A high pressure stripper bowl adapter, used in the oilfield, containing a test plug failed during pneumatic testing, resulting in a fatality. A stripper bowl is part of a snubbing unit used in high pressure oilfield applications. Metallurgical failure analysis focused on the failed threads of the adapter ring being used at the time of the stripper bowl failure. Finite element analysis (FEA) was also conducted to determine loads at the time of failure. Metallurgical failure analysis resulted in determining that deficient manufacture had resulted from improper machining of a heat treated forging. Finite element analysis (FEA) revealed that the stripper bowl design was defective.
Photograph A pressure stripper bowl.
Metallurgical failure analysis resulted in several opinions and conclusions in this matter:
- According to the engineering drawings, the manufacturer knew that several choices of materials for construction were possible for the subject stripper bowl adapter ring, hereafter referred to as the subject adapter. The engineering drawing listed AISI 4130, AISI 4140, AISI 4142 as possible materials of fabrication.
- The manufacturer allowed the machine shop to select the alloy (AISI 4130, AISI 4140, or AISI 4142) to be used to fabricate the subject adapter. The machine shop chose and specified forged and heat treated AISI 4130 for the subject adapter. The choice of AISI 4130, a low hardenability steel, by the machine shop, as the material of fabrication for the subject adapter ring was inappropriate.
- The machine shop that fabricated the subject adapter ordered a forged, heat treated ring of AISI 4130 from the steel forging provider. The machine shop should have known that AISI 4130 is a shallow hardening, low hardenability steel.
- The machine shop machined the outer diameter of the subject forging from 12.63 to 11.75 inches in order to fabricate the subject adapter.
- The machine shop should have known that when the outer surface of a forged and heat treated section of AISI 4130 is removed, the underlying inner core (or interior) metal will be of lesser hardness as well as tensile and yield strengths than the metal near the forging surface.
- AISI 4140 should have been chosen as the material of fabrication for the stripper bowl test plug. Hardness test results recorded on the hardness map of a section from the test plug show that indeed AISI 4140 has a much more consistent hardness through the full test plug thickness. The manufacturer and the machine shop should have been well aware of the better and uniform strength properties available in heat treated AISI 4140 and AISI 4142.
During laboratory testing of the failed subject adapter, tensile specimens were machined from the wall of the subject adapter.
- Tensile test results were obtained from the test specimens fabricated from the wall of the subject adapter. Tensile tests resulted in reported yield strength values of 75,300 and 71,000 psi. However, the tensile specimens were cut from an area of the subject adapter after the machine shop had machined away the harder, stronger outer rim of the forging, in order to fabricate the subject adapter, leaving a softer core of AISI 4130.
- The tensile yield strengths of 75,300 and 71,000 psi, obtained during tensile testing, reflect that testing at this location is not indicative of the original 1/4T (midradius) hardness or strength. The testing laboratory took their tensile test specimens from the only source available, i.e., from the subject adapter wall which had been fabricated from the center (core section) of the AISI 4130 steel forging.
- The tensile test results obtained were also probably low because of defects in the sample preparation and tensile testing procedure. The tensile test results also reflect the fact that the tensile test samples were taken from the center or core region of the forging, after machining by the machine shop. Tensile test specimens from these core or interior areas would be expected to have a lower hardness, tensile and yield strengths relative to the forging 1/4T (midradius) values.
- Machining of the forged ring was done by the machine shop in order to fabricate the subject adapter. Reduced tensile and yield strength properties would be expected for properly heat treated AISI 4130 AFTER machining down the outer ring diameter and increasing the forged ring inner diameter. Low yield strength results are not an indication that the steel forging company supplied improperly or insufficiently heat treated AISI 4130.
- Visual observation of the deformed threaded lifting eye holes in the upper rim of the stripper bowl adapter ring confirmed that under the influence of the 15,000 psi test pressure, the stripper bowl adapter experienced generalized yielding. Generalized yielding, of the subject adapter, indicates a design defect and is not the result of -10% reduced yield strength steel.
- Failure of the subject adapter would have occurred regardless of whether the adapter had a yield strength of 70,000, 75,000, or 80,000 psi.
- It was appropriate for the steel forging supplier to rely on the test results of their outside testing laboratory in certifying that the subject forging conformed to the required mechanical properties.
- In filling the machine shop order for forged and heat treated AISI 4130, the steel forging company followed proper and prescribed procedures. Nothing that the steel forging company did contributed to the subject adapter failure.
- As a general principle, most engineers and testing technicians believe that water is non-compressible (at normal operating pressures). However, the fact that water is slightly compressible at high pressures, such as those being used during the hydrostatic test of the subject stripper bowl assembly at the manufacturer test site, makes it incumbent that engineers and testing technicians be made aware of such facts (and inherent dangers thereof) before conducting hydrostatic testing at elevated pressures.
- It is also quite possible that the lack of an air release port on the test plug resulted in some air entrapment in the subject stripper bowl assembly prior to the initiation of hydrostatic testing.