Three Approaches for Non-Intrusive Inspection of Spheres - How Effective is the Strategy?

A question was posed to me regarding guidelines for routine external inspection of spheres, including procedural approaches along with any nondestructive examination (NDE). Firstly, I should mention that Inspectioneering Journal has published an abundance of articles related to this topic including acoustic emission testing, inspection of atmospheric storage tanks, and other NDE methods.

Here are two approaches to routine external inspection of spheres, and a mixture of the two, as a third. Firstly, one must know what the expected mechanisms are, e.g. wet H₂S damage, hydrogen blistering or cracking (this can occur in some hydrocarbon services and hydrogen services such as hydrofluoric acid storage), caustic cracking (watch out for effects of caustic carryover), even a slight amount of acid with condensing water can cause severe localized corrosion, often in the form of rivulets.

There are plenty of sad stories where owner-users performed wet mag inspection all over the internal surface of spheres, running into hundreds, if not thousands of innocuous surface indications. This kept the spheres unnecessarily out of service for months as each indication was ground out. Remember, too, that surface NDE techniques tell us very little about mid-wall or far-wall connected damage or defects.

Unless the sphere is brand new and you are performing NDE per ASME (e.g. consideration of acceptable indications such as limiting/critical flaw size), it may more appropriately be handled by the API 579 / ASME FFS-1 Fitness for Service Standard of in-service equipment. This can be applied to both in-service damage (for example, environmentally assisted cracking, fatigue cracking, pitting, local thin areas, etc.) and fabrication flaws.

Approach One: Non-Intrusive

Start with an understanding of what service limiting indications might be present, what could cause the damage, what might further drive the damage, and what other secondary dynamics might be involved. For instance, damage might start with H₂S, but secondary effects such as hoop stresses (service induced, residual stress, high hardness, and gravity load) can exacerbate the problems. If the sphere is in suspect localized corrosion service and has never been entered, an internal visual inspection might be in order.

Many use Acoustic Emission (AE) testing with an appropriate loading schedule as an initial global screen. Again, one must understand the dynamics of the types of damage anticipated to determine the feasibility and then apply this technology properly, including the AE loading schedule. An advantage of this technique is that the material stored in the sphere can often be used as test pressure for the AE test. If the stored material is not used, then one should use an appropriate inert gas.

Keep in mind that it is difficult to attach tolerable flaw size limits to results of Acoustic Emission testing.

Approach Two: Non-Intrusive

Based on past experience and engineering knowledge of fracture dynamics, perform automated UT exams of selected welds and base metal for cracking and/or corrosion. Pick the most critical areas (e.g. around leg supports and plates the leg supports are attached to, known potential problem areas for cracking or corrosion, etc.) and expand the areas of inspection based on findings.

Approach Three: Non-Intrusive

Apply a combination of #1 and #2 above. I believe this is the best approach, just in case significant damage is occurring that would not have been caught during the AE test.

Qualifiers

Make sure qualified NDE personnel apply and interpret the testing. This plan does not include considerations for insulated spheres. Make sure a qualified and experienced person that understands fracture dynamics, NDE, metallurgy, etc. establishes or reviews and approves the NDE procedures and approaches. These plans are aimed at non-corrosive services. Keep in mind these are external, not internal, techniques.

Special note: One can refer to the inspection effectiveness tables in the latest edition of API RP 581 for guidelines on developing inspection strategies that are mechanism dependent which address confidence of the overall strategy. This approach is for use as part of Bayesian updating as part of calculating a damage factor, then a probability of failure, etc. Here one might choose to address the limitations of various techniques, used separately or in a complimentary manner. With this logic, one should consider the impact of human factors (e.g. NDE qualifications of NDE technicians) and limitations of various NDE methods in detecting the possible damage and grade the effectiveness of the entire strategy. It does force the strategist to consider all relevant factors of the strategy as they relate to the probability of finding the damage.

Interested in reading more about inspection, damage mechanisms, and many other asset integrity management topics? Click the button below to download a free copy of John Reynold's updated and revised 101 Essential Elements in a Pressure Equipment Integrity Management Program guide.