a resin behind the casing 20 to seal off the leak. Preferably, operators can obtain an estimate of the washout to determine the volume of resin required.

(44) To build the model to detect the potential washout and estimate its volume, porosity (percentage) is graphed relative to a burst ratio (based on proximal to farthest detectors) for different modelled possible washout diameters 12, 15, 18, 21, 24, 27-in. etc. in the formation surrounding the casing 30. For example, FIG. 3 illustrates several graphs 70A-70E modelling burst ratio measurements that are predicted by operating the pulsed neutron logging tool (50) in a borehole in a surrounding formation. The borehole in each graph 70A-70E has a modelled diameter (e.g., 15, 18, 21, 24, 27-in., etc.). The formation in each graph 70A-70E is also filled with different fluid types, such as fresh water, salt water, different type of gases, and different types of hydrocarbons, as depicted by the different curve lines 72. For example, graph 70A shows curves lines 72 of the various fluid types correlating a predicted burst ratio of the nearest detector, DP, to the farthest detector, DL, which is called the GB15 Ratio, relative to porosity of a surrounding formation having a 15-in. diameter borehole. In this graph, the annular region between casing and borehole diameter is filled with nitrogen gas (N2),

(45) Of interest in this graph is the level of porosity 74 at zero depicted in the graphs 70A-70E, because zero porosity would be indicative of the porosity of a salt formation in the implementation of FIG. 1. As shown in graph 70A, for example, the value 74 at zero porosity of the salt formation with a 15-in diameter borehole is predicted to produce a burst ratio (GB15 Ratio) from the pulse neutron investigation of “8.51.” As shown in the other graphs 70B-7E, the values 74 at zero porosity of the salt formation with increasing diameter boreholes of 18, 21, 24, 27-in would produce decreasing burst ratios from the pulse neutron investigation. This is shown in Table I in the column labeled “GB15 Ratio with N2 filled washout”. If the annular region is instead filled with saltwater (SW), the GB15 ratios trend in the opposite direction. This is shown in Table I in the column labeled “GB15 Ratio with SW filled Washout.” The final column of Table I is the ratio of the previous two columns, and shows a strong correlation to the radial extent of the washout. This is shown graphically in FIG. 4.

(46) TABLE I: GB15 Ratios vs Modelled Diameter

(47) Modelled Modelled Radial GB15 Ratio with GB15 Ratio GB15 SW/ Diameter Washout Extent of SW Filled with N2 Filled GB15 N.sub.2 Washout Washout Washout 12-in. 0.25-in. 0.125-in. 19.03 16.35 1.16 15-in. 3.25-in. 1.625-in. 30.28 8.51 3.56 18-in. 6.25-in. 3.125-in. 33.37 5.21 6.41 21-in. 9.25-in. 4.625-in. 34.35 3.85 8.92 24-in. 12.25-in. 6.125-in. 34.67 3.18 10.90 27-in. 15.25-in. 7.625-in. 34.68 2.83 12.25 30-in. 18.25-in 9.125-in. 34.68 2.63 13.19 33-in. 21.25-in 10.625-in. 34.68 2.52 13.76

(48) FIG. 4 provides an example of the mathematical correlation between the radial extent of the washout (y) vs the ratio of GB15 for the SW and N2 gas filled cases (GB15.sub.SW/GB15.sub.N2) (x). The equation is: y=0.00085359x.sup.4−0.0196401x.sup.3+0.15946566x.sup.2+0.06112931x−0.02663991.

(49) As discussed above with reference to FIGS. 3 and 4, the borehole 12 for the salt cavern 10 extends through a salt formation so that analysis has focused on the data and modeling related to detecting a washout 15 in the salt formation surrounding the casing 20. Thus, the data and modeling has focused on the surrounding salt formation as having about zero percent porosity. As will be appreciated, the borehole 12 may be disposed in other types of materials, such as sandstone, limestone, dolomite, etc., and the analysis and modeling can be tailored to detecting a washout 15 in the formation of these other materials given their porosity, as generally indicated in the graphs of FIG. 3. This would produce comparable graphed correlations of the normalized burst ratio relative to predicted washout as in FIG. 4, but tailored to a formation of other expected or know porosity.

(50) C. Process of Detecting/Estimating Washout Behind Casing

(51) Given the modelled analysis of FIGS. 3 and 4, FIG. 5 illustrates a process 100 for logging the borehole 12 and detecting a void or washout 15 behind casing 20 with the logging tool 50 deployed in the tubing 30. (For purposes of discussion, reference will be made concurrently to the well arrangement depicted in FIG. 1.)

(52) According to the process 100, the salt cavern 10, the casing 20, and the tubing 30 are filled with brine, such as when conducting a mechanical integrity test as noted above (Block 102). For example, the pump 40 of the surface equipment pumps brine from the surface pool 42 down the tubing 30. Any residual hydrocarbon can be displaced and removed from the wellhead 28 for storage or transport.

(53) 1. Initial Logging Pass With Brine in Tubing and Annulus

(54) The pulsed neutron logging tool 50 is then run using wireline or other conveyance 55 in a first base pass down the tubing 30 (Block 104). Readings from this first base pass are telemetered to the surface processing equipment 60, which uses the information to generate a base image response (i.e., base logs). The base logs of the base image response can include: a gamma ray log; a wireline tension log; a casing collar log; burst counts for long, far, near and proximal detectors; capture counts for long, far, near and proximal detectors; burst ratios; capture ratios; thermal capture sigma (Σ) values; and the like as a function of depth in the borehole 12. Some of the logged information may not be used for the purposes disclosed herein, but may be helpful in other analysis.

(55) 2. Second Logging Pass with Brine in Tubing and Gas in Annulus Above Shoe

(56) After this first base pass, gas (e.g., gaseous nitrogen N.sub.2) is then injected from a source 44 into the annulus 22 between the tubing 30 and the casing 20 (Block 106). The gas is injected until the interface with the brine is just above the casing shoe 26 of the casing 20. In this way, the injected gas cannot leak behind the casing 20 and remains in the annulus 22 between the casing 20 and the tubing 30.

(57) This step may further account for portion of a mechanical integrity test. For instance, the pressure from the injected nitrogen is held for a predetermined amount of time, which can be governed by a given formula for a minimum detectable leak rate. In this way, a certain loss of pressure over the predetermined amount of time can indicate based on calculation that a leak rate exceeds a defined limit, which can be used to access the mechanical integrity of the casing 20.

(58) Operators make a second base pass operating the pulse neutron logging tool 50 in the tubing 30 to obtain measurements (Block 108). For this pass, the annular space 22 between the tubing 30 and the casing 20 is filled with the gas, while the tubing 30 remains filled with brine. Readings from this second base pass are telemetered to the surface processing equipment 60, which uses the information to generate a second base image response (i.e., base logs). Again, the base logs of the second base image response can include: a gamma ray log; a wireline tension log; a casing collar log; burst counts for long, far, near and proximal detectors; capture counts for long, far, near and proximal detectors; burst ratios; capture ratios; thermal capture sigma (Σ) values; and the like as a function of depth in the tubing 30, the casing 20, and the borehole 12.

(59) Base Image Response (FIG. 6A)

(60) For example, FIG. 6A illustrates a base image response 200A of logs for the logging tool 50 in this second base pass of the detection process (100) where the tubing 30 is filled with brine and where the gas is disposed in the annulus 22 between the tubing 30 and casing 20. For reference, the well schematic 210 is shown in the image response 200A as a function of depth and depicts the casing 212 with the tubing 213 extending beyond the casing shoe 214.

(61) Exemplary log data is plotted in the form of various graphical logs 220, 230, 240, and 250 in the base image response 200A. The logs show log signals as a function of depth generated in logging the exemplary borehole 12 illustrated schematically in FIG. 1.

(62) Burst logs 220 for burst counts of the long, far, near and proximal detectors (54) are shown at two resolutions. A capture log 230 for capture counts of the long, far, near and proximal detectors (54), and a sigma log 240 for thermal capture sigma (Σ) values are included. Finally, comparative logs 250 include a plot 252 of the burst ratio (GB15) between the proximal/long detectors (54) and include another plot 254 the capture ratio (GC15) between the proximal/long detectors (54).

(63) 3. Third Logging Pass with Brine in Tubing and Gas in Annulus Below Shoe

(64) Returning to the process 100 of FIG. 5, once the second pass is completed, more gas is then injected from the source 44 into the annulus 22 between the casing 20 and the tubing 30 until the interface with the brine is below the casing shoe 26 of the casing 20 (Block 110). Because the gas is now injected into the salt cavern 10 where the borehole 12 extends beyond the casing 20, the gas can enter any leaks or washouts 15 in the salt formation behind the casing 20. If a washout 15 does exist, the gas will fill the washout 15 between the cement 14 and the casing 20. Otherwise, the gas will not leak behind the casing 20 if there is not a washout 15.

(65) This step of injecting gas below the casing shoe 26 may also account for portion of the mechanical integrity test. For instance, the pressure from the injected gas can be held for a predetermined amount of time, which can be governed by a given formula for a minimum detectable leak rate. In this way, a certain loss of pressure over the predetermined amount of time can indicate based on calculation that a leak rate exceeds a defined limit, which can be used to access the mechanical integrity of the salt cavern 10.

(66) With the gas injected below the casing shoe 26, operators make a third pass with the logging tool 50 operating the pulse neutron logging tool 50 in the tubing 30 to obtain measurements (Block 112). Readings from this third pass are telemetered to the surface processing equipment 60, which uses the information to generate an image response (i.e., logs) similar to those noted previously.

(67) Image Response (FIG. 6B)

(68) For example, FIG. 6B illustrates an image response 200B of the logging tool 50 in the third pass of the detection process (100) in which the tubing 30 is filled with brine, the gas is disposed in the annulus 22, and the gas has potentially seeped into the washout 15 behind the casing 20.

(69) The well schematic 210 is again shown as a function of depth and depicts the casing 212 with the tubing 213 extending beyond the casing shoe 214. The burst logs 220 for burst counts of the long, far, near and proximal detectors are shown at two resolutions. The capture log 230 for capture counts of the long, far, near and proximal detectors and the sigma log 240 for thermal capture sigma (Σ) values are included. Finally, the comparative logs 250 include the plot 252 of the burst ratio (GB15) between the proximal/long detectors and include the other plot 254 the capture ratio (GC15) between the proximal/long detectors.

(70) 4. Fourth Logging Pass with Brine in Tubing and Gas in Annulus Below Shoe

(71) Returning to the detection process of FIG. 5, operators can run the logging tool 50 after a predetermined time in a fourth pass to check that the interface between the brine and the gas has not moved (Block 112), and a determination is made whether the interface has shifted (Decision 114). The interface between the brine and the gas in the borehole 12 can be determined using known techniques.

(72) In addition to measuring the interface, the fourth pass can obtain an image response comparable to the other passes. In this way, a comparison between the fourth pass and the previous third pass can verify the resulting image responses (i.e., that the readings of the image responses are consistent, that further leaking in voids is not occurring, etc.).

(73) Shifting of the interface or discrepancy in the image responses may indicate that gas continues to leak in a void 15 or may indicate that there are errors in the measurements. Should this be the case, some previous steps may need to be repeated. If shifting or discrepancy is not detected, analysis can continue based on the image responses obtained. In general, the step of checking that the interface has not moved is part of a mechanical integrity test (MIT). The “caliper” of a washout as discussed below can be determined from the steps for the first, second and third passes.

(74) Using the image responses, for example, operators determine that a leak does exist, detect the void or washout 15, and estimate the volume of the void or washout 15. In particular, the image response from the third or fourth pass (brine in the tubing 30, gas in the annulus 22, and possible gas behind the casing 20) (Block 112 or 114) are compared to the image response from the second pass (brine in the tubing 30 and gas in the annulus 22) (Block 108) to determine if there is a washout 15 in the salt formation behind the casing 20 (Block 118). (Should it be necessary, the image response of the second pass (brine in the tubing 30 and gas in the annulus 22) can be compared to the initial image response of the first pass (brine in the tubing 30 and brine in the annulus 22) for calibration purposes.)

(75) 5. Detecting and Estimating Washout

(76) From the comparison of the image responses of the third/fourth pass to the second pass, the process 100 determines if a washout 15 is present (Decision 120). If not, additional measurements may be necessary or the process (100) may end successfully. If a washout 15 is present, then the process 100 estimates the volume of the washout 15 so the volume of resin needed to seal the leak in a remedial operation can be estimated (Block 122).

(77) If there is a washout 15, the burst ratios (GB15) from the pulsed neutron responses of the logging tool 50 in the comparison of Block 118 are expected to be lower in the third/fourth pass (Block 112 or 114) compared to the second pass (Block 108). Analysis based on the burst ratios (GB15) may be preferred because they may show the most detectable readings and differences. Analysis could be based alternatively or concurrently on capture ratios (GC15). The capture ratios (GC15) from the pulsed neutron responses of the logging tool 50 in the comparison of Block 118 may also be expected to be lower in the third/fourth pass (Block 112 or 114) compared to the second pass (Block 108).

(78) To make the determination of the washout 15 from the burst/capture ratios (GB15, GC15) of the image response, the processing unit 60 runs the MCNP (Monte Carlo N-Particle) analysis as discussed previously for different potential washout diameters in order to characterize the expected responses of the logging tool 50 for the burst/capture ratios. Regression is performed on these characterizations and applied to the actual tool measurements in the image responses to create a caliper-behind-casing estimation of the washout 15.

(79) Image Response (FIG. 6C)

(80) For example, FIG. 6C illustrates an image response 200C of the logging tool 50 after processing to determine a radius of a washout 15 behind the casing 20. The well schematic 210 is again shown as a function of depth and depicts the casing 212 with the tubing 213 extending beyond the casing shoe 214. The burst log 220 for burst counts of the long, far, near and proximal detectors from the third/fourth pass are shown at one resolution. The comparative logs 250a from the third/fourth pass are shown and include the plot 252 of the burst ratio (GB15) between the proximal/long detectors and the other plot 254 the capture ratio (GC15) between the proximal/long detectors.

(81) Other comparative logs 250b-c from the third/fourth pass are also shown. These include the logs 250b having the plot of the burst ratio (GB14) and capture ratio (GC14) between the proximal/far detectors and the logs 250c having the plot of the burst ratio (GB24) and capture ratio (GC24) between the near/far detectors.

(82) Finally, another well schematic 260 depicts the casing 262 with the tubing 263 extending beyond the casing shoe 264 as a function of depth. The region labelled 265 illustrates a washout (i.e., caliper enlargement) created after the installation of the casing 20. The estimated radius 266 of the salt formation outside the casing 20 as predicted by the analysis disclosed herein is plotted as a function of depth relative to the casing 262. As can be seen in this plot, the estimated radius 266 indicates that a r washout 265 exists in the borehole behind the casing 262 uphole of the casing shoe 264. The volume of this washout 265 can be estimated so a volume of resin needed to seal the washout 265 can be calculated.

(83) As would be expected, any pre-casing openhole log in this region would not have suggested that any abnormal borehole diameter value would subsequently develop. Any pre-casing logs would likely have recorded low count rates in the region 265, but these pre-casing logs are poor indicators of what would result following the break-up of the formation in the region 265 as a result of post-casing washout.

(84) On the other hand, the comparative logs 252, 254 that are plotted following post-casing and cementing indicate a dramatic increase in the borehole diameter behind the casing 20. From such readings, it is possible to estimate the addition to the volume of the borehole 12 caused by the caliper enlargement. From this estimated volume, it is also possible to estimate the cost of remedial work, of the kind outlined above, on the borehole 12.

(85) As shown in the process 100 of FIG. 5, operators pump resin to seal the leak and fill the washout 15 behind the casing 20. Pumping of the resin can involve using a bradenhead squeeze to force the resin into the leaks in the formation outside the casing 20 where the resin is then allowed to harden. Various types of known resin can be used that can harden in the aqueous brine.

(86) After the resin is pumped and hardened, operators can make a number of additional passes with the logging tool 50 to determine how the resin has sealed the leak and filled the washout 15. For example, comparison of burst data can be performed over intervals of time in a number of passes to determine how the resin has filled the washout 15. The burst ratio (GB15) is expected to indicate the higher density across the pumped resin filling the washout 15. In the end, the caliper analysis may indicate that only a minor void remains after the resin has been pumped so the salt cavern 10 can be returned to storage operations.

(87) D. Alternative Processes of Detecting/Estimating Washout Behind Casing

(88) In the process 100 of FIG. 5, four passes of the logging tool 50 have been described. However, fewer passes can be used to detect the leak or washout 15. For example, more or fewer passes may be needed to make baseline measurements. Also, a pass to check for shifting of the interface may not be necessary or may be performed more than once. In general, the image response from the third pass (brine in the tubing 30, gas in annulus 22, possible gas behind the casing 20) is at least compared to the image response from the second pass (brine in the tubing 30, gas in the annulus 22) to detect the washout 15 and estimate its volume.

(89) The caliper of the washout 15 behind the casing 20 can be estimated with fewer passes than necessarily disclosed above. For instance, the fourth pass to check that the interface has not moved may not be necessarily performed and may not be used in the estimation. In general, either the base pass of Block 104 or the second pass of Block 108 need not be performed in order to estimate the caliper.

(90) For example, there may be implementations where the base pass of Block 104 or the second pass of Block 108 cannot be performed. In these cases, a water filled base pass can be simulated by considering other log data, such as Capture and Burst counts/ratios, GR and other available Open Hole log data. A simulated baseline can be established from these log responses by normalizing uphole and downhole of the suspected washout in the borehole. With this simulation of the simulated baseline, it is now possible to estimate the washout 15 using the same methodologies disclosed in detail previously, albeit there may be higher error and less accuracy with this method.

(91) As can be seen, the disclosed process for estimating the washout behind casing uses a comparative procedure between passes of a pulse neutron logging tool, where the casing annulus is filled with brine, filled with gas above the casing shoe, and filled with gas below the casing shoe. As will be appreciated, this comparative procedure can use passes of a pulse neutron logging tool when the casing annulus is filled with brine and gas in other ways.

(92) FIG. 7 illustrates another process 100 for logging the borehole 12 and detecting a void or washout 15 behind casing 20 with the logging tool 50 deployed in the tubing 30. (For purposes of discussion, reference will be made concurrently to the well arrangement depicted in FIG. 1.)

(93) The tubing 30 and an annulus 22 between the tubing 30 and the casing 20 are filled with a liquid, such as brine (Block 102), and a first image response is obtained of a portion of the borehole with the annulus 22 filled with the liquid by operating a logging tool 50 in the tubing 30 along the borehole (Block 104).

(94) A gas, such as gaseous nitrogen, is then injected into the annulus 22 between the tubing 30 and the casing 22 to a point below the casing shoe 26 of the casing 20 in the borehole (Block 110). As noted herein, the gas can then enter any leak or washout 15 behind the casing 20. Eventually, the injected gas can be removed (extracted) from the annulus 22 by displacing the gas in the annulus 22 with the liquid supplied through the tubing 30 or by some other procedure (Block 111).

(95) To inject the gas (Block 110), the gas can be injected down the annulus 22 to a point below the casing shoe 26. Once the gas is allowed to enter any potential washout 15, the gas remaining in the annulus 22 can be removed or extracted. For example, the gas in the annulus 22 can be displaced with the liquid supplied through the tubing 30 so the gas can be removed at the wellhead 28 (Block 111).

(96) As an alternative to inject the gas (Block 110), the gas can be injected using a capillary string run down the annulus 22 to a point below the casing shoe 26. Sealing or packing in the annulus 22 can then concentrate the gas to enter any potential washout 15. Once the gas is allowed to enter the potential washout 15, the capillary string can be removed, and any residual gas remaining in the annulus 22 can be removed at the wellhead 28 (Block 111).

(97) In the end, the gas that has leaked into the washout 15 in any of these procedures can remain so that comparative logging responses can be obtained between the washout 15 first filled with brine and then filled with gas while the annulus 22 is filled with brine. At this point, a second image response is obtained of the portion of the borehole with the annulus 22 filled with the displacement fluid by operating the logging tool 50 in the tubing 30 along the borehole (Block 112). The washout 15 would potentially hold injected gas. If desired, an additional pass after a time can be performed (Block 114) to determine if any shifting has occurred (Decision 116) due to a larger leak/washout.

(98) In any event, the process 100 continues with comparing the second image response from the second pass to the first image response of the base pass (Block 118). The void or washout 15 can be detected in the formation behind the casing 20 based on the comparison due to the injected gas potentially trapped in the washout 15 (Decision 120). If the washout 15 is present (Yes at Decision 120), then estimates can be made of the resin required to fill the voided space (Blocks 122, 124, 126).

(99) The foregoing description of preferred and other embodiments is not intended to limit or restrict the scope or applicability of the inventive concepts conceived of by the Applicants. It will be appreciated with the benefit of the present disclosure that features described above in accordance with any embodiment or aspect of the disclosed subject matter can be utilized, either alone or in combination, with any other described feature, in any other embodiment or aspect of the disclosed subject matter.

(100) In exchange for disclosing the inventive concepts contained herein, the Applicants desire all patent rights afforded by the appended claims. Therefore, it is intended that the appended claims include all modifications and alterations to the full extent that they come within the scope of the following claims or the equivalents thereof.