Abstract
Negative pressure tests are an important step in proving well integrity for some well operations. Although the concept is straightforward, there are no standard procedures for conducting and interpreting negative tests. Recent industry experience has shown the potential for these tests to be misinterpreted with disastrous results. This paper describes proper quantitative interpretation of both successful and unsuccessful tests demonstrated on two full-scale test wells.
Tests with and without water leaks are simulated in two vertical wells with 2971ft and 5884ft depths. Test results are interpreted quantitatively to distinguish successes from failures. In addition, the impact of leak rate and the differential pressure between the leak source and well are investigated. The impacts of gas as a leak fluid, different well volumes, different surface piping arrangements, and different measuring tanks are also shown.
The success of a test should be verified with both flow and pressure checks. For the flow check, a trend between bleed off pressure and volume can be predicted based on fluid compressibility, tubing expansion, and drainage from surface piping. Deviations from that trend will identify the onset of a leak. Early detection and control of a high rate leak can reduce the formation fluid volume entering the well and the consequent risks. Detecting a low rate leak is difficult if the test duration is too short. Therefore, a 30 minute test duration was evaluated for detection of low rate leaks.
The fundamental differences between a leak and no leak are conclusively demonstrated with quantitative results. Approaches for overcoming complications to achieving conclusive tests are explained and demonstrated. These test results and the proposed interpretation methods are applicable to vertical wells. Drilling fluid properties, multiple fluids, and thermal expansion can also influence results but were not investigated in this study. Nevertheless, these results should provide a basis for improving field procedures and minimizing the risk of misinterpreting a leak as a successful test.
