Casing Coupling and Casing Grade
Field and laboratory experience suggests that most API couplings fail in tension following the application of compression load during the heating cycle. During joint makeup, both API Round and Buttress threads are loaded on the upper flank of the thread form. As the temperature increases, the pipe expands and the normal loading changes from tension to compression. As a result of load reversal, the thread flanks eventually unload and so-called ‘thread shift’ occurs, i. e.. the opposite flank becomes loaded. When steam injection stops, the well cools and the tension force comes into play and the threads load onto the opposite flank and a loose joint results.
Generally, parting of the joint does not occur during the initial steam cycle, although there have been cases where this has occurred after between 3 and 7 cycles (Carnahan, 1966).
There is no generally applicable formula to estimate when a joint will fail under
|
До (Т = constant) Fig. 4.39: Stress-temperature diagram of API Buttress connection during cyclic heating and cooling. (After Goetzen. 1987: courtesy of ITE-TU Clausthal.) |
cyclic loading. As a result, casing manufacturers test their products in the laboratory under simulated bottomhole conditions to assess joint performance. Typical performance analysis of API Buttress coupling. API Extreme line coupling. VAM Premium coupling, Mannesman’s BDS and MUST couplings, and Nippon Steel’s NSCC coupling is presented in Table 4.10 (Goetzen. 1986).
Each coupling was subjected to compressive and tensile stress cycling at 480°F and/or 670 °F followed by cycling at room temperature. The couplings were capped at the ends and subjected to an internal pressure and an axial pressure during the cyclic thermal loading. The test specimens were held at high temperature for 60 to 100 hours.
The performance of API Buttress and MUST couplings during the first and subsequent cyclic loading can be visualized in terms of the stress-temperature diagram shown in Figs. 4.39 and 4.40. The first-cycle compressive and tensile stresses are shown by a solid line and successive cycles by a dashed line. From these figures three important phenomena can be observed. First, during the hold interval at elevated temperature for about 100 hours a stress relaxation occurs. The stress temperature path moves from point 2 to point 3.
Upon cooling, path 3-4-5-6 is followed which results in higher tensile stress than
|
Fig. 4.40: Stress-temperature diagram for the MUST connection during cyclic heating and cooling operation. (After Goetzen. 1987: courtesy of 1TE-TV Clausthal.) |
that observed in Fig. 4.27. During the next heating operation, path 6-7- 8-9 is followed which is different from the previous heating cycle due to the residual tensile stress present at point 6. Subsequent cooling and heating paths are similar to 3-4-5-6-7-8-9. This suggests that the stress-temperature loop settles down after the first cycle and further changes take place slowly.
Finally, paths 4-5 and 7-8 in Fig. 4.39 indicate that a shift of loads on the thread flank of API Buttress coupling occurs during cooling and heating cycles. Load shifting on the thread flank was observed with all the couplings tested except the MUST coupling.
All the couplings within each grade of steel exhibit excellent tensile properties. From the test results, the calculated average axial stress per degree Fahrenheit change in temperature (ТЕ) during the heating cycle between 120 °F and 300 °F in the elastic range are as follows: BTC — 397 psi. BDS — 485 psi. ELC’ — 500 psi, and MUST — 500 psi.
The tests highlighted the poor gas-sealing performance in six of the couplings tested. Only the NSCC and MUST couplings retained their gas-sealing characteristics after the testing.
Based upon the twin requirements of gas-sealing and high structural strength,
the MUST and NSCC couplings in C-75-ST, C-75-TR and C-95 grades can be recommended as the best couplings for temperatures up to 670 °F.