What's new in production

In 2017 and 2018, I served on an ad hoc study committee of the National Academy of Engineering (NAE), to conduct an independent review of several incidents involving the failure of high-performance flange bolts (or studs) that occurred around the world on subsea drilling blowout preventers (BOPs). The U.S. Bureau of Safety & Environmental Enforcement (BSEE), who sponsored the study, was concerned about the potential for such incidents to compromise well control in the future.

The failure source. The NAE study committee concurred with the conclusions of oil industry-sponsored studies that hydrogen embrittlement of these bolts was the culprit. Three conditions are required for hydrogen embrittlement to occur in steel: 1) A susceptible steel; 2) A source of hydrogen; and 3) A state of high tension. In these incidents, high-strength steel used for the bolts or studs was susceptible; tensioning of the bolts and studs to over 50% to 60% of the steel’s minimum yield strength was used to assemble flanges; and the source of hydrogen was the offshore environment and manufacturing processes.

BSEE also asked the study committee to consider the entire life cycle of such bolts—including design, manufacture, quality control (QC), installation, and operations stages for subsea BOPs, to identify any improvement opportunities in those processes.

The study committee identified several potential general failure mechanisms for flange bolts in subsea BOP service, including: 1) Fatigue crack initiation and growth; 2) Environmentally (hydrogen)-assisted crack initiation and growth (hydrogen embrittlement); 3) Stress corrosion cracking; and 4) Combinations of all the above.

Flange bolts in subsea BOP service operate in an environment that makes them susceptible to hydrogen embrittlement. Once a microcrack is nucleated at a particle or defect and attains a critical size, it serves to concentrate the stress and can rapidly propagate across the steel. Even at low stress, brittle fracture can occur. Subsea flange bolts are commonly manufactured from high-strength steel that may be susceptible to hydrogen embrittlement. With steel this strong, the hardness needs to be controlled to a very tight band—a tighter hardness band than is required in various specifications.

Subsea flange bolts are usually highly stressed when placed in their operational environment. For surface wellhead flange applications, API specifications allow bolts or studs to be pre-loaded with an axial tensile stress up to 50% of their minimum yield strength (based on thread root diameter). This pre-load is for the initial mating of the flange faces. Bolt or stud axial loads may not exceed 83% of their minimum yield strength under operating conditions. For subsea wellheads, Christmas trees, and risers, the initial tensile pre-load cannot exceed 67% to 73% of their minimum yield strength, and 83% of their minimum yield strength under operating conditions.

So where does the hydrogen come from in deep water? Common sources of hydrogen atoms in subsea environments are: 1) Improper bake-out of hydrogen after electroplating; 2) Cathodic and anodic protection systems may provide sufficient voltage to cause water to electrolyze out atomic hydrogen from seawater in small crevices (such as threads). The study committee recommended that the oil and gas industry consider using low-voltage sacrificial cathodic protection systems recently deployed on Italian and U.S. Navy ships; and 3) The presence of zinc—there was significant agreement among the committee members that galvanizing, even hot dip galvanizing, contributes zinc into the environment. The study concluded, “… zinc coatings act as a sacrificial anode, when the steel fastener is exposed driving the potential of the steel surface to more negative potentials and increasing reduction reaction rates.”

Torqueing of bolts and studs to pre-load flange bolts can also be problematic. Although torqueing was not thought to be a factor in the BOP flange bolt failures being studied, it’s worth digging a little deeper. The tensile loading on a bolt or stud is:

Bolt stress = Torque / (Bolt diameter x Nut Factor)

The estimated value of the Nut Factor, referred to as “K,” can be problematic. It accounts for a multitude of different parameters: thread geometry, manufacturing tolerances, thread plating, surface roughness, and thread lubricant. Based on several studies, including one by Sandia National Laboratories, the NAE study concluded, that “…torqueing is a very inaccurate method for achieving bolt pre-load. The ±25%-to-30% accuracy range of using torque to pre-load bolts and nuts should be considered when determining the suitability of 20.5%-to-50% (pre-load and operating) safety margins. It is problematic to consume 50% to 60% of a very narrow bolt pre-load safety margin with pre-load variability.”

There are better methods of preloading a bolt, the most favored of which is pre-tensioning. In this method, a flange bolt is pulled to the proper tension, then the nuts are hand- tightened, thus trapping in the correct pre-tension.

Please be aware that this is only a very brief summary of a 252-page NAE report, which delves into considerable detail. The full study is available online at: https://www.nap.edu/catalog/25032/high-performance-bolting-technology-for-offshore-oil-and-natural-gas-operations.

In memory of Robert Schafrik, Ph.D., P.E., NAE–Study Committee Chair.