Let’s Be Frank: I Was So Much Older Then

After my last article discussing “subject matter experts,” I got to thinking more about how I, and others, have gone through various career paths. This got me thinking about how changes occur in one’s life, both professionally and personally. Then I was driving around last week and heard the Byrds cover of a Dylan song, “My Back Pages.”

Ah, but I was so much older then
I'm younger than that now

I had never really listened to the lyrics before, and I admit I was always a little confused by the chorus. But this time, I listened, and it clicked with what was already going on in my overactive thoughts.

Bear with me on this, so many of you are so much more intelligent than I am that I am pretty confident you probably figure out where I am going before I do.

Early on in my petrochemical/materials engineering/inspection career, I found everything exciting and wanted to learn so much. I also had the infectious optimism of youth and the added bonus of having a high degree of attitude from serving in the military. Not in the bad sense, just in that sense all young soldiers have from being physically fit and having tested some of their limits and finding they can go past them. Although, to be honest, I probably had a bit of a bad attitude as well. I brought all of that into schooling and then into my career.

I was doing failure analysis with a very senior engineer who had been in the metallurgy and failure analysis field for over 40 years at that point. We would get a job to analyze a failure and report back to the client, and he would stroll by with a coffee cup in his hand, take a look at the failed components, and dump it on me. He would usually end this very one-sided conversation by telling me he was busy with some other job and that he wanted to see my preliminary report by the end of the week.

I loved this, and it also scared the pants off of me. I had taken failure analysis courses in school and really enjoyed it. That’s one of the reasons I worked to get a job that would put me in that role. I felt I knew quite a bit and could figure out the rest. The very first time he did this to me, the job was the failure of a production wellhead that was in sour natural gas service. As an aside, the consequent explosion and fireball after the failure leveled a pretty good chunk of wilderness at this remote site. It was one of those the world never hears about because there was no one around to witness it. By the time the remote operators got out to the well site, there were just some burning trees and a geyser of flame coming out of the breached wellhead. Not even a peep on the news.

There I am in a metallurgical/NDE lab, looking at a crate full of all the recovered components from the wellhead area with an insurance company client who wanted to recover some or all of the 50 million claim the well operator had. As I was a recent hire, someone else had gone out and gathered field data (photos, measurements, etc.). There was almost no information on operating conditions anywhere, so I got on the phone with the upstream company and finally got through to an operations engineer who said he would send me all the information they had and all of their sampling and flow rate data for the previous year.

As another aside, some big refineries were the worst clients to have to work for. They would usually send little to no operations data and just kind of expect that the personnel doing failure analysis would be able to figure it all out off of the one cracked valve – don’t be that guy when you are hiring a third-party lab.

Back to young me and my first real failure analysis job. Again, I was feeling scared and nervous, but also feeling that I can do this. I can get him a preliminary report in the next 4 days, and I know I know this.

So off I went. There was another senior engineer at this company who was an expert in burn patterns. He had started his life as an electrical engineer but got into research over electrical fires, and then that rolled into a career in fire investigations. I went to him with all the scene photos and information gathered and asked if he could help me figure out, based on burn and blast patterns, where the initiation may have occurred. He said it was a bit difficult but that the initial event did happen at or adjacent to the wellhead.

I started looking at all the components they had brought in. Like in any fire or explosion investigation, there are tons of red herrings and meaningless information to be had, but you have to go through and either take things out by evidence, or they have to stay in your “scope” of investigation because they may have played a primary or secondary role.

Most of the gaskets had been burnt and blown out. One of the gaskets on the instrumentation tree under the main control valve of the wellhead looked promising. But I couldn’t say for sure it started there or knocked out other evidence. There was a lot to go through. Local and remote instrumentation set-ups, flanged connections, remote and hand valving. It was feeling a little overwhelming.

When everything was said and done, I was pretty sure it had been the connection to the flow meter that had been the initial point of failure. This was a flanged connection, and the gasket was gone. I thought maybe I could get a hold of servicing and maintenance records, and those records might help out.

I quickly put together a preliminary report of my findings and path forward and excitedly took it to the senior metallurgical engineer. He took one quick glance and said, “It’s interesting you ended up on the connections to the flow meter; maybe you should look more into that flow meter?” He then gently and kindly told me to get back to work and not to bother him until I had looked into the flow meter in more detail.

I felt good about tracing the origin back to the approximate right area, but I was wondering what could be up with the flow meter. I gathered the part numbers and worked with a local NDE company to get radiographic testing (RT) shots of the flow meter before I disassembled it.

The internet was young and kind of useless at this time, so trying to look up numbers just got me to the company’s website, but they didn’t have their full catalog online at that point, and the part number was bringing up no hits. I left a voice message with the local technical sales guys asking for some information.

The RT came back. The company had an X-ray tube in their shop and used it to take a bunch of variable-density shots from different angles to maximize the details we could see. There was an internal bearing system that rotated a part with the flow for the instrumentation to pick up, and the bearing cage in that sealed bearing looked messed up, as did some of the balls. Everything else looked pretty good.

Now I had to wonder: was the bearing failure an initiation event or a secondary consequence of the loss of containment, explosion, and fire?

I started dissembling the flow meter, documenting it every step of the way (as an aside, all parties involved in the investigation and insurance claims at this point had agreed to let us work alone as the testing lab – that meant I didn’t have to get pre-approval for every step of the investigation like I had to on some jobs). There were eight small ball bearings in what should have been a sealed bearing. They all were discolored by fire, and many had visible cracks. Again, I wondered if this was part of the fire damage, or did this damage lead to the fire?

I performed a penetrant inspection to identify and document the cracks clearly. I then took one of the least cracked bearings, cross-sectioned it, and made a metallurgical sample. I could then put this under both the inverted light microscope for viewing of the structure and cracking. But I could also do microhardness testing and use a scanning electron microscope to take semi-quantitative chemical analyses at the microscopic level.

As I was prepping to do some of this, the tech sales guy from the flow meter company called me back. He told me that the flow meter was designed for sour gas service and met the requirements of NACE MR0175 (which I am sure many of you are very familiar with). Once I told him I was doing an investigation, he said he would send over the technical brochure. He then asked me to talk further to their legal department, and that was the end of what he could tell us in this matter.

I was not super familiar with this NACE standard at the time, so I started reading.

One of the things that NACE MR0175 does as a standard is lay out the material requirements for sour service use so as not to cause cracking. The bearings were a low alloy tool steel material that needed to be manufactured and heat treated to have a hardness lower than Rockwell C-scale 22 (22 HRC) in order to avoid cracking problems in sour service.

I am now very interested in getting some microhardness done. And guess what? The bearings were too hard for sour service, and I was finding more evidence (micro-scale chemical analysis in the crack tips, etc.) that the cracks were sulfide stress cracking.

I quickly edited my original preliminary report and took it to the senior metallurgical engineer. He said “Now, this looks good. You have a provable cause and have proven all the other leaks/fires were likely secondary.” He then told me he had been involved in a lawsuit against the flow meter manufacturing company because they had a batch of meters made with these bearings that could not be in sour service. It turns out the bearing manufacturing had problems with some of their heat treat processes and hadn’t been doing proper quality control to check batches. So, they had been supplying these sub-standard bearings to a bunch of companies.

I did ask him why he didn’t just tell me that in the first place. All he said was having a predetermined outcome bias in an investigation is the worst thing you can have, and he didn’t want to bias me. Gee, thanks.

Because of his experience, he had a starting point I couldn’t even come close to. Did that stop me from being cocky as hell? No. In fact, it made me go into the next investigation with some biases that did not help, and that took me a while to get myself out from under. Through that process, I learned a bit of humility and realized as I matured and gained experience, the only thing I really know is how little I really know.

That’s my take on the lyric, “I was so much older then, I’m younger than that now.” I acted with way more firmness and authority than I should have when I was younger. Now I (generally) take a calmer view -- although there are those who know me that might argue about my calmness.