40 CFR §146.8 states in part as follows:| Home | Tech & Tips | UIC MIT Menu |
"(b) One of the following methods must be used to evaluate the absence of significant leaks under paragraph (a)(1) of this section:
(2) Pressure test with liquid or gas…."
The most common internal, or prong one, mechanical integrity test (MIT) done pursuant to the underground injection control (UIC) program is by far the standard annular pressure test (SAPT). The SAPT is a creature of the incredibly broad language at 40 CFR §146.8(b)(2), the same language that allows the arguably superior "ADA" Pressure MIT to be utilized as a "variant" pressure test. The SAPT is a simple test wherein the annular space between the tubing and casing is initially subjected to pressure, and if the pressure can be maintained for a specified period of time within a specified percentage of loss or gain, the well is deemed to have passed. Most commonly, the SAPT is complimented by the cementing records external, or prong two, MIT.
The SAPT relies on the common sense proposition that pressure applied to a closed system with a fixed volume like an injection well annulus (sealed at the bottom with a packer and at the top with a wellhead) will be maintained even once the pressure source is removed, if there are no leaks in the system. Pascal's Principle, also called Pascal's Law, describes the effect of applying pressure on fluid in a closed container; stated simply, pressure applied to a container, such as a sealed injection well annulus, is transmitted with equal force throughout said fluid filled container. This is true even if the annular fluid is Annulus Gel (bentonite). It is Pascal's Principle that makes the SAPT appealing to regulators, along with the relative simplicity of the test. While a sealed annulus with no leaks is closed to matter transfer, it is not closed to energy transfer, because it is free to exchange heat with its surroundings. Heat transfer has not proven to be a major problem with the SAPT, presumably because of the short time duration of said MIT, but it is a major problem with regard to the Monitoring of Annulus Pressure (MAP, APM, SAPM, or SAMT) MIT where huge pressure variations have been demonstrated as a result of temperature changes over time. Because the regulations do not proscribe all leaks, only "significant leaks", a small allowance for pressure change is permitted, solving the heat transfer problem (and other problems as well).
The above discussion should make it clear that the SAPT has a sound scientific basis (though it is arguably a very inadequate MIT as discussed below). On the other hand, the cementing records external MIT, the nearly constant companion to the SAPT for Class II injection wells, is based on sheer fantasy. Cementing records simply do not prove anything, certainly not external well integrity. When the cement top log is used as a substitute cementing record, at least the presence of cement can be proven with a high degree of certainty (but only a tiny minority of Class II wells have cement top logs performed on them).
It is difficult to find any logic to the SAPT procedural variations from state to state or EPA region to region. Test pressures from 200 psi up to 2500 psi are used (1989 EPA Final Report of the MCE Workgroup), and in EPA Region IV, the test pressure is a fixed 300 psi no matter how deep the well is or what the authorized injection pressure might be (in the Region IV direct implementation (DI) states, like Kentucky). This creates the bizarre outcome that some wells are tested at a fraction of their normal operating pressure, while other very shallow wells are tested at a pressure that actually exceeds their normal operating pressure. The testing interval is most commonly thirty minutes, but the percentage of pressure gain or loss again varies widely. EPA Region IV allows a maximum pressure gain or loss of 3%, but other regions allow 5% and even 10%. The SAPT is most often performed with water in the annulus and water as the pressuring agent. However, most jurisdictions allow the use of nitrogen or compressed air as the pressure source if the annulus is liquid filled. The regulation does not actually specify what the annular fluid or pressure source is to be, only stating a "Pressure test with liquid or gas." The Region V SAPT procedure includes a provision to check annulus fluid returns because, "In the past criminal charges have been brought as a result of investigations inspired by the observation that very little liquid was returned." Incidentally, in response to FOIA requests, Region IV EPA has been unable to provide a single document showing the basis for the 300 psi, 30 minute, 3% regimen used in Region IV DI states, though they do point to an old 1984 report.
But the major problem with the SAPT is what it tests. It is one of the absurdities of the UIC program that the SAPT has a resolution in the neighborhood of one gallon per day, yet the well under test may be leaking virtually all its injectant past the casing shoe. The only required demonstration of external integrity for Class II wells is the ridiculous cementing records external MIT, often faked or altered, but worthless even when it is not (the silly cementing records external MIT was retained in EPA's 1980 final regulations for only Class II wells due to the oil industry's political clout at that time). The SAPT / cementing records combination is profoundly lame because said pair lacks any mechanism whatsoever to detect bad primary cement jobs. Leakage bypassing the casing shoe or behind casing channeling from perforations are both disturbingly common; neither is detected by the SAPT / cementing records combo. While doing research on a specialized application of the radioactive tracer survey (RATS), we discovered a well that would pass the SAPT and the cement records check with flying colors, yet said well was leaking most of its injectant past the casing shoe. How many more are there?
FOIA requests, and depositions of EPA Region IV personnel, reveal that EPA violated its own mid-course evaluation (MCE) regulation at 40 CFR §146.25(c), now repealed, by not collecting certain well logging data intended to shed light on the efficacy of the cementing records external MIT. Even so, the August 22, 1989, Final Report of the Mid-Course Evaluation Workshop contains the gem "...cementing records in themselves do not provide a test of MI...." However the MCE workgroup declined to recommend an end to the practice. Ironically, EPA Region IV personnel have resisted the use of meaningful MITs like the RATS, a few even going so far as to make false and misleading statements.
TIP: If you must perform SAPTs, keep in mind that they are easier to pass when testing with gas. Fill the annular space with water, but provide the actual test pressure with bottled nitrogen. It may give you that edge that makes the difference between a pass and a fail.
DO NOT USE OXYGEN as an MIT gas; it is a potentially deadly practice.