John F. Kiefner
Kiefner & Associates Inc.
Worthington, Ohio
Abandoning or replacing electric-resistance welded (ERW) line pipe in service is impractical and undesirable despite the abundant evidence that large quantities, especially if produced before 1970, are inferior.
Part 1 (OGJ, Aug. 10, p. 45) of this two-part series reviewed the ERW manufacturing process and many of the service problems encountered where this pipe has been used.
This conclusion examines the phenomena of flaw growth and pressure reversals in ERW pipe and offers suggestions on how an operator can ensure that his installed ERW pipe will operate safely.
It must be made clear, however, that when the objective of ERW pipe manufacturing has been met, namely, when the microstructural appearance and the mechanical properties of the bond line region are virtually indistinguishable from or superior to those of the parent skelp, the quality of the product is at least equal to and can be superior to pipe made by any other method.
CRITICAL FLAW SIZE
The effect of low toughness in the bond line region of older ERW materials can be seen by comparing critical flaw sizes in Figs. 1 and 2. The curves in these figures are based upon the well-known surface-flaw equation developed through research by the Pipeline Research Committee of the American Gas Association.1
Both figures are failure pressure-vs.-flaw size relationships for 16-in. OD by 0.250-in. W.T. X-52 pipe.
The difference is that Fig. 1 is based on a full-size Charpy V-notch specimen equivalent toughness of 25 ft-lb representing the level of toughness which one might expect for the skelp, while Fig. 2 is based on a full-size Charpy equivalent toughness of 2 ft-lb as was found for the LF seam examined in Reference 6 of Part 1.
Compare the critical flaw lengths at the maximum operating pressure (MOP). A longitudinally oriented flaw in the skelp which is 50% through the wall thickness would have to be nearly 6-in. long to cause a failure.
In contrast, a 50% through-wall flaw in the bond line which was only 2-in. long would be expected to cause a failure. Also, note the location of the leak-rupture dividing line.
Based on Fig. 1, a flaw longer than 3 in. is necessary for a failure to occur as a rupture at the MOP.
In contrast a flaw in the bond line (based on Fig. 2) needs to be only about 1 1/2 in. long to cause a rupture at the MOP.
The combination of an abundance of small bond line flaws in older ERW materials, particularly low frequency (LF) and direct current (DC) welded materials, and the effect of the low toughness as demonstrated by Figs. 1 and 2, tends to explain why older ERW materials have a rather dismal track record (References 1 and 2 of Part 1) compared to ERW materials of more recent vintage and to line-pipe materials made by other processes.
PRESSURE REVERSALS
A phenomenon regularly associated with the retesting of pipelines containing older ERW materials is that of a "pressure reversal."
This phenomenon can result following the raising of the pressure level of a defective segment of pipe to a level high enough to cause the flaw to fail, but the pressure is not held long enough for all of the necessary time-dependent growth to take place.
Instead, before the flaw has time to grow to failure, the pressure is removed.
Upon repressurization, the flaw begins to grow at a lower pressure level because it has been enlarged by the previous loading. It can grow to failure at a pressure level below that reached on the previous pressurization, and if it does, a pressure reversal is said to occur.
In the testing of older ERW pipelines, pressure reversals are common, although reversals larger than 10% of the initial test pressure are rare. As a result, once a hydrostatic test has been successfully completed, an operator can be reasonably confident that no remaining flaw will be large enough to fail at the MOP.
If a pipeline contains numerous flaws, as may be the case in an older pipeline comprised of an ERW seam material with one or more of the problems in Part 1, it is possible that those flaws may grow in service if the service consists of many large pressure fluctuations.
The potential for this kind of flaw growth, if it exists, can be dealt with by periodic retesting of the pipeline. The interval for such retesting can be predicted on the basis of an existing model.
TEST MARGINS
For the pipeline operator who must deal with keeping a pipeline comprised of a problematic ERW material operating safely, some guidance is available.
First, hydrostatic retesting is a prudent and reliable means of establishing the integrity of such a pipeline.
As has often been discussed, the higher the margin between test pressure and operating pressure, the more effective the test in establishing serviceability. 3 But the conventional ratio of test pressure to operating pressure of 1.25 may be insufficient if the pipeline has a demonstrated track record of service failures or a high rate of seam splits during testing.
As a rule of thumb, when the hydrostatic-test failure rate exceeds one per mile, it is a sign that the margin of test pressure to operating pressure needs to be more like 1.35-1.4 to ensure an adequate level of safety.
Second, when an older ERW pipeline is retested, an operator need not hold the test pressure for long periods of time (more than 30 min, for example).
Also, it is a good idea to avoid cycles of test pressure because experience shows that large pressure reversals are more apt to occur as cycles of test pressure become more numerous. Therefore if possible, raise the test pressure to the target level and hold it briefly.
If it is necessary to conduct a leak test, do so after lowering the pressure to no more than 90% of the maximum test pressure.
It should be obvious, of course, that such a scenario is not feasible if test failures begin to occur before the target test pressure is reached.
If they do, it becomes necessary to continue testing and breaking flaws until the target level is reached. Otherwise, the desired test pressure-to-operating pressure margin will not have been demonstrated.
It should also be obvious that repeated test failures introduce numerous cycles of test pressure. This is unfortunate because the cycles undoubtedly induce more flaws to grow.
But it is never a good idea to sacrifice the margin between test pressure and operating pressure.
If the pipeline operator feels that the target test pressure must be abandoned in favor of a lower one to reduce the number of failures, he should be prepared to lower the subsequent MOP as well to maintain an adequate margin of safety.
Finally, in a program of retesting of older ERW pipe, it is a good idea to keep a record of all leaks and ruptures, especially if related to seam flaws.
At the least, it is useful to document such items as failure pressures (referenced to a common elevation for a given test section), numbers of failures per test section, size and location (on the pipe) of the splits or seepers and visually to ascertain, ii possible, whether the failure is the result of a cold weld, a hook crack, selective corrosion, or some cause not readily identifiable.
When one knows the failure rates and causes that are typical for a given pipeline segment, it is possible to obtain an improved assessment of the serviceability of the segment and to forecast when, if ever, the segment should be considered for future rehabilitation.
REFERENCES
- Maxey, W. A., Kiefner, J. F., Eiber, R. J., and Duffy, A. R., "Ductile Fracture Initiation, Propagation, and Arrest in Cylindrical Vessels," Fracture Toughness Proceedings of 1971 National Symposium on Fracture Mechanics, Part II, ASTM STP 514 (1972), ASTM, pp. 70-81.
- Kiefner, J. F., and Forte, T. P., "Hydrostatic Retesting of Existing Pipelines," Journal of Energy Resources Technology, Transactions of ASME, Vol. 106, September 1994.
- Kiefner, J. F., and Maxey, W. A., "Evaluating Pipeline Integrity-Flaw Behavior During and Following High Pressure Testing," proceedings of the 7th Symposium on Line Pipe Research, Houston, Oct. 14-16, 1986.
Copyright 1992 Oil & Gas Journal. All Rights Reserved.
