Episode Transcript
Rhett Dodson: On this episode of Pipeline Things, we take you to Centreville, Virginia, a place you might not have been, but in a scenario, you might find yourself as an operator. If I asked you how many times do you excavate dents? And if you don't find anything or if the final deformation after the dent re rounds is less than reporting criteria, do you just recode it and move on?
What if I told you there are certain situations where that might not be the best idea? Hang on as we take a look at the NTSB failure investigation into Colonial Pipeline in Centreville, Virginia. It's a good one. Thanks.
Rhett Dodson: I do a little thing like you do. Yes. All right. So, you know, we never...
Christopher De Leon: You just noticed that, right?
Rhett Dodson: I did. I did not notice that until we did the one-year anniversary high five-episode cool thing.
Christopher De Leon: Why don't you pay attention to me more often, Rhett?
Rhett Dodson: You know, I think it's because the intro for our audience is actually more difficult to pull off then you realize we have to navigate not knocking over...
Christopher De Leon: All kinds of stuff.
Rhett Dodson: Ms. Aaron Rodgers. Is that what we call her?
Christopher De Leon: No, see that’s terrible. Wasn't it four plays then he was out?
Rhett Dodson: Yeah, we actually we can't go with Aaron Rodgers anymore even though she said Turf won't stop her. Ms. Producer will get a different a different name. We got to go with something else.
Christopher De Leon: And come up with something good.
Rhett Dodson: Yeah, but it's harder to navigate all the stuff and sit down after you do a high five. So yeah, I never, ever paid attention to your little riff that you do after.
Christopher De Leon: When you love somebody, you know everything about them.
Rhett Dodson: Isn’t that a song? When you love someone...
Christopher De Leon: And you know you should know everything about me and you. Oh, yeah, you're an idiot.
Rhett Dodson: Okay, so you know what season it is, Chris? Not fantasy football.
Christopher De Leon: It is fantasy.
Rhett Dodson: It's running season.
Christopher De Leon: It is also run ins and it's been hot. Yeah.
Rhett Dodson: Dude, it's so bad in the morning right now. And my daughter Aubrey, my second one, you know, she's doing cross country, right? And so, I have to watch
Christopher De Leon: It builds her character.
Rhett Dodson: It does build character. She I'm actually really proud. She's she is struggling through cross country in every sense of the word.
Christopher De Leon: For the record, we all struggle through cross country. And just at what point do you begin to struggle?
Rhett Dodson: I don't think my kids I didn't blessed them with mental toughness like I did it myself. Whenever I did a Half Iron Man, I, I struggled through it. But I have my own struggle coming up. So, I'm running a Ragnar race. Oh, I told you about this.
Christopher De Leon: Yeah, Yeah, It's amazing. It's so honest with you.
Rhett Dodson: It's the worst game of relay ever.
Christopher De Leon: It's not worse. I mean, it like, it makes me feel more. There's more like.
Rhett Dodson: Well, the name Ragnar McShane. Yeah, it's like 300 men grab some swords but it racing for 24 hours.
Christopher De Leon: So, you think you’re going to wear just like the little underwear looking type shorts like the 300 guys in, like, leather sandals? Dude, that would be so cool. There's a wait Now. Do you think someone will? It's Ragnar. Do you think so? And wear like, the leather?
Rhett Dodson: Some people do in full blown outfit. Some people do it the podium. Not me. Other people. Teams will do it, like fully dressed up.
Christopher De Leon: So, we legitimately need to consider this next year.
Rhett Dodson: We need six more people. You want to take the whole. This could be an announcement. I hope our team member. So, we're a team of seven now. I hope that all of our D2I, I really want you to recognize. Yeah, you'll have one year before we do the Ragnar race in San Antonio next year, and you have to be able to run 15 miles in a 24-hour period.
Rhett Dodson: Is that what we're telling them?
Christopher De Leon: All fully outfitted? How could that be? That can be a fundraiser. Like if we got so much like we would have to do it, give it to our local charity? Yeah, that would be cool.
Rhett Dodson: Right now, I'm just thinking it was a bad decision. Yes, I think it's about Aubry.
Christopher De Leon: Aubry is great. Hey, Aubry, cheers to you for taking this.
Rhett Dodson: Oh, yeah, absolutely, man. So, but it's funny because, you know, running is just it's like one of the things you just have to do it and do the give them make yourself get up and go run. Like, you really have to be diligent.
Christopher De Leon: Know, I also ran cross-country in high school.
Rhett Dodson: You know, we taught you to play soccer.
Christopher De Leon: No, I use it for strength and conditioning. So, I ran cross-country in high school. Then we practice three days a week games on Saturdays and Sundays. And so, I felt like, well, why not run cross-country? Point is, give you my perspective on this, I liked cross-country. It was organized and there was a start and end.
Rhett Dodson: Isn't that every race ever created humankind as a starting man.
Christopher De Leon: But for example, like when we just get up to go run as part of our daily health, like we also do my medical changes all the time.
Rhett Dodson: Oh, no, I know what I'm getting into before I start the run.
Christopher De Leon: See, that's my thing, right? Is like, I'll wake up and be like, Dude, it's 107 and it's 2:00 in the afternoon. I wanted to do five miles. I'm going to do two and a half today. So, I totally changed the rules of the game. Or is like when you're in a race, you're in, you know, like I get to determine when I stop Monday through Friday. On Saturday, you don't get to pick.
Rhett Dodson: It goes both ways. So, yeah, you know, again, I've walked last five miles of the seven three and that's a miserable experience. I was wishing that one would have ended earlier. There's always what's so bad.
Christopher De Leon: Is that when you just lay over like the soccer players, just start rolling.
Rhett Dodson: Oh, shut up. Oh, gosh, I was so angry with that race. Anyway, so we're continuing on our NTSB arc that we're doing here today, Chris. And this failure is interesting for our audience out there. We're going to be actually taking you to the events and Centreville and in 2015, dealing with a leak in a liquid.
Pipelines probably more famous than most people actually realized at the time, I would say.
Christopher De Leon: Or the impacts.
Rhett Dodson: Of the impacts of it. Yeah, we're probably more famous than what most people would have known. But what's funny is this NTSB report is like normally there's 70, 80, 100-page monsters. This one's not. It is honestly very straightforward and to the point. So, a storied time with right might be a little bit condensed today is what I'll say. But what I do want to know...
Christopher De Leon: So, everybody has to switch their audio from 1.5 to back to one.
Rhett Dodson: Yes, because it will be a little bit shorter in contrast to the one-hour long episode that we did following Kalamazoo. This one is obviously much, much, much shorter. But the reason that I say everybody is probably aware of this episode is most Americans aren't aware of the pipelines around them at all.
Christopher De Leon: We've talked about this a couple times.
Rhett Dodson: Right. This one brought them to the front. And the reason why is that this actually involves a pipeline that is crucial to the transportation of refined products from the southern part of the United States. It feeds into one of the pipelines, you should say, that's crucial to the transportation of refined products on the eastern United States and the failures in the system actually resulted in gas prices increasing on the East Coast.
Christopher De Leon: So, a lot of people knew about this because it wasn't about that happened to them. Now it's like it happened to my pocket.
Rhett Dodson: Like, “What? There’s a pipeline that has to move product from the south part refineries to the East Coast?” Yes, there is. And that actually plays a role here. So, I'll take you to Centreville, Virginia. You've been there.
Christopher De Leon: I think so. All I mean, a Centreville, Texas, it's on the way to Dallas.
Rhett Dodson: Interesting. Yeah. Well, this recall case, it's 2015, 12:00. And I think what defines this story, Chris, isn't so much the incident itself as the challenge and finding the cause of the incident. So, what happens is they get a it all begins with a 911 call and somebody says that they smell a gasoline odor. The fire department responds to the report.
And it's kind of interesting. It was it was an employee at a restaurant that called in the 911 call and they immediately suspect that it's coming from a nearby gas station. So, they think the next door, gas station is either dumping gas into the drains or as a leak thing. So, they spend all this time for the first 3 hours trying to figure out if it's something related to the gas station.
And they rule out the gas station pretty quick and then they realize, hey, there are two fairly large liquid pipelines right next door. Maybe it's from them, right? So, they contact Colonial Pipeline in this case. And Colonial spends 3 hours trying to determine if the gasoline odor is coming from one of their pipelines. And there's two pipelines in this quarter, line three and line four.
And these are not little pipelines. These are, from a liquid standpoint, giants. These are 36-inch diameter, 281 wall thickness pipelines. is big, big diameter, high d/t, which will play a part in it later. And Colonial spends 3 hours and they're actually not able to determine by looking at their state of data if they have a leak.
They are doing borehole testing where they're going along the pipeline, trying to find it. They have trouble with the rocky soil getting confirmation of anything. They can't get the probes deep enough. Long story short, basically over 3 hours they're not able to confirm if the leaks coming from their pipeline. So, they keep this process of escalating. So originally it was just the Colonial district office.
It expands from the Colonial district office to basically a company-wide response. About eight and a half hours later, once corporate gets involved, they decide, hey, we're actually going to do a static pressure test between these two block girls, three and four. And when they do the static pressure test, they're still able to not conclusively rule in favor or exclude a leak possible in either pipeline, which is amazing.
So, line four, they say, hey, we're positive it's nothing on line four are positive. There's nothing going on four. But it's possible we have a legal monitoring, just possible is all they're able to get to. So next thing that they began doing, this is where a senior integrity engineer gets involved. So, it's kind of cool. They get a senior integrity engineer and he begins prioritizing locations for them to dig in based on historical ILI. They ended up selecting and excavating six dig sites the afternoon.
So, at this point they are choosing six points based on previous features and appurtenances valves, anything they have in that area, they can go and dig in around 7:00 that evening. So, seven, full hours later, they're excavating line four around a dent that had been determined in a previous in line inspection. They find contaminated soil at this location and they keep digging all the way through the night to expose the full pipeline diameter.
And then they find product dripping underneath the pipe after they manage to get the full circumference of the pipe, uncovered. So, it took almost over a day to find the right location and a full two days to completely confirm that they had a leak in the pipeline. And of course, what they do is that they go ahead and they put a sleeve over it temporarily so they can do some other work and move on.
But it begs the question and we'll talk about a little bit is two things come up to play here. One is the nature of the feature that failed. And I'll go ahead and tell you because we're going to get into it. It's a dent, right? So, we find out that it was actually a dent that failed.
And then the question is, well, how did they not know the pipeline was leaking? So, the NTSB report spends a lot of time on the leak component of this particular pipeline. And I think it's interesting, our audience may find it interesting to know that they found that the SCADA system, that the leak was actually so small that the SCADA system wasn't able to actually find it.
The leak rate was about 0.004% of the total flow. So remember, this is a big 30-inch diameter pipeline, but even that tiny amount of flow was enough to obviously contaminate a significant amount of the soil. And obviously the resultant 911 calls and an NTSB report, if you will. And so they determined that, hey, we can't find this in the skater system and it's about that 0.004%, about 550 times lower than the SCADA leak detection limits.
So, it explained, you know, again, Colonial was off the hook from a sense of being able to detect and identify it. But the dent is really what becomes the real story. Yeah. And so, as we get there, before we get in, I want to ask you, Chris. Because you were an operator, where were dents on you guys’ threat spectrum? You didn't know I was going to ask you this question.
Christopher De Leon: Yeah, I’d say it is fun. I enjoy dents.
Rhett Dodson: You enjoyed them?
Christopher De Leon: A little bit.
Rhett Dodson: Where do they rank on the threat spectrum? I mean, you had metal loss, you had SCC...
Christopher De Leon: They were high. It was high because I feel like you never know what you're going to find, right? And so, I mean, let's think about this just in general that would cover maybe different type operators, right? Is it's like at the time when I was an operator, we would leverage different technologies from one run to the other. So, an example of that would be, hey, we had a caliper run down on this date with this technology.
Now we have a more modern caliper or another MFL. We didn't detect it, didn't metal loss previously, but the dent was there. Now we have a metal loss that we can correlate to. So, it's dent with metal loss and we know that code and HCA says you need to dig it. And so, as an engineer wanting to investigate things and understand, and as our friend Ted Lasso would say, let's be curious.
So, I'd ask a lot of questions.
Rhett Dodson: I thought you were going to say, be a goldfish. And I'm like, I don't know how that works here.
Christopher De Leon: Maybe that comes up later. Maybe we'll bring that up later. But I ask a lot of questions. So, I’d call analysts and we pull up the data and I would start asking logical questions, right? Is this a deformation? Is this a dent?
Rhett Dodson: Well, let me ask a question. Did you have instances where you dug up the dent, found something and then recoded the pipe and moved on?
Christopher De Leon: Absolutely. And not only that, but I would I would do my best to leverage that information on future ones. So, if we found shallow dent at the bottom of the pipe in a certain terrain that it was a break in the coating and we found light metal loss that was triggering any metal loss associated with that deformation, we would characterize those, right?
So, if we saw on the shoulder of the dent versus in the dent or the deformation, we would integrate all that and begin to classify them and say, you know what, these are higher property and these are not.
Rhett Dodson: And I think what you just described, actually represents what a lot of gas operators do. I think it is actually directly what contributed to the failure here. So let me catch the audience up...
Christopher De Leon: But I will say this: It's also fair for us not to say that that's that common either, right? Because I would say there's a...
Rhett Dodson: What's not that common?
Christopher De Leon: It's almost a little bit of the Wild West, if you think about it, right?
Rhett Dodson: Oh, dents?
Chris: Yeah, dents.
Rhett: Oh, absolutely. I'd say particularly around this time and before even now, we might say that dense are the Wild West with operators’ response is looking different. But what caused this failure? We take you back a bit. So, remember I told you it was a dump? Yeah. It turned out that this dent had actually been excavated before 2015.
It had actually been excavated in 1994. You know, that's what that's 21 years later, 21 years earlier. The dent had been excavated. It was below the repair criteria. So, once they excavated the dent, it might have been deeper before, but once they excavated it, the pipe re-round because it was a bottom side dent.
Christopher De Leon: So, if it was constrained or unconstrained...?
Rhett Dodson: Yep, so most bottom side dents, the dent re-rounds.
Christopher De Leon: You said it was rocky, right? So, what does that matter?
Rhett Dodson: Because it just explains...
Christopher De Leon: What if it was like a nice farm ground or something like that?
Rhett Dodson: Oh, say it's, I mean it's less likely that you're going to get a bottom side restrained dent that would re-round, but it's not impossible either, right. I mean you could have something as simple as a wooden two by four that was left in the trench to support the pipe can create a bottom side dent that when removed, will pop out.
That's what happened in 1994. So, they went there to excavate this down. It re rounds to round 1.6%. And they're like, “Hey, regulatory doesn't require us to do anything. Let's move on.”
I'll say for our audience out there, it has been almost without a doubt proven, not in all circumstances, you can't say it every time, but in most circumstances when you have a deep bottom side restrained by deep I mean greater than 3% or 4%, maybe even 2%.
And then you unrestrained that didn't allow it to pop back out and then just re bury it. You've actually made the situation worse when you have the threat of fatigue cycling present, which is this was a liquid pipeline, heavily fatigued cycle. So, what happened is over the next 20 years, that dent was now free to flex in a way that it wasn't free to flex before.
Christopher De Leon: Free to flex? Or Fridaflex?
Rhett Dodson: Free to flex.
Christopher De Leon: Not Frida. So, I kind of thought of like Frida’s like unibrow where it's like...
Rhett Dodson: Okay, I feel like you're making fun of my Louisiana accent that comes out every now and then.
Christopher De Leon: Free to Flex is doing this.
Rhett Dodson: Yes, free like free like this. Free to flex, Free to flex. And as it does over the course of the next 20 years, it generates fatigue, cracks on the outside surface of the net that grow from the outside surface to the inside surface, resulting in the condition that they observed.
Christopher De Leon: So is the idea there and we'll get to more questions later. So the idea there is that had it not been excavated, would have had a longer life.
Rhett Dodson: In this particular case, yes. Or I'd give you another alternative. Had it been preemptively repaired when excavated, it would have had a longer life. The point is that changing the conditions of the dent and I'll say in fairness, I don't think in 1994 they would have been as aware of this. But we definitely have the research to understand this now.
And the funny thing is, early on in my career, you know, I used to get questions like this from operators a lot.
“Rhett, If I excavate a bottom side down and then I don't repair that, I make it worse?”
That is actually a common question that I would get early in my engineering career. And the answer is again, in most cases, yes.
So, in this case, the NTSB report, it's kind of interesting because we'll get into this more later. They eventually cut the dent out. The NTSB report concludes that it's corrosion fatigue. I thought it was interesting that they labeled it corrosion fatigue. They didn't have a lot of information from the metallurgy report. I think it's fair to say that the corrosion was very, very, very minor. If anything, it might have shortened the fatigue life by giving like an initiation site to the cracks. But this is not metal loss like 20 or 30% or anything like that.
Christopher De Leon: So, what you're saying is, is that...
Rhett Dodson: It's predominantly fatigue.
Christopher De Leon: So, there is a basis, if there's any metal loss there can make it worse.
Rhett Dodson: Yes, unless you're a Polish steel, you know, anything from Polish steel down to light aerial corrosion is going to impact the fatigue life of the more than again, if you had perfectly polished steel. So, but you know what? We're going to take a short break right here and come back and let's dive into a little more information on what other things were present in this particular failure.
Rhett Dodson: So, welcome back. So, we're picking up in Centreville, where I'm telling you the story that led to the failure of the right. So, again, it's pretty straightforward. This particular dent had been excavated in 1994. They didn't find anything other than a dent. Final depth was around 1.6%. No regulatory criteria required them to do anything, recoated it in the field.
I'd say for 1994, that was pretty common. Shallow dents, you know, so again, it's important to consider that all the depth base regulation, even up until this time 2015, it's just depth based, right? So, criteria both liquid and gas pipelines is really only driven by the depth of the dent.
Christopher De Leon: And what would have been the trigger to want to go after it, right?
Rhett Dodson: Yeah.
Christopher De Leon: Well, the whole point of like generally in mentality, we're going to talk about integrity in not just the operations and maintenance integrity is normally correlating dents to mechanical damage, right? Typically, on the upper half of the pipe or something that struck the line. And so, I mean, naturally, I guess we could kind of see where if there's something on the bottom of the pipe, it wasn't struck, the pipe re-rounds, it looks like it’s mint condition.
Rhett Dodson: After 1994, they continue to pick up the stint in ILI inspections, but it comes back at 1.6% bottom side, even if it been top side, it wouldn't have mattered. So, what we see here is a major gap in regular artery response to dents, right? What comes out and is a bit shocking...
Christopher De Leon: I guess the question here is, is it a gap, it's regulatory response or an integrity engineering analysis?
Rhett Dodson: Well, let's say it's both, because let me give you a little more information. So, after as FEMSA starts to dig, and they recognize, okay, there wouldn't have been hazardous digs, so starts to dig in to Colonial's response versus investigative. I'm sorry, that wasn't even intended pun. How did you play that that way? As they do that, what they find out is colonial comes to the table and says, yeah, we recognize we're not required to do anything in dents that are less than depth of 2%, but we've had failures in four other locations with dents that have depths less than 2%. And so, in Pelham, Alabama, Phoenixville, Louisiana, Simpsonville, South Carolina, also had through wall cracks in pipeline, in the pipeline with depths less than 2%. So, this is actually the fourth incident that they've had in a shallow dent.
Christopher De Leon: So, do you think that was the trigger for the senior engineer to say, why don't we go look in these dents?
Rhett Dodson: Oh, I'm sure it's probably one of the reasons why they included a dent and where they had looked. And so, I think that kind of becomes the crux of this episode lessons story. But then the impact that it had on dent integrity management moving forward, Chris, and that's where I want to drive the conversation.
So, one last thought to close it out is I told you that at the time that the incident happened, they sleed the down. Yeah, but the reason they did that is so they could actually run a crack inspection tool. So, they ran a crack inspection tool through the pipe and they actually was ultrasonic. So, UCD cause it's liquids, obviously, and the UCD tool was able to identify the crack in the dent.
Christopher De Leon: UCD?
Rhett Dodson: UCD.
Christopher De Leon: UCD stands for what again?
Rhett Dodson: Ultrasonic crack detection. And so, but then after that, obviously they cut it out and that’s what we get today. Let's get straight to the recommendations. So it says to PHSMA: “PHMSA is to work with pipeline trade and standards organizations, to modify pipeline acceptance criteria, to account for all factors that lead to pipe failures caused by dense, and promulgate,” such a good word...
Christopher De Leon: Promulgate pro means like.
Rhett Dodson: For propagate could have an initiate but promulgate “promulgate regulations to require the new criteria to be incorporated into integrity management programs.”
So that's the first one.
Christopher De Leon: And we know at the time the amps were generally focused on each year, which is interesting as well.
Rhett Dodson: And then their second recommendation which flowed down to Colonial as well, is “require operators to either repair all excavated defects.”
That's interesting because that would require some data gathering, “or install leak detection systems at each location where it is not repaired to continuously monitor for hydrocarbons.”
I mean, that seems a bit impractical, right, if I'm going to have to.
Rhett Dodson: I mean, unless the dig depth is like really deep or really expensive, it's almost seems like it has to be cheaper to dig than install a continuously monitoring system. But anyway, and then the third one is they issued a directive to the Association of Oil Pipelines and API, and they said, “communicate to your members the finding of this report on the susceptibility of dents to fatigue cracking even when the dent is under acceptable current criteria.”
I think I would say even when the debt doesn't fall into regulatory requirements.
Christopher De Leon: Lessons learned.
Rhett Dodson: Lessons learned.
Christopher De Leon: So, they went to the industry organizations to say, hey, let's try to share this information so we don't get there. Yeah, there's a gap in information that we're responding to in IMPs, and people need to be aware of this. That was a good one. So, what happened?
Rhett Dodson: Well, so I think since this time, oh, we've had definitely a change in the dent landscape. So, most people might be aware that Colonial was one of the leaders in the development of API, ARP 1183. And I mean, that standard has its challenges. There have been a lot of people that have raised some, I would say both objections and challenges to some of the methodologies in that standard.
But it was absolutely a move towards trying to standardize some of the dent assessment methods. So, you see that as number one in API 1183 and Colonial leading the way in some of that. The second thing that you see taking place is obviously we already talked about it is 192 712 and the updated regulations on dents. So, there was a lot of talk that that PHMSA had wanted to look to 1183 in order to develop that.
But to be frank, 1183 was both took too long and I don't think there was enough there was enough industry consensus around 1183 to actually make it into regulation. So, PHMSA kind of did their own thing in 192 712 C.
Christopher De Leon: So, that's some good background. So, let's turn it up a little bit. So, let's structure the conversation. So, if we are to look at like we always like to do, right, right. We look at liquid regs in 195 and gas regs in 192. I feel like we're kind of finding see them converge a bit. You know, we talked a little bit at some of the other episodes we've done where it kind of seems like you we now see that maybe because of the regulatory process, how long it takes.
We're seeing regulators reference one code and one language i.e. gas on liquids and stuff because the rulemaking process. So, maybe they're coming together. But at this point and for our audience, I mean, let's kind of clarify this. What is the biggest discrepancy between 195 regulation on dents and 195 as it stands today?
Rhett Dodson: Well, as it stands today, there is no real difference, to be frank, not appreciable write both standards are depth based, both standards are interaction based. That's where they go. So again, I'm excluding the proposed ECA process for dents, which is still proposed is my understanding is not in final adoption.
Christopher De Leon: So, we'll go this way, we'll say it this way. So, I think the way we say it is, it's in let's break it down. So, if you're a liquid guy and you're dealing with pipeline integrity management, you're in 195 452. And if you go to paragraph eight, you see these are all the prescriptive things that you need to respond to. So it's our database right where it is on the pipe. And is there something interactive?
Rhett Dodson: And again, I've said it multiple times in many publications, I've done all things being equal. Yeah, depth based has some merit. Deeper dents are typically more susceptible to fatigue, all things being equal, the challenge is all things are never equal.
Christopher De Leon: Okay, so, so let's unpack that. Why, why does depth matter? Right? Because we have integrity engineers. Maybe they haven't applied the mechanics yet or whatever it may be. Why are deeper dents worse?
Rhett Dodson: All things being equal, things. If you have two unrestrained dents, the deeper one will generally be more severe. Generally, if they're the same length, same diameter, same wall thickness, then the deeper one will be more severe.
Christopher De Leon: And you said restrained?
Rhett Dodson: I said unrestrained. So, dents come in two flavors. Dents come in the restrained flavor, which is not free to flex, something is holding it in place, not free to flex. And then your unrestrained ones are free to flex. Which means there's no restraint.
Christopher De Leon: They're not being held down. They're not being held together.
Rhett Dodson: It's often used synonymously with constrained, constrained or restrained. I prefer the restrained definition, but both are used interchangeably in the industry.
Christopher De Leon: So and so that was that we kind of reference 452. So, if we look at the gas side, traditionally we look at Subpart O 933, which is the second one, 9293, to find your response criteria. Same thing, right?
Rhett Dodson: Same thing.
Christopher De Leon: And or you're interacting with something. And it's fair to say that we understand if it's interacting with something, why does that matter? So, if you have dent with metal loss, what do you think is the basis for a while that set in that's more, more a higher risk to integrity?
Rhett Dodson: Well, so let's if we say again dealt with metal loss has a greater possibility of being associated with mechanical damage but we can we can shelve that, shelve the mechanical damage concern to just deal with this threat.
Christopher De Leon: Because that wasn't what happened here.
Rhett Dodson: Yes, it's that corrosion and of itself would reduce the fatigue. Life in an unrestrained interaction with a long scene would reduce the fatigue life. These are all things that would reduce the fatigue life. So, the origins of the depth-based regulation or twofold. Right? I think it's important to recognize that around this time, too, you do have PRCI, who's been doing a substantial amount of full-scale test work that I was involved in very early on, even at Stress, around the dents and their fatigue life, but that work’s ongoing. It hasn't culminated into a report. Yes, it went on for a very long time, started in 2009 and wasn't completed in 2017.
Christopher De Leon: And is it fair to say those probably mechanical damage? If you guys want to Google that PRCI MD program starts with four...
Rhett Dodson: But before that testing, you have to go all the way back to Joe Fowler and Chris Alexander together. Actually, at Stress Engineering did some very early work that gave way to I think it was the API standard 1153, if I'm not mistaken, which was bottom side rock dents. It was later recalled. But in that document, they basically confirmed some of what appears in regulation, namely that fatigue of dents that wasn't really a threat in gas pipelines, if it was under 6%.
So that's why, you know, that 6% criteria kind of appears in there. But again, it was it was it was limited testing and...
Christopher De Leon: It was empirically based is what you’re saying.
Rhett Dodson: It was empirically based. And a lot of that confirm that. Again, deeper dents typically have lower fatigue lives. The challenges this pipeline, I told you, is a different animal. Most liquid pipelines eight, ten, 12-inch...
Christopher De Leon: Diameter.
Rhett Dodson: Heavier wall.
Christopher De Leon: Diameter to thickness ratio.
Rhett Dodson: Bingo. You end up with an eight-inch line that's got a quarter inch wall that only has a d/t of 42. Yeah, right. You take this pipeline here, 36 divided by 281. We're actually up over the hundreds. And when that happens, the fatigue becomes much more susceptible. So that’s why I say a 2% dent in an eight-inch pipeline, you can probably live with.
A 1.6% dent in this pipeline, you can't live with. So that's what depth is. That's why it's so there's so much, I think.
Christopher De Leon: So, there's a couple of things you're unpacking for us here. It's kind of the this is hopefully not people aren't listening to 1.5 and they're at the 1 so, we can kind of break this down a little bit. Right. So traditionally we think about dense. Normally our minds potentially go through mechanical damage. And that's not the case here. Where mechanical damage.
Rhett Dodson: You have a time dependent failure mechanism.
Christopher De Leon: Yeah, we would almost say depth is almost could be argued independent if you're dealing with mechanical damage. Because you may be dealing with cold work, some strain hardening initial cracking. So, depth at that point maybe doesn't matter. So that's one mindset. This mindset is a bit different, right?
I think what you're seeing here is it's you can have a dent independent how it got there. Top or bottom. The threat is fatigue.
Rhett Dodson: Yes.
Christopher De Leon: And so, depending on how your pipe is able to manage the fatigue or address the fatigue and any additional concentrators of stress, i.e. a weld or corrosion in this case. So, for example, me, when I was doing this, we weren't really concerned about fatigue.
Rhett Dodson: And that’s common with most gas operators.
Christopher De Leon: I'll say this right, if I had a date with metal loss, I look at the same as if I was on the shoulder, it was very light, I would say, “Hey, they probably broke that broke the coating system. We've got a little bit of mild corrosion here. I'm not really worried about this.”
You know, that's not the case here.
So, if you have a dent with metal loss, I mean, walk us through that. What's the thought process through dents with metal loss in a in a in a liquid line where you're, or you in a gasoline where you’re susceptible fatigue?
Rhett Dodson: So, I think in both cases and we've seen that change so 192 712 C now requires fatigue and I'm going say look, nine times out of ten, fatigue is going to be a check the box operator for a check the box operation for a gas operator. I have yet to come across a gas pipeline that has fatigue cycling in any sense of the word, like a liquid pipeline.
Christopher De Leon: Well, I'm going to break your thought here. Have you done analysis on more than ten dents? Have you done a high level, level three type engineering on more than 10 dents?
Rhett Dodson: Yeah. I'm trying to think where you're going here.
Christopher De Leon: Have you done it on 100?
Rhett Dodson: It's a hundred individual level threes. Probably getting close to that I’d say.
But again, when we talk about stress concentration factors, I don't consider that to be a full level three. We have done thousands of those. Som I had my eyes on thousands of dents.
Christopher De Leon: Okay, so...
Rhett Dodson: And that's not exaggeration because I know we did at least 2500 at one point in time.
Christopher De Leon: Okay. Because what I'm trying to get to is I'm trying to qualify you a little bit for the approach to if I have an interesting situation where I have a high D/T, I have a dent that's either potentially restrained or constrained or it's unrestrained, this process of vetting. Because again, when I look at this failure, I think you had somebody with their hands on this that saw it and said, I don't think this is an integrity issue.
So, at what point do you start identifying the integrity?
Rhett Dodson: Look, I think what it's dealing with a couple of things were unique here. One, that had previously been excavated and wasn't prepared. So as an operator, if you knew you had previously excavated dents that weren't repaired, those were a threat. Right. And I'm going to say that even in gas pipelines, they're potentially not a good deal. And you say, well, why Rhett? If that gasoline ever changes service or changes operations, yeah, then you have potential challenges there, right?
So, but in a liquid line for sure, if you have previously excavated dents were the practice of the IMO was just to leave it recoded and move on. I think that's potential. The higher your D over T, the greater your potential.
Christopher De Leon: So is the risk in that it re-rounded. So, there was some material property changes or just that you've now allowed it to become susceptible to fatigue.
Rhett Dodson: Again, if the dent is held in place, if it's restrained, yeah, the failure mechanism will still be fatigued, but the fatigue has to grow from the inside surface, the cracks generate on the inside surface and grow to the outside surface. That's how we know restrained dents fail. The damage accumulation is lower for the restrained dent, if it's like, let's just say 4%, then if you allow it to re round and you don't subject it to a hydro test or force it to any additional re rounding, that dent that re rounds under, let's just say MEOP or MOP, that dent is now going to be free to flex and its fatigue susceptibility would be much higher than it was when was restrained.
Christopher De Leon: I think we've driven that home.
Rhett Dodson: We have. I hope so.
Christopher De Leon: That's the goal. Hear it three times, maybe it sticks.
Rhett Dodson: So.
Christopher De Leon: Okay, so now we know if you dig dense and you...
Rhett Dodson: Whether your gas or liquid...
Christopher De Leon: You should repair them.
Rhett Dodson: And so, I think the only way you don't is if you literally can't see it anymore. So that operators that have come in and say, hey, look, a reduction of the famous rifle right here, bam. And then I can't find it. If you literally re-round the dip and you can hold a straight edge to the pipe and you can't find it anymore, I'm not going to say install a sleeve or install a composite wrap.
But if you if you excavate the dent and you can still see the deformation, I would just say, hey, do your operators in the public service I mean, you already got the ditch open, the expense is already there. Wrap it or put a steel sleeve on it and make sure you fill it with a compressible material.
If you do that, you will have probably checked off honestly, everything that would you would want to learn from this particular NTSB failure report.
Christopher De Leon: So, what if you dig it up and there is cracking, what do you do?
Rhett Dodson: So, I mean, most operators if you dig it up and it's cracking, you’re probably going to look at excavating and getting it out of the pipeline. If it's a critical line.
Christopher De Leon: Like getting it out. What do you mean?
Rhett Dodson: I mean cutting it out, right. So, cutting it out. I think most operators would choose to cut it out. You're not bound to do that. I mean, you could put a if I knew it had cracks in it. My personal recommendation would be a steel sleeve. That's what I would go with in that in that case with again a sound filler material and sleeve, and you would mitigate most of the fatigue growth that you would expect if you did that at that point.
Christopher De Leon: Yeah, I think probably worth considering if you can buff it out too. So, do some MPI, get some Phased Ray, try to get an appreciation for how deep it is if you're if you're operating procedures allow you to buff it out too. I would say probably consider that because at that point...
Rhett Dodson: No, it's important because operators have asked me that. But you said found a dent with a crack. Again, I mentioned a lot of times restrained dents the cracks were grow from the inside surface of the pipe. So, if you excavated, restrained it and don't find any cracks, there may not mean you're out of the woodwork. I often tell people you need to get Phased Ray or something like that.
Christopher De Leon: Well MPI and Phased Ray right. So, if you do some kind of phase ray or some other crack detection in the E technology to understand what's going on, you always want understand what's going on before you start buffing.
Rhett Dodson: Exactly right. And then obviously you couldn’t buff in the inside surface.
But let me ask you a question. If you were it, let's say you were liquid operator and I dump this on you and you went back and looked and you're like, yeah, we probably have at least 300 deformations that we historically excavated and just recoded. What do you do with this?
Do you just go and start digging all of them up? What do you think about the potential to use ILI? Since we’re the ILI guys.
Christopher De Leon: I'll be honest with you. Right. So first and foremost, if I'm an operator, I'm probably I've got a time constraint and I've got a program that I'm already trying to deal with, and this is probably being dropped on my lap. And you know, my favorite lifeline is phone a friend, right? So, I would get on the phone and I start calling a bunch of people, right?
I would start understanding what records that we have in the field and we found it is it you want at least my experience at the time has it still remains that a lot of the field crew are still in place. And so, you can call those guys and say, “Hey, I'm looking at a dent that's across County Road, pick in something one, three, three.
It looks like using Google Maps, it's about a mile north of there. Like, do you remember that dig that was 12 years ago?
Rhett Dodson: I didn't find anything.
Christopher De Leon: What was it exactly? But that begins to allow you to begin to understand the gravity. Now, you going to go on the phone and call all 300? No, but I would say I'd phone a friend understanding what's going on in the field, what records I have. I'd probably call a guy like you and say, Hey, this is the ILI data I have.
I need to start building a program. I will tell you this. It's normally going to turn into some kind of program. And then what I would do, is I would turn to industry to see what kind of work has been done so I could try to start risk ranking these. And I would probably go to something that you brought up.
You brought up API RP, a recommended practice, 1183. I'd phone a friend and say, “Hey, I've got 300 of these, how would you help me start risk ranking?”
Rhett Dodson: And honestly from there it actually gets a bit easier, right? So, I'd say if you did that you'd be in a good place. You look at doing pressure cycle analysis on all of your pump stations. Generally, what's closer to the pump station has higher cyclic fatigue. It usually drops down below that. Unless you have significant elevation changes, it drops down.
So, it does, there is a method there where you could begin to prioritize which ones that you would go after.
Christopher De Leon: And I would say this is it's I'm depending on my program. So one, I'd phone friends, I'd kind of try to understand information what's available, some lessons learned. That's kind of data gathering mode for me. And I like to be curious, so I like to talk to people. I'll learn as much as I can, and then I'll try to propose a program. And being an ILI guy, I've kind of always be an ILI guy...
Rhett Dodson: So, what do you feel about the potential for ILI tools to be able to find cracks and dents?
Christopher De Leon: And that's where I was going, right? So, I'd start trying to look at what my ILI program could be or would be, right? So, it's possible I'm already doing other allies, i.e. ultrasonic wall thickness. It could be some wells, maybe some multi data set in metals where have two different magnetic components looking at it, maybe some, in this case we heard there was some ultrasonic crack.
Look and see if I can take advantage of existing data sets in place. I'd call my vendor and I would say, “Hey, I think I have a dent that has this profile. I can send you the radii files for it so you can understand what we think the geometry to be based on how your since your carry is mounted on the tool, is there some flex there to where that's to my advantage or disadvantage if the tool is going over a dent so that I could understand with the play associated with the dent, will my sense of carrier law for me to get good data acquisition in the in the deformed area?
Because a lot of times at that point you might not even care. It's the sizing, right? You might just be looking for detection or if I detect a discontinuity anywhere in the region of the dent, whether it be idea odd, I cannot begin to risk rank these.
Rhett Dodson: Know, they didn't say they just said that the inspection detected the crack. They didn't say what size to that. So, what do you size the through wall crack at all. So, you have God honest truth, man. You know, we're going to close them. So, what I will say is I do know from experience that Colonial undertook a massive program to address their announced a multi-year, massive program.
But one of my favorite stories, you remember this. We generated some cracks in a pipe, three of them specifically, and I knew one of them was through Wall, and I called out the Phase the Ray guy, and told him to size of them and he comes back and he sizes all three cracks and he gives it to me. And I just looked at him and I said, “One of those is through wall all week.”
He's like, “You're kidding me.” I was like, “Oh, no, buddy, you better go back. Figure out one of those is actually a through wall crack.”
And the deepest he'd gotten was 50%, which is really great. So, then he goes back and interrogates the data even more and he's like, “Oh, it's this one. Because of this signal.”
I just laughed. It was really good fun.
Christopher De Leon: So, but I would say for the audience, because you touched on the inside, right? Is it some ask a lot of questions because there are a lot of opportunities there. And if you do end up in a situation where you're dealing with a high number of dents or deformations, I would say treat them all individually. Look at any data that is or could be available.
And again, don't assume the technology can do something. And a lot of times they may not be able to get a full spec for you, POD/POI in sizing, but a lot of times they can give you enough to where you can risk rank them. And I think this is a great case where the engineer got involved and he's like, “Hey, you know, we found instead of been leaking already, they were in these environments, let's go dig these holes.”
And you could do something very similar like that in your program. You find yourself in that.
Rhett Dodson: So, the last plug for operators, but I will say the 1183 is about to undergo a revision. If you got strong opinions on dense, I would recommend that you reach out to API and get on board with that committee as it goes underneath any revisions because it is in need of a lot of work. The first round was it was a good stab at getting things in there, but it definitely needs some, it needs an overall and some operator input.
All right. So, this was the sixth in our series. We have one more episode coming up for you guys. Yeah, we'll be taking on another gas pipeline failure and I'm going to say another let's call it emerging integrity threat. But we hope you enjoyed this episode. We'll see you again in two weeks.
