In the 24 years after the Olympic Pipeline Explosion of 1999, the industry has advanced in some ways and stayed stagnant in others.
On the first episode of this special season of Pipeline Things: The Failure Files, we begin a new chapter discussing notable pipeline incidents throughout history and breaking down the NTSB failure reports that accompany them.
This story begins at 3:28PM on June 10th, 1999 in Bellingham, Washington near the pristine and lush outdoor space, Whatcom Creek. Olympic’s 16-inch gas pipeline in Bellingham experiences a rupture that dumps 237,000 gallons of gasoline into the creek and ignites. Rhett and Chris walk us through the events that led up to that disaster and remind us that despite this event occurring over twenty years ago, we always have something to learn.
Highlights:
- What series of events led to Olympic Pipeline Explosion of 1999?
- How did biases play a role in the failure? When can you trust past experiences and how can data help prevent bias?
- What strategies do we have today and what lessons can we learn to prevent history from repeating itself?
- What effect did this failure have on regulations that liquid and gas operators contend with today?
- How has data integration improved (if at all) since 1999?
Connect:
Be sure to subscribe and leave a comment or rating!
Pipeline Things is presented by D2 Integrity and produced by ADV Marketing.
D2 Integrity (D2I) is providing this podcast as an educational resource, but it is neither a legal interpretation nor a statement of D2I policy. Reference to any specific product or entity does not constitute an endorsement or recommendation by D2 Integrity. The views expressed by guests are their own and their appearance on the program does not imply an endorsement of them or any entity they represent. Views and opinions expressed by D2I employees are those of the employees and do not necessarily reflect the view of D2I or any of its officials. If you have any questions about this disclaimer, please contact Sarah Roberts at sarah.roberts@advmarketing.com.
Or maybe, what influenced regulations? Have you ever wondered where the requirements, the prescriptive requirements on DEP might come from or the incidents that drove them? And dents with metal loss? And focus on mechanical damage? Well today we dive into the first of our series of significant events, and we take you to Bellingham, Washington, in 1999, and we hope you get a good learning episode out of this. Thanks for joining us.
Intro Music
Rhett Dodson: All right, Welcome to the Pipeline Things podcast. I’m your host, Rhett Dodson. My co-host Christopher De Leon and we are excited to be in this arc with you. This is going to be a different arc for Christopher and I as you’ve already heard from the first episode and we’re going to do. This is going to be the first of a series that we’re going to do. Oh, and I’m going to say notable pipeline incidents and the in the NTSB failure reports that accompany them. I’m excited about that.
Before we get into that, I always want you guys to know, because I know that kind of like the paparazzi and other people, you guys want to know what’s going on in Christopher and I’s life. I know you’ll want to know that. You want to know how our summers are going, how we’re surviving this Texas, what do they call this? The heat dome? The heat dome or something like that? My goodness, man, it has been warm. But, Chris, before you go, I mean, how is your how’s your summer going? How are you kicking things off?
Christopher De Leon: Things are great. We just wrapped up vacation season for us in the De Leon household, so we got to go to Stargazing near Big Ben. That was fun. Barton Springs in Austin.
Rhett Dodson: How many stars did you see when you went stargazing?
Christopher De Leon: So, if you have ever looked into this, I – apparently most people in the world have never seen the Milky Way. That’s apparently a fact. So, I was like, you know what, this summer I’m taking my kids to see the Milky Way. And we have a dark zone around Big Ben, which is fantastic. And so, we went, and it was I mean it was a beautiful night. The moon was completely gone. And there was just these cloud formations were just incredible. They created a perfect blanket across the entire sky, which is not great for stargazing. So, we drove –
Rhett Dodson: I’m sorry. So, when are you heading back?
Christopher De Leon: Well, we did get to see Venus. So, my kids every night look up at the sky,
Rhett Dodson: The blue dot.
Christopher De Leon: and they find Venus before it disappears. And so, all they didn’t learn about Venus, we’ll say that. So that was cool. They learned about Barton Springs. We have seen your
Rhett Dodson: So, when you drove back where you like, “We’re going to go back and try this again,”?
Christopher De Leon: I was like, nope, I think that’s kind of still in the air. Yeah, it was a long drive. They enjoyed the peaches in Fredericksburg, they enjoyed Barton Springs because we made a road trip out of it. But we’ll see. And then obviously we went to we went to beach vacation, our family beaches.
And now what’s coming up is we have this outdoor patio which has a nice creek that runs right along our fence. And so, we’re getting ready for a lot of outdoor activities now, birthday parties, family get togethers, all that fun stuff.
Rhett Dodson: Yeah, we get the run of birthday parties in my family. It kicks off with Reece in May, and then we get both Aubree and Caleb. Peyton’s the only one that’s born out of cycle in February.
Christopher De Leon: Ours are towards the end of the year too. So, it works out.
Rhett Dodson: Yeah. And so, you know, it’s interesting because you mentioned birthday parties and I remember you telling me about a birthday party you went to because it’s going to tie into this episode.
Christopher De Leon: Yeah.
Rhett Dodson: You guys had like the birthday party on top of the right of way for a pipeline. I remember you sending me a picture and, like, there’s tents up in the background, there’s kids running around, and there was the pipeline right away, marker right there. You know, like a little warning high pressure pipeline underneath you.
Christopher De Leon: As a pipeline integrity engineer and former pipeline operator, it was incredible to see how the home had integrated with the pipeline right away. And as you
Rhett Dodson: Knowingly or unknowingly integrated with the pipeline right away?
Christopher De Leon: So that’s a great question. So, I started taking pictures with the pipeline marker and like doing a certain like angle so you could see how the party and our friend’s yard literally is integrated with the right of way. Right. And so, I tried to get an angle picture where I’m using my phone, taking a selfie or an “Us-ie” if you watch Ted Lasso. And I mean, you can totally see how there’s like this entire setup. And if you could see two pipeline markers in the picture, one of them which the family had built a brick structure around the pipeline marker where the CP station was with full access. So, it didn’t it didn’t eliminate the ability to do maintenance and monitoring, but they wanted to pretty it up a little bit because it was such an integral part to their front yard. And it dawned on me.
So, I went and I spoke to our friend who also in the party and said, “Hey, do you know what that is?”
And she’s like, “You know, I never really thought about it, but I know my dad, you know, built the whole little like, casing around it, so it didn’t look so bad.”
I was like, “Yeah, you have a high-pressure pipeline there.”
She’s like, “Is that a big deal?” And I’m like, this is fantastic.
And I’m like, “How long have you lived here?”
She goes, “We’ve lived here our whole lives.” And that’s kind of what you want, right Rhett?
Rhett Dodson: I mean, that is look 100% in the eye, a stone’s throw from my house, I probably could throw a baseball and hit it. I have the Tennessee gas pipeline system, the TV system. All three of those lines come through. And so, every time I pull it in my subdivision, I can see all three of the markers right there. It runs right behind the subdivision. And I’m always pointing out to my wife and she’s like, “Whatever, I don’t know what any of that is, and I’m not really interested,” which is, I think to some extent, how you want the public. That’s how you want it. They’re not aware that there is a high pressure either gas or liquid pipeline and they feel, let’s say, safe enough to host a birthday party on top of it, routinely drive it’s –
Christopher De Leon: Safe enough or ignorant enough?
Rhett Dodson: I mean, ultimately, I think that’s the question, right? You know, I think it’s fair to say that maybe the unawareness and ignorance go hand-in-hand to some extent. Right. I mean, the pipeline company obviously makes them aware because you’ve got the marker there. And so, the pipeline company does its job, but they don’t care enough because they don’t have to interact with it on a daily basis, which is exactly how I do believe we want the public.
Christopher De Leon: Yeah, And I mean, these interactions happen in all kinds of ways, right? You mentioned the TGP system we have and again, it’s all fair because everybody’s doing a good job right now is it’s you know, we have a creek right here next to side of our home. Almost every day my kids say, “Hey, dad, take us out on the creek on the Polaris.” So we open the back gate and we hightail it and we see the pipeline mark right where it goes under the creek. And it’s an enterprise pipeline and it’s properly marked and all that fun stuff. Or, the Walmart down the street, we have a pipeline that literally cuts across the Walmart parking lot, right? And says high pressure gas pipeline. It’s pretty interesting because you see these yellow, the white poles of the yellow signs. I mean, they’re lined up, I mean, all the way through the parking lot. And so, I mean, yeah, these pipelines are fully ingrained in our lives.
Rhett Dodson: And it’s interesting because you mentioned the one with a creek. So that’s going to be the focus of today’s episode. So, know going from there, I think it’s fortunate when we live our lives that way. But today’s accident or today’s report of union doesn’t go that way. So, we’re going to be talking about Bellingham, Washington. I think for some of our liquid operators, it’s going to be very, very familiar, very, very intimate with them. I think our goal is to hit a lot of the younger generation that maybe aren’t familiar.
The reason we chose to start with Bellingham is it’s actually a long time ago in terms of 25 years. So, for a lot of pipeline integrity engineers that are entering our workforce right now they were certainly too young, maybe not even on the planet at an early enough time to understand. So, I’m going to take you through the story.
Christopher De Leon: But I want to set the framework a little bit as well, too. So, I mean, one of the things we’re often say on this podcast is this knowledge transfer is fundamental. And when we first started this episode one, you know, we kind of set the framework and said we want the objective of this podcast to bring up relevant events, knowledge transfer. So that’s why we do things like the PPIM Arc Tour. If you guys didn’t have a chance to go or you don’t get a chance to read the papers, it’s a place for you to go and say, you know, “I didn’t get to go to that, but what are a couple of the takeaways?”
And so, for me, when I look at this stuff, it’s there’s a lot of really good information that we find in some of these historical events where a lot of people put a lot of time and effort into these things. And what we don’t want to happen is one of the things that we heard in the last episode, right, which is it’s you know, it’s a question around are we learning from our mistakes? So, what’s that timeline of how far back we look to learn, right? Is it as recent as it does our does our attention span or our memory only go back to San Bruno, or can we go further behind in this case? We’re talking roughly 24 years between now and when this Bellingham incident happened.
So, again, as we said in the first episode, you know, our goal is to highlight the big takeaways from these right. So what are the things that we can learn from these? And if you guys want to go back and read it, by all means, please do. But the goal is to surface some of these things, like for example, in this one and I’m going to jump a little bit to the cases. It’s like, why are dents with metal loss important? Where did that come from? That’s really the spirit of what we’re trying to do. So, we’re going to try to extract some of that stuff.
Rhett Dodson: So as one of our friends says, the one thing we learn about history is that we don’t learn from history. I think the NTSB reports are they’re actually a good read. They’re good factual information. If you’re interested in what we present here, I’d say you should dig into the NTSB report – any of the ones that we talk about, they are the timelines are well-written. The facts are presented in a way that’s actually, I think, sometimes easy to identify, even yourself there. So, what happened is –
Christopher De Leon: Let’s jump right into the story. Let’s see if you’re good storyteller.
Rhett Dodson: Let’s indeed do so. You have to imagine that it’s June 10th, 1999. I don’t know where you were. June 10th.
Christopher De Leon: I was in high school.
Rhett Dodson: I was a junior in high school as well. Yeah, we’re in Bellingham, Washington, near Whatcom Creek. Right. And so people that are from Washington, it’s not quite like Houston, a little more wooded. This is a very, I would say, a pristine outdoor area. It’s a popular barracks for people to be outdoors, very lush park. I mean, kind of like feels like Rivendell out of Lord of the Rings to some extent when you see it in the key,
Christopher De Leon: I think you said is it’s June, it’s summer.
Rhett Dodson: Yeah. And people are outside and it’s. 3:28 p.m. So, the incidents that surround this particular pipeline failure are when people are out and about. You have kids coming home from school that might be going things after you have people driving home from work in contrast to other failures that might happen in, say, the middle of the middle of the night.
Christopher De Leon: Exactly.
Rhett Dodson: So, what happens is next to what concrete or actually within Bellingham, I really should say. Yeah. There’s a pipeline operated by Olympic Pipeline. It’s a 16-inch line that transports refined products. In this particular case, it’s gasoline. It experiences a rupture and dumps 237,000 gallons into Whatnom Creek and flows into the creek and it ignites.
But the series of events that led to that is really what’s kind of interesting, Chris, because what we see and I’m going to take you through the timeline is you see this, it’s not in Bellingham, it’s not one thing that happened. It is a cascade of things that have happened here. Almost any one of them, if you took out of the picture, wouldn’t happen.
Christopher De Leon: It’s characteristic Swiss cheese model, right?
Rhett Dodson: Yeah, but you don’t always use that in every failure. But in this one, it’s like as you read, you’re like, “oh man, just if that … oh if just that didn’t… that if just it…” You probably wouldn’t have ended up with the precise events at least on this day that we had happened.
So, starting off they’re operating the pipeline, they’re actually switching the product. So, they’re going to be moving the product from one refinery to another refinery. And as they’re doing that, they have to make some changes in the SCADA system at the exact time that they decide to switch over the product, somebody decides to edit some of the SCADA files, some of the historical stuff in SCADA system. That edit apparently sends the SCADA system into kind of a panic mode and it actually freezes for about 5 minutes. They lose control of certain operations of the pipeline going from one location to another. They actually lose the ability to see the data, the backup. They bring the backup system online while they’re trying to find it out. It’s during this precise time that you actually get the overpressure event that causes the failure. But because of what’s going on in the SCADA system, it obscures their ability to see what’s going on moving forward.
They don’t know if the data they’re getting is a function of what happened badly with the SCADA system or something’s actually happened to the pipeline. But that’s when the actual incident happens. From that point on, they move forward with trying to commission operations again. So they’re like, “Hey, we need to recommence this transfer of product back up the thing.”
So, they get the system up and running to get to a point. They feel comfortable, they hit the button and restart the pipeline – fulling not knowing that the pipeline had already ruptured, which then puts more gasoline into the creek. It’s around this time, and it’s we’re not talking significant time differences, from the time that the system shut down, they restarted it 13 minutes later. I mean, only 30 minutes think. But how much happens in 13 minutes? They get a leak alert. So, the system was equipped to have a leak alert at the same exact time that they get the leak alert or nearly, they’re getting a call from one of their employees who’s driving over the bridge and says, “I’m smelling gas. Something is going on here.”
Around the same time that residents start lighting up the 911 call center is when people start to realize. And so, we’re only talking again. The leak happened, I told you at, precise time was 3:28 by 4:30 is whenever they realize that something is going on, the leak alert has happened, the systems start shutting down the pumps and they realize that something has happened and that’s when things go into motion really fast.
So, at this point in time, they start at 4:30, they start sending people out to shut off this area, to cordon it off. But unfortunately, there is so much product in Whatnom Creek in and around the surrounding area, the fumes are so heavy that it’s a matter of time. And so, you get ignition. And ultimately what happens is that you get ignition.
They start getting reports at 5:00. It’s only an hour and a half after the original rupture happened, roughly 30 minutes from when they got the leak alarm and they were able to start putting events together, it ignites. And you burn a mile and a half of the full creek. And for those of you who haven’t seen the pictures of the pictures of what this fireball looked like, it wasn’t like I mean, you got to think of a mile and a half, right? A mile and a half, 7000 feet. That’s a minute and a half in your car. It didn’t just all ignite at once. It ignited starting at a water treatment facility, which we’re going to talk about in a minute, and the plume rivals down the creek. And obviously, as that happens, everybody starts calling 911 at this point reporting what’s going on.
And they put full incident command goes into effect. The actual fires continue to burn for four days. Near the water treatment plant before they were able to get in there and actually fully put them out. And so unfortunately, as a result of this, you probably wondering I told you this this summer it’s 3:30 people are coming home from work. Unfortunately, there were casualties and some of them that would be very close. That’s kind of because you mentioned you were a teenager. I was a teenager the same time.
Christopher De Leon: One was an 18 year old. So, there were three fatalities, three fatal fatalities, two 10 year old boys and an 18 year old male as well. Young adult who lost their lives in the event. So, again, right around 1999, I was in high school right there, and I found myself playing outside in creeks. Ten-year-old’s, my daughter, Gemma, Gemma’s ten and she’s outside this creek by my house all the time. And yeah, I mean, this stuff, this stuff hits close to home. And so it kind of puts a lot of meaning. I mean, I love the soil, right? I mean, I wanted to pause here because this is why we do what we do, guys.
Rhett Dodson: This is what we don’t want to happen. Right. And it’s really interesting because to the two boys actually eventually succumb to burns that they suffered. Right. So, they were airlifted. Yeah. The older gentleman that died succumb to asphyxia and drowned, they believe, as a result. So, it’s kind of even apart from the fireball, which is the really visual devastating effects, just the inflow, even if that hadn’t happened, you still would have had at least one loss of life here. And it really brings home the concept of how important pipelines I doubt any one of those three. I doubt most of the residents, the families of those boys realize how close they had a pipeline to their home, realized they had potential for the pipeline to dump material hydrocarbons into Whatnom Creek, which is the way we go back to say we want to do it.
So, most of the audience now, we told you is the what happened. We haven’t got into the why for most of our pipeline integrity engineers you’re probably wondering, how? How did that happen? What are the events that led to failure. Before we get into that, this is actually where we’re going to take a break and join us after and we’re going to get into the how is it that you end up with a release into the thing? How did this failure happen and what steps even today do we need to take to continue to prevent it?
Music
Rhett Dodson: All right. Welcome back to Pipeline Things where we’re going to pick up our discussion. So, where I left you off is I told you what physically happened at Bellingham, Washington, in 1999 in terms of the events that took place. What we haven’t gotten into now, is what led up to that, how it failed, and the fallout.
Christopher De Leon: Yeah, the downstream effects of it. So, a couple of things. If you are a listener to our podcast, we really encourage you to go to YouTube and watch the Bellingham Disaster video. It’s old school. You get to see a lot of cool cars like Crown Vics and old metro busses and all that fun stuff. It’s a little bit nostalgic, but the story, it really hits home. And we have a customer of ours that has told us that, you know, that used to be kind of part of the rite of passage. When you were a new employee going into the organization, you had to watch this video to kind of ground you a bit into understanding why what we do is important and the duty of care in everything that we do right. And having this mentality of public safety, social license, all that fun stuff. So do us a solid, you know, go to YouTube, google Bellingham disaster, and, you know, take a look and listen. I think that’s really helpful.
One of the other things that was important from this is it’s as part of the, you know, the effects of the Bellingham failure was a judge awarded effectively $4 million that was used to establish the pipeline safety trust.
And the pipeline safety trust, as we said, is a is a direct berth of this incident. And the way they kind of describe themselves or they describe the industry is they’re kind of like a watchdog for pipelines. And so, whenever there’s new regulations or there’s happenings, you know, you can trust that the pipeline that the Pipeline Safety Trust is providing insight always with this idea of public safety and let’s not let this happen again, you know, to
Rhett Dodson: absolutely encourage our listeners to check those videos out. So let’s jump in in those resources out. And it is it is a good video to watch. Let’s jump into the why, obviously, because things don’t start in 1999 in this and I want to encourage your listeners, I think some of you guys again find yourself right.
The tendency is always to say, oh, that’s other people, but take a look. And I think you can find elements of yourself in each of these that we can learn from. So, like most pipelines in the United States, this pipeline was not built in 1999, it was actually built well before that been the particular section that failed was rerouted in 1966.
But that doesn’t have anything to do with today’s story. This story really actually begins in terms of causal factors, in 1991. There was a water treatment plant, Bellingham, Washington only has one water treatment plant, and they were doing a significant expansion of that water treatment plant in 1991. As a result of that, they needed to cross over Olympic pipeline.
Any time you’re crossing over or around the pipeline and exposing it, that’s why we have what we have today. Chris, what number do you always call?
Christopher De Leon: 811.
Rhett Dodson: 811.
Christopher De Leon: Why do we do that again?
Rhett Dodson: So that the pipeline operator can tell you where their pipeline is and make sure that you’re aware of it so that you don’t accidentally hit it.
Christopher De Leon: It’s fundamental and public awareness programs, right? So again, if you’re a young integrity engineer and you work for an operator, you are aware of public awareness, right? If you’re not if you don’t have expensive pipeline operator equal to of equal importance to pipeline integrity plans, you have public awareness plans. Yeah, right. With the whole intent of saying, hey, the public needs to know where we are so that you don’t excavate or do something like build a pool or install a fence.
You don’t hit the pipeline, then don’t tell anyone, cover it up and then that is now a new integrity threat that needs to be managed, right? So, we want… we like 811 and this is exactly what plays out here.
Rhett Dodson: So, what happens is they expand. They have a 72-inch water line multiple PVC pipelines. They have some challenges. And where the water plant wants to put the actual tie in and believe it or not, they’re actually really good communication with Olympic pipeline throughout all of this. Right. An Olympic pipeline actually has records of sending their investigators out there to watch while they’re doing excavations. But because of the challenges and maybe because of a little bit of complicity, who knows?
There are several instances where it’s pretty clear that Olympic pipeline wasn’t out there while their pipeline was exposed. It’s also clear that they had to make changes to where the water pipeline or the water treatment facility needed to dig didn’t let Olympic pipeline know. So Olympic Pipeline was aware of a lot of stuff, it’s pretty clear they weren’t aware of some other stuff.
There are also conflicting reports when they interviewed people, one guy said, “Oh, I remember when they dug it, they hit the pipeline multiple times,” as an electrician and he said, “and they had me go in there and coated with mastic.” They went back. Other people couldn’t corroborate his story. At the end of the day, it doesn’t really matter what actually people said.
What we know happened is that at some point during these excavations in 1991, Olympic pipeline was struck not once but multiple times along approximately, I think it’s a 27-foot stretch. There were multiple dents, multiple gouges, where it was pretty clear that the excavators that were used there, hit the pipeline.
Christopher De Leon: So, pipeline integrity, that threat is what?
Rhett Dodson: Third party damage.
Christopher De Leon: Third party damage. And we have tools to assess the integrity of a pipeline as it relates to third party data.
Rhett Dodson: Absolutely, and believe it or not, that’s where this story gets more interesting, right? Because after that happens in 1991. And in November of 1991, they contract with Tuboscope to perform an inspection, to scope your linalog.
I think you say that did that was a lot Look there when you were around –
Christopher De Leon: I’ll be honest with you. So, my my career was in line inspection. We ran, we ran a lot of Tuboscope based products through line log. But that wasn’t until probably ten years later, almost in 2008.
Rhett Dodson: So, before they dug, they had an original 1991 inspection, with Tuboscope. They ran again in 1996 after the construction of the water treatment plant with a different or with the same Tuboscope tool that they ran here. And then they also ran an MFL inspection as well, right?
Christopher De Leon: So, both of those were MFL, right? So, the 1991 inspection I think was dubbed as a standard resolution MFL. And in 1996 we have reason to believe that it was an improved MFL based technology. And they do highlight the intent that they –
Rhett Dodson: You’re right. Thank you. On the Enduro was an is the caliper tubing scope was in MFL. Enduro, they ran separately a caliber tool following the 1991 improvements to the water treatment plant. Which interesting is the Tuboscope feature found, the Tuboscope inspection, I’m sorry, found multiple features.
Christopher De Leon: The 1996 one, to be clear.
Rhett Dodson: 1996 one found multiple features. They classified one anomaly as possible mil mechanical that was 0.4 inches long with a 23% wall loss. They reported a second feature possible mesh. I don’t. Yeah. They actually don’t even define possible mash here. I’m wondering if that’s like some sort of British term. Yeah. I was like, never heard of that before. And then another feature identified as a possible wrinkle bin. What’s significant about these is none of these features existed. The 1991 inspection prior to the expansion of the water treatment plant.
Christopher De Leon: So big, big takeaway. Right. And again, we have all different level of use for the people that listen to our podcast, right? But the idea here is it’s we were integrating data even back in the nineties, right? So, a lot of the stuff that we should be complete expert at I feel like unfortunately back to you know what are we learning from our lessons learned it’s we still struggle with data integration today.
I mean how often do you get an ILI report where you challenged or requested or made part of your scope that the most recent inspection that you provide the vendor with that data for integration? You know, there’s always this chasm, I feel like, between blind trials to make sure they’re doing a good job versus let’s have this mindset of integrate all the data and make the best decisions possible.
So, when they did that back in 1996, which I think they labeled the person who did the work as an engineering assessment.
Rhett Dodson: So, it’s good because I didn’t finish. That was the – you you’re referring to the to the scope inspection. There was an enduro inspection that also called a caliper related deformation.
Christopher De Leon: Yeah. What I was going to build there is it’s so they were doing data integration back then, right guys. And so, what they did with this engineering, this analyst assistant did is the way they called him is it’s he did what we would do today. He takes the 1991 run MFL. He compares it to the reported features. It’s unlikely he did signal analysis because they were using cards in scrolls back then. But cards to compare magnitudes like they draw lines on it and then as they go through the log, they’d see how big the deflection was.
But he realized these are new features.
Rhett Dodson: Yep, he did recognize the features, but they were new features with a pressure ratio of 1.21.
Christopher De Leon: Yeah. Which is above one. Right.
Which back then I think it was be B314 that was used as a governing document. They would use B31G to do [unintelligible] calculations. But the point was this. But they took the right action, guys. Right? So, I mean, none of this stuff is new. So instead of having combo tools, they were assessing particular threats. So, when they record wrinkle bends, possible deflection, what do they do? They go and say, we need a caliber tool. And that’s where Enduro comes into play.
Rhett Dodson: And then they run the caliber tool. As I mentioned to you, Enduro finds similar features not at the exact same long distances. For obvious reasons.
Christopher De Leon: Why is that obvious? I don’t know if that’s obvious.
Why is not every tool’s odometer system is identical, right? They’re not going to record it. You know, if it reports that at 100 feet, one inspection, it’s not going to be exactly 100 feet in the next inspection.
But Enduro reported a 0.45 inch total, sharp V4, sharp defect and same area. So now they have two inspections that are telling them there’s something here. And it begs the question, what actions do they take? What’s interesting from this point is they do begin taking actions on some of the anomalies in the line, but they don’t end up taking specific action here because of, again, of a few oversights.
You mentioned that it was a non-degreed person performing integrity calculus.
Christopher De Leon: Well, we don’t know if they were non-degreed.
Rhett Dodson: No, we didn’t set it in here.
Christopher De Leon: I think there’s two elements that we need to get. We’re looking back and we are we’re sifting through this, right. We’re trying to find lessons learned that we think are applicable. So, as you’re hearing this, you think back at your organization and say, how do I address this? Right? And one of my favorite episodes that we’ve done was during the World Cup arc, and we talked about SMEs and RIN2 and FEMSA has clearly communicated they want to start documenting what tasks or processes and what qualifications you have to make decisions are being a SME.
And so, the question here is one, it was an engineer and an engineering assistant, which again, we’re not judging, we’re raising questions, right? What qualifications were needed? But I think the other thing that was important here, would you have it highlighted here is it’s where were these features were found were in an area facility. It’s in the water treatment facility grounds. And so that’s a great opportunity to start integrating information. So, when we think is integrity management, I would like to bring that component to this. We often think we’re an integrity management and we use risk. And risk is probability of something happening by the probability of or the consequence of that thing. And there’s other metrics we can do, but we’re trying to keep it simple today.
And so that’s when you start integrated is not just was it there or not, but where is it in how can that become relevant?
Rhett Dodson: And he took those steps, which what’s fascinating to me is he took the steps of going look at the alignment sheets. He noted that there were crossings there. Yeah, but the crossings didn’t have the date on the alignment sheet.
So, he incorrectly assumed or didn’t look into when the third party crossings happened and didn’t attribute the change to the crossings. Furthermore, and I think this is what’s really fabulous when you read why he decided not to take action on the fact I mean, even absent he had a short defect. He had multiple call outs. Why didn’t he take action?
He says: “I had lost faith in tubes, scopes, ability to identify deformations because the tool had previously failed to find a buckle that Olympic pipeline had experienced a failure with.”
So, the guy has one bad experience with the tool. And is now biased towards writing off its results. So, he has information in front of him and he says, “Yeah, you know what? These guys missed this back in the day. It’s I’m not going spend time and money here.”
Christopher De Leon: But we don’t see that anymore, though, right? Right. I mean, we learned the lesson, right? We never see them, or we never see where they’re in an ally system. Which is the mandrill or you guys hear me saying this all the time, right? The sensors evaluation process and the people. Right. I mean, how many times when we were at a former employee, do you hear customers say, we need to change our lead analyst? This isn’t working for us, right? I mean, there’s so many components going to play. So, this concept, you bring this up and we talked about that during the semi podcast was biased, right? It’s real. And how do you account for that?
Rhett Dodson: Yeah, how do you recognize whenever you have information that’s warning you? But you’re biased a certain way, again, because you’ve had this particular experience, one bad experience where they got caught with a really bad failure. Yeah. And it didn’t call it.
Christopher De Leon: So, I actually, you know, one of the things that come to mind, and we won’t go down this road too long as it’s you know, API 1163 requires, you know, level one, level two, or level three validation. And in that it says that, you know, you’re supposed to leverage past experiences. And so, you would say, hey, there’s grounds. I have experience either with this vendor or this pipeline or this threat to where I can make a decision. And that technically 1163 allows you to do that. So, one of the things I’d maybe give us some food for thought there is it’s well, that’s true, what it also requires you to do is to maintain a database. With the specific data. With the specific data, you have to have specificity in how you’re going to do this, this disvalidation, independent of it’s a level one.
Rhett Dodson: This is like saying, hey, I got caught up once in traffic on the Hardy Toll Road. I’ve never ever taking the Hardy Toll Road again.
Christopher De Leon: Don’t talk about Hardy Toll Road. That’s how I travel. Exactly. You guys stay on 45 and 99 of the Beltway.
Rhett Dodson: So, I think it’s interesting, but let’s take it a step further. In the end of the day, they do decide to issue a report and an excavation for this feature. And it still doesn’t get taken care of.
Christopher De Leon: Swiss cheese, baby. Swiss cheese.
Rhett Dodson: It’s amazing It doesn’t get taken care of because they go out to dig and they find the area is too wet, too difficult to dig in. So, it gets delayed. We know we were delayed. Yes, it is. It actually never gets dug. It actually falls off the integrity management radar until eventually it fails on its own accord, right?
I think what’s interesting there is again, how many things had to happen. If it hadn’t been wet, if maybe they would have had somebody was a little more proactive. Maybe if they hadn’t had the previous failure on Enduro. Any one of those things, they’re out here and they’re excavating it. But you get all of them together and you get a feature that six years earlier, no sorry, eight years earlier had been struck on the pipeline. You have data six years later that strongly suggests you need to take action. And then three years later it fails on its own. And we end up with, as we mentioned, three casualties and significant changes to pipeline regulation.
So, I did not know this. I found this out, Chris, because we mentioned and I’m curious how many of our viewers know. The requirements, the mandated requirements on debts that came out of this. I’ve always had people ask, why are our dents descriptive? The dent that failed ended up being right at 3%. And so now in the in the federal oversight section of the NTSB report, they write: “Since the accident, the OPS has issued an integrity management rule that added requirements for the evaluation of defects identified in line inspections.”
So, we see OPS, it was OPS at the time, so OPS responded by saying, “Hey, now you will do this.”
And here’s what they changed: “Specifically, an operator will schedule for repair or evaluation within 60 days, all dents, regardless of size, located on the top of the pipeline.”
So, what do you see? We have a huge failure. We have catastrophic consequences. FEMSA says, descriptively every dent top of the pipe liquid pipeline 3%. You’re going to dig it.
Christopher De Leon: And it’s prescriptive. So, what does that mean?
Rhett Dodson: It means that whenever in another scenario where maybe you have proper data integration, maybe you really know the cause of it, you’re now bound to do it.
Christopher De Leon: Yep. We’re going to over dig it out. Have we seen that?
Rhett Dodson: Oh, yeah.
Christopher De Leon: I mean, we see that on both sides, right? This is a liquid line which is more susceptible fatigue. And therefore, if you have deformation, you have stress concentration, you have, you know, initiation of crack, etc., etc., etc. We’re actually, I’ll share we’ll probably to talk about another incident that had motor failure based on that. But today we also see that on the gas side, right. So where, you know, we’ve seen regulation before in two came out that basically states that if you have any dent with any metal loss, you’re going to go dig it. And look, if we go back 24 years ago, that’s what we had here. And so we are really, it really begs the question, is it if we are still being held to processes that are that conservative, you know, it’s a rhetorical question here, you know, what role do we have in that? And so apparently there are regulators feel that they need to continue to manage us with prescriptive regulation to ensure that the public remain safe.
Rhett Dodson: And I mean, in the face of when you go back and in an analysis like this, Chris, it’s almost hard to argue like it really kind of is because you look and you’re like, man, we weren’t good enough at that time, clearly, with what we were doing here.
Christopher De Leon: But things are different now, right? We have we have 1183.
Rhett Dodson: Yeah. So, well, before we do that, I want to leave our audience with Chris and I, I’m going to say glossed over a number of other contributing factors because they’re… It’s harder to get into the weeds of the repair and bypass valve and the roles that those played and some of the other things. The release valve. I would really encourage, you know, again, the audience, if you want more information on that, definitely dig in. But I want to talk about the ones that I feel, you know, again, we say the only thing we learn from history is we don’t learn from it.
Chris, I want to read this. In order to ensure the integrity of a pipeline, the operators should assemble and evaluate all of the pertinent pipeline data available that was written 24 years ago. That is in regard to how you should evaluate in-line inspection data with respect to this failure. Chris, we are still talking about data integration 24 years later.
Christopher De Leon: Yeah, and it’s still not solved. And how many times have we talked to an operator and again, typical consultants, right. Well, we’ll take them to have a meal or spend some time with them on a disc golf course or something. And we say, hey, what’s your biggest challenges? Oh, data integration, right? Our guys are communicating with our database where we’re making integrity decisions.
Rhett Dodson: Our GIS is outdated. It’s three years old, even in this. I mean, you can see now six years is a pretty long time. Yeah, but imagine that this was three years. Somebody that’s siloed doesn’t realize a third-party cross.
Christopher De Leon: And we’re not knocking GIS, right. What we’re saying is, is that 24 years is a long time. We’re still having these issues. The takeaway is data integration is fundamental. We see that all over the place, right? We see a move towards probabilistic risk-based assessments for interpreting where we need to allocate funds and resources. And we also see it now in more modern, we’ll call deterministic models for determining burst pressure calculations of fitness or service. We’re now beginning to see these probabilistic. And you can almost say the basis for that is to kind of start supplementing for some of these unknowns a bit right. If we don’t have all of the disparities data or the discrete data that we need to integrate and make absolute decisions, get it all, either you get it or you start running probabilistics.
Rhett Dodson: So, it’s also interesting too, because some of the you know, like I’ve recently seen this, we’re getting better with the advent of satellite imagery and historically satellite imagery of Google Earth. Man, you know, it’s really easy when I do an SCA now for operators on a dip, I’ll go back and look at all the available satellite imagery which usually goes back to the early nineties to see. And it’s not foolproof.
Christopher De Leon: Yeah, of course.
Rhett Dodson: But you’d be surprised the things we call caught them like, “Oh, this homeowner dug a pond literally adjacent to your pipeline. Right next to it.” So that’s a good resource. You know, again, I hope what people take away from this is data integration was real back then. It was a causal factor here. I want to believe that if the gentleman who had made the decision knew with certainty that those lines were installed between the time the two inspections, I want to believe he would have taken action.
Christopher De Leon: And again, we’re not saying that’s a smoking gun here.
Rhett Dodson: No, it’s not right.
Christopher De Leon: It’s one of the it’s one of these Swiss cheese effects. It’s like the I think you said it earlier. If one of these things maybe would have been done a little bit differently, the outcome could have changed drastically.
Rhett Dodson: Now, let’s hear about one that’s probably closer to your heart. Yeah. And that’s the ILI one. Because it’s really interesting again that he had lost confidence in the ILI technology. So really factually speaking he’s blowing it off because he had a bad experience. And I say “blowing it off” because even when the dig was delayed Chris, he could have been the one to fight. There’s things he could have done, like, “Hey, guys, I put this as a dig. We’re taking care of it. When are we taking care of it? We said we’re rescheduling it. When are we rescheduling it?”
There’s processes, there’s things, I mean, he could have wanted to stay on top of it if he could have wanted to. I’m not blaming it. It’s the saying. What do we see FEMSA responding with? Well, hell, if you’re not going to dig it, you’re going to feel sorry. 60 days, you’re going to be digging all of them.
Christopher De Leon: So, here’s some thoughts. Yeah, I won’t give too many my opinions here, but what I’ll say is it’s what is it that we see or hear from our customers? You’ve always heard me say your consultants are only as good as their network in their experience. And obviously understanding their own bias and saying, “hey, I have strong opinions here. You need to know that before I offer you my advice.” And some of the things that that we’ve heard directly from operators or through the request they give us. One of them is it’s before we sentence a feature. We like a third party to look at it. We have cold eyes, right? We see that at least two of our operators, right?
So that’s one way or one thing that you could think about is it’s you have this process where typically the average staff member that we’ve heard is very busy. There are time constraints, there are resource constraints. We need to make decisions, but there’s only finite resources, so we can’t just go and dig everything, right? So, decisions have to be made.
Maybe you think about, well, what’s the role of a third party there? Right? And I’m not that’s not a plug for us. A third party could even be somebody outside of your normal process that has to come in and take a look.
Rhett Dodson: We talked about that in the SMB. In the SMB episode, going back to there is a role for third parties to play, maybe in terms of identifying where you had bias. Yeah, you know, you ask my impressions of that, I feel unfortunately sometimes like operators line up behind ILI vendors, kind of like they line up on college football teams, right? And they’re really forgiving of the college football teams that they like and they will penalize the heck out of the college football teams that they hate.
The truth is, I think when and until you fully discredit an inspection and a robust and thorough process, you should consider anything it does as legitimate.
Christopher De Leon: So, the other thing we hear from our clients is we hear them say, or you know, I’ll ask some of this informal question. I’ll say, “well, have you looked at the signal data?”
And they say, “No, because I’m not qualified,” then I would say that that’s actually a very valid statement. I actually really like that. You know, 1163 tells us that ILIPQ should have processes in place to qualify your analysts. But that doesn’t keep you from interrogating the process by which the call was made and looking at the data to see if it passes the smell test.
Are there other reference features, right? Questions you can ask are: how was your ILI specification qualified for this threat? So, in this case, when Enduro calls a sharp geometric flaw, call them up and say, “Hey, I’m not an analyst, but I want you to walk me through this, why do you believe…?” Get confidence in the process, right?
Have the analyst talk you through it, look upstream, downstream. Are there other features like this back to where it’s you can use historic information? Hey, Enduro, do you have other features or pool tests that give me confidence that I should believe in this to get me past my bias of last time he did this he didn’t do a good job. How do I qualify it now?
So that’s one of the takeaways. And I would say a third one that I would bring in is it’s for more modern, right? That’s looking back. There’s a lot of information out there that’s been published since then. Right? We’re talking 24 years since then. There’s a lot of IPC papers. There’s a lot of PPIM papers. Call your SMEs. You don’t have to always call them for a solution. Call them and say, “Hey, I’m dealing with this scenario. Are there any good industry works out there that I can review that can maybe shine some light on what we could have or have learned about this, whether it be an RP, a standard, an IPC paper, etc.?”
Rhett Dodson: Yeah, absolutely. So, and again, I think, you know, if there’s one jewel maybe that came out of here that is kind of nice, they did come in that the emergency environmental response to the release was effective and well-managed. So, we at least got that. I do think that’s one thing that that response was good on the part of the pipeline operators.
Christopher De Leon: Well, yeah, I mean, if you watch the video, I mean, you see everybody mobilize. They knew something was coming, right. But at that point, it’s what’s the what’s the likelihood…
Rhett Dodson: We wish we didn’t need to be good at that. That’s the key. We wish we didn’t need to be good. And you could have limited the consequences. But again, that happens.
What you take from here, and I want people to know is if you’re in the pipeline industry, this is the first I’m going to say maybe in our lifetime. I recently our biggest fear is that many of us that to creation of the pipeline safety trust, you see influence of prescriptive regulation coming down. But this isn’t the only one. So, I’ll put a plug going next.
Christopher De Leon: Well, it’s also kind of what we talked about in the last episode, right, is it’s the NTSB findings or those scenarios or cases where the NTSB get involved often end up influencing regulation one way or another. And this is just one of those first or groundbreaking, you know, path setting NTSB investigations where again, you see mandates afterward by either OPS or FEMSA or Congress saying we need to make this shift.
Rhett Dodson: Yep. And that goes back to the when one of us fails, all of us fail because we still, 24 years later, live under the umbrella of what was created by these regulations.
Christopher De Leon: So, to wrap up the threat here. Right. So, this threat was third party mechanical damage, there were clear signs of construction related events, dents with metal loss, sharp dents. ILI data even as back as almost 25 years ago identifying this stuff. We have better processes. You guys can look at API RP 1183 There’s a lot of industry papers out there. We even have a podcast on this stuff. So, you guys can go back and look at that.
Rhett Dodson: So, yeah, Chris, you’re correct. The threat was mechanical damage by a third party. That is, I think, the cause or one of the causes that we would attribute, at least from the pipeline –
Christopher De Leon: From an integrity perspective.
Rhett Dodson: From an integrity perspective. Yes, absolutely. And I think, you know, again, the takeaway is, is what did this threat is still relevant today?
I do think we’re much better today with dealing with a third party, not third-party damage… We’re much better at dealing with the threat of mechanical damage as it pertains to dents. We know a lot more than we did back then. We have things like an ECA process that allow us to handle non injurious dents. This would hopefully be the type that would fall out of that very quickly, like you’d identify and be like, no, you need to take action here.
But we are, I think, much better. But you know, again, in this failure report, you see some of those things that we hope we would do continue to improve on such as data integration, such as our own internal bias. So, what we’re going to go next –
Christopher De Leon: And thinking about safety, right? It’s a whole concept of, yeah, the numbers might say something, but that’s only half of risk management or an element of risk management.
The other is consequence, right? And having that safety mentality and obviously just one note is it’s, you know, you said ECA, I mean that is code now. It is, it’s you know under certain circumstances you can invoke this ECA process for an engineering critical assessment process on dents to make good data and engineering based decision on if you need to take it or not.
Rhett Dodson: But it was20, 23 years before we got that.
Christopher De Leon: But we have it. The point is we have it now, right?
Just be careful about using that as a scapegoat versus the having self-integrity, right. Of saying let’s do the right thing and when is the right time to do it or not.
Rhett Dodson: So, if you’re going to hang out with us, I’ll give a plug for our next episode. I’m going to say shortly after this failure, it’s the gas industry’s turn to have an a very unfortunate accident related and things that I’m going to say we’re still talking about today. Non diggable pipeline, internal corrosion, and unfortunately with catastrophic effects again.
So, we hope you’ll join us for our next episode of Pipeline Things as we continue to talk about the notable historical accidents, the NTSB reports that accompany them in the pipeline industry. So, thanks for joining us. We look forward to seeing you again in two weeks.