On the second episode of Pipeline Things Arc 3, Rhett and Chris welcome Matt Romney, the Product Manager for Global Pipeline Integrity at T.D. Williamson. Join them to learn about how T.D. Williamson enhanced its Gouge versus Metal Loss (GvML) Classifier and how it is changing the pipeline industry for the better. Dents with metal loss have been a pain in the pipeline industry for years. Tune into this episode to find out more about how this innovative technology is improving how full systems can be validated, tested, and measured.
Highlights:
- How does T.D. Williamson’s specification change the way that operators can detect and classify dents with metal loss?
- What are the questions you should be asking yourself when in the classification process?
- What technologies are required for the MDS™ platform?
- What does the future look like for T.D. Williamson and its technological advances?
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 the D2I or any of its officials. If you have any questions about this disclaimer, please contact Lina Adams at lina.adams@advmarketing.com.
Rhett:
Alright, on today’s edition of Pipeline Things, we meet with Matt Romney from TDW talking about the Dent with Gouge specification and talk a little bit about his mad disc golf skills on the TDW disc golf course. Or lack thereof, Mr. Romney, next time, we will definitely expect you to have played disc golf by the time we show up on that course.
Chris:
Yeah, we want to see MDSB multi -disc system.
Rhett:
Absolutely. Enjoy the episode. Thank you.
Rhett:
We are continuing our arc at PPIM 23 talking with various technology vendors and today is going to be super exciting episode you’re going to be hearing from Matt Romney with TDW and we’re gonna be talking about their dent with gouge specification and so again welcome to Pipeline Things. I am your host, Rhett Dotson, my co -host Christopher De Leon and our guest, Mr. Matt Romney. What’s your official title? Introduce yourself real quickly.
Matt:
Matt Romney, the product manager for Global Pipeline Integrity at TD Williamson.
Chris:
Nice.
Matt:
From Salt Lake, live in Salt Lake, been Salt Lake for about 10 years.
Rhett”:
Awesome.
Chris:
What was before Salt Lake?
Matt:
So I’m originally from Arizona. I grew up in Northern Arizona.
Chris:
Okay. Excellent. So you went north a little bit?
Matt:
Went quite a bit north.
Rhett:
Yeah. Absolutely. Salt Lake, that’s the office that has the disc golf course, right?
Matt:
We have a disc golf course. That is correct.
Chris:
Do you know we played it?
Matt:
I do know you guys played it. I think I saw a picture of that one.
Chris:
Yes. For the record, I did beat him in that round too. For everybody that listens, I beat him on that round. I’m pretty sure I beat you.
Rhett:
I think that’s under debate. I think we might have to go back and look that up. But Matt, have you ever played it?
Matt:
I have not played our disc golf course. I just haven’t made the time out there, yeah.
Chris:
It’s a good nooner. Like you can get in there, get it done quick, and get in and get it out.
Matt:
All right, all right.
Rhett:
Can just anybody show up and play that disc golf course?
Matt:
So not anybody, I know we get I know we get a lot-
Chris:
If we show up and have a random tournament that’d be frowned upon.
Matt:
That would be frowned upon for sure. That might grow. — You may get some attention from that.
Chris:
Think of the opportunities. It’ll all be pipeliners
Matt:
Maybe we’ve got to talk offline
Chris:
That’s good. Well, we appreciate you joining the podcast. This is something we enjoy having.
Take us away.
Rhett:
Yeah, especially here at PPIM. You know, this is such a great venue you’re gonna kind of feel the energy you want some random people that are gonna walk by maybe take pictures in front of the cutouts, but that’s not why you’re here. You’re here today because we want to talk about the dent with gouge specifications So I’ll catch the audience up briefly that was an IPC publication from September and thought you honestly did a fabulous job in that presentation.
Matt:
Thank you
Rhett:
Really enjoyed it and that’s what we want to focus on and the reason why so we chose different vendors for this technology art and that in my opinion being a dent guy I felt like was kind of groundbreaking which you guys did with that gouge back –
Matt:
I appreciate that
Rhett:
-to be honest feel like it’s long overdue in the industry if I’m being on like we’ve been dealing with dents with metal loss and the pain in the butt that they have been for a very long time and so to see somebody kind of break ground and push I’m gonna say really I think push the industry forward is refreshing so what we want to talk about is talk to the audience is I want to know how you guys got there so if you don’t mind can you show us what I mean Matt how did you guys develop the specifications y ‘all sit around a table one day it was like the gathering of the minds and you’re like you know what we need we need? a gallon specification?
Matt:
Not quite what that’s not quite the way it happens –
Rhett:
It is in a smoky back room right you’re all smoking cigars after you finish playing disc golf okay
Matt:
No I mean it’s a journey right so we the we first you’ve got to start with the technology, the MDS platform, our multiple dataset platform, we leverage off of that so we collect six primary technologies with that platform. That gives us a good basis for data. From there, then it’s really market pool.
Rhett:
So, I want to interrupt you real quick. I’m going to do that a lot, by the way. Chris might do it too. The MDS platform, is the gouge specification only applicable for the MDS platform?
Matt:
It is, and it requires several of the technologies. Specifically, the Spiral MFL, the Axial MFL, and the low -fill technology that are on there. All of those turned out to be pretty, pretty critical. But yeah, so that’s the basis, right? So you’ve got data, you can collect the data. Next is really pool. So, specifically around this specification, a number of years ago, FIMSA, an R&D forum, identified Dense with Metal loss as a challenge and so that was awarded some funding. We were subcontracted to do some testing with that and out of that came an initial classifier. So from there now we’ve got a classifier, we’ve got data, we’ve got a classifier. It’s really about getting more data and that’s really what’s happened in the last couple years.
Chris:
So can we pause a little bit but they’re getting clarification. So when you say a classifier, is that synonymous with POI? So when we think of an ILI system, we’ll think you detect something, but that doesn’t always mean you can identify what it was. When you say classifier, is that more along the lines of POI, we can see this?
Matt:
Yeah, absolutely. Yeah, so we’re talking dent with metal loss, right? So the first question’s gonna be, let’s assume I can see a dent, Can I see metal loss located within a dent? What’s my confidence? That’s really gonna be your detection piece. Once you’ve detected the metal loss, now the question is, what is it, right? Is the metal loss, I’m looking at something that has characteristics of a gouge, which might pose a greater integrity threat to the pipeline, or is it something that maybe is coincidental you know, different assessments can show, aren’t maybe as critical. So that’s really what we’re trying to classify, is or is not a gouge.
Rhett:
So was that an internal classifier, or you were given that to clients at the same time?
Matt:
We were providing, so once we had that classifier, we provided that as a service. As, and really it was in an R &D format, right? We didn’t have enough data to make a statement to provide a specification around that but we had enough data to say look there appears to be some value here there’s you know there there’s an opportunity here to separate these out really that’s what allows us to gather more data and so for the last several years we’ve had the opportunity to work with a number of operators and apply that classifier to their data.
Rhett:
So it’s kind of interesting, you know, and I didn’t realize that. I think it’s really interesting that the, there was research dollars, federal research dollars, right, that kind of went into that. I mean, that, I want to say, I’ve heard very few times where I really feel like I’ve seen the fruits of some of that research, and so it’s really cool that you guys were able to latch onto that and bring, I think again, something much needed to the industry, and see it manifest itself that way. So y’all went from idea, to research dollars, to classifier, to where you’re at now. And I assume the difference between classifier and where you’re at now is like a specification. Is that what the…
Matt:
Yeah. Yeah. So what we were able to do in 2021, we released a performance specification which is based on the data that we’ve been able to collect. So not laboratory data. This is real world anomalies. I believe there’s, you know, we’re approaching somewhere around a thousand different dents that we’ve been able to apply this to. Not all those we got—
Rhett:
Well, that’s a good, robust data set.
Matt:
We got dig feedback on a good portion of those, several, you know, hundreds and hundreds of those that are dent with the gouge, and hundreds of them that are dent without a gouge. And so, based on that data, now we’re able to make a statistical statement, develop and publish a performance specification. So that’s kind of where we’re at now and we’re really excited. I believe it’s the first in the industry. What also comes out of that is metal loss depth sizing. So we’ve gathered enough data, we can now make a statement around our ability to detect, our ability to classify, is or is not a gouge but along with that same data we have enough data to say here’s our sizing depth sizing for metal loss located coincident with the dent again which I don’t think there are a lot of statements out there currently
Rhett:
No I would say I think you’re right I am not aware of anybody that has a gouge — confidently has a gouge specification I will say usually we see POIs, PODs are like maybe 50 % on the POI, but I’m not seeing anybody that says, “Hey, I can confidently identify a dent with gouge with this percent confidence.” So when you talk about sizing for metal loss, is that sizing a metal loss irrespective of whether it’s a gouge or a general metal loss, or is it only for a general metal loss?
Matt:
It’s both, actually. So our specification, because we’re able to make that distinguish between gouge and non -gouge, We have sizing for both and they turn out to be a little bit different, but it’s—
Chris:
I’m not an operator, it’s like I wanna run this thing.
Rhett:
Yeah, I’m telling you, the dip part of me gets excited. The dip part of me does.
Chris:
If somebody runs them that we get to observe, we get to see what’s going on.
—
Chris:
We see what’s going on. You know, I just wanna say this, Matt, congratulations. thank you because I’ll just make a couple of statements I mean we’ve been doing this a long time as well and just some feedback is it’s one I want to echo what Rhett said is it’s not that the FIMSA R&D dollars don’t always provide fruit but this is a very tangible example of how it was a catalyst whether y’all we’re gonna do it or not or not doesn’t matter that’s secondary it was a catalyst that now led to
something that is a gap and I’ll speak to that gap and the gap is a lot of times you’ll hear companies say we use this technology because it can do this and when you look at things like 1163 for example when you’re qualifying the system you need to understand its performance spec. If there’s no performance spec then how do you qualify it? As an example not saying there aren’t ways but what you guys have done is it’s you’ve now said here is a line in the saying that you can begin to build off of it. I just want to give you guys kudos for that cause you’re saying, hey, I can do it and here’s what I can do. And I just want to say, I think that’s fantastic, especially with something like algae.
Matt:
I appreciate that. Yeah, we’re really excited about it obviously as well.
Chris:
That’s awesome.
Matt:
It’s been a long journey. We’ve had a number of people that worked on this for quite a while, curating the data, pulling it together and making sure that we make the right statement. And it’s been, you know, at the expense and through the support of a lot of operators, you know, ultimately they provide us the dig feedback on these features that allow us to go back and say, “Okay, how are we performing? What’s our ability? Where is this classifier performing? Where are its strengths? Where are its weaknesses?” So it’s an exciting time for sure.
Rhett:
So that operator feedback is a really important component in this process, did you find that to be challenging or did you have certain operators you really partnered with that got behind you and helped you or do you feel like you were really having to struggle to get the data that you needed?
Matt:
You know I don’t think it’ll be a surprise you know dense with metal loss, it’s a problem right people face that every day and there’s a lot of opportunities where we’ll run an ILI tool, and there will be a number of them. And so operators often came to us and said, look, we need help. How do we prioritize this? How do we work this list down? So that those opportunities certainly kind of come organically as you work with different operators.
Chris:
And just kind of looking for it. I mean, you guys are providing characteristic data that can now be used in different ways. Right, that differentiation between gallage and metal loss, the fact that you have a spec, again, you can begin to draw a line and say, this is mechanical damage, I can address this this one way, and this is co -located, not interacting, but co -located features, so.
Rhett:
And you know, actually, that’s where I’m gonna insert a brake, cause I wanna pick up talking about 712C and how this is gonna tie into that, cause I think that that specification, and particularly your ability to provide confidence and size and it’s going to play an important component but we’re going to take a break real quick where we hear from Ben and BJ our sponsors on Pipeline Things at PPIM 23.
BJ Lowe here at the Clarion Technical Conferences and if it’s February it must be PPIM. We’re here that’s right for the 35th time in Houston, Texas starting in 1989 Here we are in 2023 with another record breaking turnout for the world’s largest pipeline technology related event. We’ve got almost 3 ,500 people coming for four or five days, almost 200 companies exhibiting on the trade show floor, another sell out. It’s not too early to begin thinking about 2024. We sell out around June or July, so clarion.org is where you go to sign up and we hope you can join us for the next conference if you’re not here this year.
Another big part of the conference is our training and education program, which takes place on the Monday and Tuesday of the week, and this year we have eight really really great courses, world -class instructors, and almost 300 people showing up for the specialized courses on different aspects of pipeline integrity technology. I’m here with Ben Strowman, my colleague at Clarion. Hey, Ben. What’s up?
Hey, doing good. I’m always happy to be here because it’s, as we say, PPIM season. It’s February. So I would like to remind everybody that we also have a very specialized technical conference that’s associated with PPIM and this year I’m proud to say that we have the largest conference program that we’ve ever had. It has 94 technical papers that were chosen from a large selection of submissions that were very high quality, so we’re very pleased to be putting on these 94 papers during the week. I’d also like to mention that we do have online training if you aren’t able to make it to PPIM each year. We Present online training throughout the year. You can check more of that out at clarion .org and We hope to see you in one way or another online or in person and if you’re not here at the 35th version of PPIM I Hope you’re here at the 36th. Definitely
Rhett:
Welcome back. We are here with our guest Matt Romney from TDW Picking up where we left off talking about the Dentwood gouge specifications. It’s been a great conversation so far. So right before the break, Christopher, you alluded to how operators would use this data. I want to tie it very directly to the updated gas rule. So in 192.712C, they released the new ECA requirements for dents. I’m not sure how familiar you are with them. I assume you’re probably somewhat familiar?
Matt:
Yeah.
Rhett:
That had been a terrible question, by the way. If I’d asked You’d have been like I have no idea yeah, yeah, I’d have been like oh I failed as a host –
Chris:
That’s your job.
Rhett:
Yeah, I should have cleared this question beforehand. You supposed to ask the question
Matt:
192. What’s that?
Rhett:
I don’t know
Chris:
192 galges no no no no no no no Change the threshold
Rhett:
So that 192 712 that new you see a process. How do you see this fitting in there? You see a lot of demand coming you see operators asking for it because of 712 or what do you say?
Matt:
Yeah, I hope so right so 712 talks about the operator’s responsibility to understand not just the dent anomaly, but the features that are in proximity, right? I think it specifically talks corrosion crack cracking those sort of things so I think understanding, you know that metal loss feature that’s in co-located with your dent, whether a gouge or not a gouge, it’s going to become very valuable, right? As operators put together their plans on how they’re going to manage dents, the ability to say look I’ve got a dent with a gouge and I’ve got this mitigative action that I’m going to take, or I’ve got a dent where I have confidence that it’s not a gouge. I’ve got some other metal loss feature, Maybe I have a dent with metal loss, but I’ve got confidence it’s not actually a gouge, and so I’m gonna take a different action. I think the specification enables that as a defendable position.
Rhett:
And I completely agree. I think one of, in there, I think the thing that really catches on to me is when you, in 712C, I think it’s like paragraph eight, if I am not mistake, and says that the operator has to account for tool sizing tolerances and other things. I think as operators try to employ that ECA process from 712 and you need some real meat for the sizing tollerances that is when that specificication is going to become so important.
Chris:
It is an enabler to do it right? It is an enabler.
Rhett:
Right, its an enabler and I would even say that there is a chance of even setting presidence. Again, I don’t have enough positive things to say. I really hope that is when you guys continue to push industry forward and other people catch on. I do have –
Chris:
I know exactly what some of the integrity managers are thinking. They’re like, where was this during baseline assessments?
Matt:
Yeah, yeah.
Chris:
It takes time, right?
Matt:
It takes time.
Chris:
And they’ve done it. You’ve got the feedback you’ve needed. My perspective is it’s, you know, industry’s giving you guys the feedback and you know the operators that you’ve been able to work with to not just do that small scale testing, you know to establish a spec but to me it sounds like on the magnitude that you’re talking about I mean that’s really more like validation right I mean you can use historical information that these guys have given you to say this is why we feel comfortable so it almost feels like it’s been validated you know it’s an operator should be able to use that under certain circumstances so just you guys just keep bringing value to this game
Matt:
Yeah I mean you know one end of the big the benefits that I see is often in laboratory, you’re focused on the performance of, let’s say, the science of the classifier. The way that we approach this, the way that it really played out, we leveraged the entire system. So we’re measuring our performance, not just on how this algorithm works and doesn’t work in a laboratory environment, but how it’s applied and utilized through the data science organizations. So the full system’s really being validated, tested, and measured, and that’s what we’re able to—
Rhett:
I got a question.
Chris:
Make it real for us.
Rhet:
I’m gonna do what I know a lot of operators are gonna do. — Hold on. I didn’t run your MDS tool. I ran something else. Can you give me the baby specification? Do you guys see trying to push that bound or you just don’t think that’s ever really gonna happen?
Matt:
– No, so the—
Rhett:
– Cause you know, the operator’s always trying to push it. I know you haven’t already gotten that. I know it’s coming.
Matt:
Yeah, no, we get the question quite a bit.
Rhett:
I just asked it for all the viewers on the thing. I’m like, I know somebody is asking this right now.
Matt:
But, no. And the reason being is we’re, you know, the development of the algorithm, we leveraged the machine learning modeling, you know. It’s not, you know, it’s not super extensive, but we’re feeding in due to it from the data that we have, from the MDS data, in that process you identify what’s critical. Turns out a lot of the MDS technologies and the amplitudes and the way that those waves are produced are pretty critical. The SMFL data is leveraged pretty heavily, the axial MFL data, and actually the low -fill technology. It plays a big role in really into that classifier is or is not a gap. So not able to apply it to something that we don’t have that data set.
Rhett:
Thanks for the explanation.
Chris:
I feel like we should address that a little bit and everybody’s technology is different so I’ll be very broad right so for the listeners is it’s if you’re saying well hold on well I’ve got two of those three right so maybe I don’t have the SMFL or maybe I don’t have the low field why does that matter again what you want to think about is not just the integration but also differences right when you’re thinking about classifying you’re trying to say it’s this and not that right and so those other technologies are really what are allowing you guys to say this is why we have confidence that it’s this i.e. gouging versus not so it’s not just a one system not just one technology but rather the system as a whole yeah so if you guys if they figured it out that way use them that way that’s what we would say right you have a solution get the solution
Matt:
So what I will say is we didn’t make a hardware change with this so the MDS tool we ran last year we can’t apply it so
Chris:
They brought back the residuals so no no no
Rhett:
You tell me I can go back to like 2010 and take my MDS data
Matt:
If you believe that data is still relevant to your pipe condition today the tool the tool It didn’t change. This is not a hardware change.
Chris:
Didn’t the first MDS have a residual field versus a low?
Matt:
Yeah.
Chris:
-So you kept residual, but now it’s value.
Matt:
So we still have residual on two of our technologies. And it applies the same way. The way that that feeds in, whether it’s our low field technology or the residual technology, they both apply-
Chris:
– So actually, that’s a good technical point. Does the spec apply to both the residual and the low -field or just the low -field?
Matt:
It applies to both.
Rhett:
He got really excited about that match. He wanted to make that point.
Chris:
That’s so good.
Rhett:
You need to settle down.
Chris:
Will you make this practical for us? So I got a question for you.
Rhett:
Make this practical.
Chris:
Make this practical. So if an operator ran the MDS and they find that under the prescriptive measure they may have to respond to a feature, they did with co -located metal loss and they ran the MDS. How do you see an operator using this as it relates to ECA?
Rhett:
Right, so as you walk down the process, I mean, first step, you’re gonna run the strains and find out whether or not you’re gonna be able to fall within that process, right? So if you pass the strain limits, then you’re gonna kick into the ECA process and you’re gonna have to redefine your reassessment interval. And in doing that, that’s where it’s gonna come really critical, so you’re gonna have to identify, it says to review all of your high resolution data, you have to identify the threats that are associated with it, then you’re going to have to assess them. And it’s really when you get into that assessment part where I think it’s going to matter the most because if you’re going to assess a dent with metal loss, that’s actually going to be a little bit different than a dent with gouging. But in both of them, you’re going to need a sizing specification because then in the back half you’re going to have to assess more than likely conservative estimates of the metal loss sizing. I think where TDW really set themselves apart with this and again I keep and I feel like I’m just like — I’m excited because this is what this is what happens when you don’t I have a dent with metal loss and the operator’s like okay I want to analyze that what’s the size of the metal loss I’m sorry I can’t tell you what it is well wait I have a dent with metal yeah I just can’t tell you what size it is well what do you mean and I’ve run into situations where it’s a below spec metal loss and a below spec dent I run into situations where if the dent box hadn’t been drawn, the metal off signature wouldn’t have been called. And I’m like, wait, so the only reason you called this metal off is because there was another box here? Well, yeah, I wanted to help the operator. You did not help the operator. You might have just cost them a whole lot of money for something that’s not significant.
Chris:
Benign, yeah.
Rhett:
Benign, yes. This, for me, really starts to turn that on its head because now it’s dent with metal loss. And here’s the information you need to complete the picture and confidently complete the picture. That’s big.
Chris:
I feel like we get it.
Rhett:
We do. Yeah, let’s talk about the future, right? So we’ve got a couple minutes left. Where do you guys see this going? I mean, is this it? You’ve done with it or how do you see this going? No, I wish. It’ll be continuous improvement, right? So we’ll continue. I hope operators leverage it, continue to leverage it. We’ll continue to gather data, re -evaluate the specification, and continue to improve it. We’ll look for other opportunities, other ways to apply other critical dimensions or data sets to further refine the spec. Obviously, dent with metal loss isn’t the only challenge out there, so there’s other ways, other processes. One of the things I’ll be talking about at PPIM this week is our Selective Seamworld Corrosion classifier.
Rhett:
So are you going to be meeting with us again next year to do the same thing about Selective Seamworld Corrosion?
Matt:
We’ll have to see, but you know, it’s the same story, right?
Rhett:
It all depends on how many ratings the show gets.
Chris:
No, no, you know what I was thinking, how many people in the next three weeks from when this airs, they get called and say, “Hey, can you pull that 2019 date and do this analysis for me?”
Matt:
Yeah, that’s right.
Rhett:
For our audience, we get absolutely no returns from anything. We have not monetized the podcast. There’s nothing here for us. This is just for fun. TDW paid us no money, and we are paying TDW no money to appear in the podcast.
Matt:
All those disclaimers, no. So it’s continuing to leverage the technologies we have to solve problems, and that’s always fun.
Rhett:
And you’re not done.
Matt:
We’re not done. That’s fun.
Chris:
We need to go play disc golf again over there.
Rhett:
Are you gonna play with us next time we come?
Matt:
If you guys come out there, I’ll see if I can find a way.
Rhett:
No, no, I need to know if you’ll come throw a hole with us.
Matt:
I will throw a hole.
Chris:
We’re gonna get the one over there by the intersection.
Rhett:
The one that’s right next to the street so you can throw into the hole.
Matt:
I’ll hit cars. I know I’ll hit cars. I haven’t played that course I have played disc golf I’m not that good.
Chris:
It’s okay it’s about having fun. If you haven’t figured this out it’s about having fun.
Matt:
I get it.
Rhett:
It’s fantastic. All right Matt I want to say thanks for joining us on the show today we really appreciate it.
Matt:
Appreciate the invitation this has been fun.
Rhett:
It’s been fun I hope we do it again. To our audience out there I hope you guys enjoyed this episode in the ARC where we’re talking to vendors at PPIM23. I’m your host is my co -host, Christopher De Leon, and we will see you again in two weeks.
Thank you!
Alright, on today’s edition of Pipeline Things, we meet with Matt Romney from TDW talking about the Dent with Gouge specification and talk a little bit about his mad disc golf skills on the TDW disc golf course. Or lack thereof, Mr. Romney, next time, we will definitely expect you to have played disc golf by the time we show up on that course.
Chris:
Yeah, we want to see MDSB multi -disc system.
Rhett:
Absolutely. Enjoy the episode. Thank you.
Rhett:
We are continuing our arc at PPIM 23 talking with various technology vendors and today is going to be super exciting episode you’re going to be hearing from Matt Romney with TDW and we’re gonna be talking about their dent with gouge specification and so again welcome to Pipeline Things. I am your host, Rhett Dotson, my co -host Christopher De Leon and our guest, Mr. Matt Romney. What’s your official title? Introduce yourself real quickly.
Matt:
Matt Romney, the product manager for Global Pipeline Integrity at TD Williamson.
Chris:
Nice.
Matt:
From Salt Lake, live in Salt Lake, been Salt Lake for about 10 years.
Rhett”:
Awesome.
Chris:
What was before Salt Lake?
Matt:
So I’m originally from Arizona. I grew up in Northern Arizona.
Chris:
Okay. Excellent. So you went north a little bit?
Matt:
Went quite a bit north.
Rhett:
Yeah. Absolutely. Salt Lake, that’s the office that has the disc golf course, right?
Matt:
We have a disc golf course. That is correct.
Chris:
Do you know we played it?
Matt:
I do know you guys played it. I think I saw a picture of that one.
Chris:
Yes. For the record, I did beat him in that round too. For everybody that listens, I beat him on that round. I’m pretty sure I beat you.
Rhett:
I think that’s under debate. I think we might have to go back and look that up. But Matt, have you ever played it?
Matt:
I have not played our disc golf course. I just haven’t made the time out there, yeah.
Chris:
It’s a good nooner. Like you can get in there, get it done quick, and get in and get it out.
Matt:
All right, all right.
Rhett:
Can just anybody show up and play that disc golf course?
Matt:
So not anybody, I know we get I know we get a lot-
Chris:
If we show up and have a random tournament that’d be frowned upon.
Matt:
That would be frowned upon for sure. That might grow. — You may get some attention from that.
Chris:
Think of the opportunities. It’ll all be pipeliners
Matt:
Maybe we’ve got to talk offline
Chris:
That’s good. Well, we appreciate you joining the podcast. This is something we enjoy having.
Take us away.
Rhett:
Yeah, especially here at PPIM. You know, this is such a great venue you’re gonna kind of feel the energy you want some random people that are gonna walk by maybe take pictures in front of the cutouts, but that’s not why you’re here. You’re here today because we want to talk about the dent with gouge specifications So I’ll catch the audience up briefly that was an IPC publication from September and thought you honestly did a fabulous job in that presentation.
Matt:
Thank you
Rhett:
Really enjoyed it and that’s what we want to focus on and the reason why so we chose different vendors for this technology art and that in my opinion being a dent guy I felt like was kind of groundbreaking which you guys did with that gouge back –
Matt:
I appreciate that
Rhett:
-to be honest feel like it’s long overdue in the industry if I’m being on like we’ve been dealing with dents with metal loss and the pain in the butt that they have been for a very long time and so to see somebody kind of break ground and push I’m gonna say really I think push the industry forward is refreshing so what we want to talk about is talk to the audience is I want to know how you guys got there so if you don’t mind can you show us what I mean Matt how did you guys develop the specifications y ‘all sit around a table one day it was like the gathering of the minds and you’re like you know what we need we need? a gallon specification?
Matt:
Not quite what that’s not quite the way it happens –
Rhett:
It is in a smoky back room right you’re all smoking cigars after you finish playing disc golf okay
Matt:
No I mean it’s a journey right so we the we first you’ve got to start with the technology, the MDS platform, our multiple dataset platform, we leverage off of that so we collect six primary technologies with that platform. That gives us a good basis for data. From there, then it’s really market pool.
Rhett:
So, I want to interrupt you real quick. I’m going to do that a lot, by the way. Chris might do it too. The MDS platform, is the gouge specification only applicable for the MDS platform?
Matt:
It is, and it requires several of the technologies. Specifically, the Spiral MFL, the Axial MFL, and the low -fill technology that are on there. All of those turned out to be pretty, pretty critical. But yeah, so that’s the basis, right? So you’ve got data, you can collect the data. Next is really pool. So, specifically around this specification, a number of years ago, FIMSA, an R&D forum, identified Dense with Metal loss as a challenge and so that was awarded some funding. We were subcontracted to do some testing with that and out of that came an initial classifier. So from there now we’ve got a classifier, we’ve got data, we’ve got a classifier. It’s really about getting more data and that’s really what’s happened in the last couple years.
Chris:
So can we pause a little bit but they’re getting clarification. So when you say a classifier, is that synonymous with POI? So when we think of an ILI system, we’ll think you detect something, but that doesn’t always mean you can identify what it was. When you say classifier, is that more along the lines of POI, we can see this?
Matt:
Yeah, absolutely. Yeah, so we’re talking dent with metal loss, right? So the first question’s gonna be, let’s assume I can see a dent, Can I see metal loss located within a dent? What’s my confidence? That’s really gonna be your detection piece. Once you’ve detected the metal loss, now the question is, what is it, right? Is the metal loss, I’m looking at something that has characteristics of a gouge, which might pose a greater integrity threat to the pipeline, or is it something that maybe is coincidental you know, different assessments can show, aren’t maybe as critical. So that’s really what we’re trying to classify, is or is not a gouge.
Rhett:
So was that an internal classifier, or you were given that to clients at the same time?
Matt:
We were providing, so once we had that classifier, we provided that as a service. As, and really it was in an R &D format, right? We didn’t have enough data to make a statement to provide a specification around that but we had enough data to say look there appears to be some value here there’s you know there there’s an opportunity here to separate these out really that’s what allows us to gather more data and so for the last several years we’ve had the opportunity to work with a number of operators and apply that classifier to their data.
Rhett:
So it’s kind of interesting, you know, and I didn’t realize that. I think it’s really interesting that the, there was research dollars, federal research dollars, right, that kind of went into that. I mean, that, I want to say, I’ve heard very few times where I really feel like I’ve seen the fruits of some of that research, and so it’s really cool that you guys were able to latch onto that and bring, I think again, something much needed to the industry, and see it manifest itself that way. So y’all went from idea, to research dollars, to classifier, to where you’re at now. And I assume the difference between classifier and where you’re at now is like a specification. Is that what the…
Matt:
Yeah. Yeah. So what we were able to do in 2021, we released a performance specification which is based on the data that we’ve been able to collect. So not laboratory data. This is real world anomalies. I believe there’s, you know, we’re approaching somewhere around a thousand different dents that we’ve been able to apply this to. Not all those we got—
Rhett:
Well, that’s a good, robust data set.
Matt:
We got dig feedback on a good portion of those, several, you know, hundreds and hundreds of those that are dent with the gouge, and hundreds of them that are dent without a gouge. And so, based on that data, now we’re able to make a statistical statement, develop and publish a performance specification. So that’s kind of where we’re at now and we’re really excited. I believe it’s the first in the industry. What also comes out of that is metal loss depth sizing. So we’ve gathered enough data, we can now make a statement around our ability to detect, our ability to classify, is or is not a gouge but along with that same data we have enough data to say here’s our sizing depth sizing for metal loss located coincident with the dent again which I don’t think there are a lot of statements out there currently
Rhett:
No I would say I think you’re right I am not aware of anybody that has a gouge — confidently has a gouge specification I will say usually we see POIs, PODs are like maybe 50 % on the POI, but I’m not seeing anybody that says, “Hey, I can confidently identify a dent with gouge with this percent confidence.” So when you talk about sizing for metal loss, is that sizing a metal loss irrespective of whether it’s a gouge or a general metal loss, or is it only for a general metal loss?
Matt:
It’s both, actually. So our specification, because we’re able to make that distinguish between gouge and non -gouge, We have sizing for both and they turn out to be a little bit different, but it’s—
Chris:
I’m not an operator, it’s like I wanna run this thing.
Rhett:
Yeah, I’m telling you, the dip part of me gets excited. The dip part of me does.
Chris:
If somebody runs them that we get to observe, we get to see what’s going on.
—
Chris:
We see what’s going on. You know, I just wanna say this, Matt, congratulations. thank you because I’ll just make a couple of statements I mean we’ve been doing this a long time as well and just some feedback is it’s one I want to echo what Rhett said is it’s not that the FIMSA R&D dollars don’t always provide fruit but this is a very tangible example of how it was a catalyst whether y’all we’re gonna do it or not or not doesn’t matter that’s secondary it was a catalyst that now led to
something that is a gap and I’ll speak to that gap and the gap is a lot of times you’ll hear companies say we use this technology because it can do this and when you look at things like 1163 for example when you’re qualifying the system you need to understand its performance spec. If there’s no performance spec then how do you qualify it? As an example not saying there aren’t ways but what you guys have done is it’s you’ve now said here is a line in the saying that you can begin to build off of it. I just want to give you guys kudos for that cause you’re saying, hey, I can do it and here’s what I can do. And I just want to say, I think that’s fantastic, especially with something like algae.
Matt:
I appreciate that. Yeah, we’re really excited about it obviously as well.
Chris:
That’s awesome.
Matt:
It’s been a long journey. We’ve had a number of people that worked on this for quite a while, curating the data, pulling it together and making sure that we make the right statement. And it’s been, you know, at the expense and through the support of a lot of operators, you know, ultimately they provide us the dig feedback on these features that allow us to go back and say, “Okay, how are we performing? What’s our ability? Where is this classifier performing? Where are its strengths? Where are its weaknesses?” So it’s an exciting time for sure.
Rhett:
So that operator feedback is a really important component in this process, did you find that to be challenging or did you have certain operators you really partnered with that got behind you and helped you or do you feel like you were really having to struggle to get the data that you needed?
Matt:
You know I don’t think it’ll be a surprise you know dense with metal loss, it’s a problem right people face that every day and there’s a lot of opportunities where we’ll run an ILI tool, and there will be a number of them. And so operators often came to us and said, look, we need help. How do we prioritize this? How do we work this list down? So that those opportunities certainly kind of come organically as you work with different operators.
Chris:
And just kind of looking for it. I mean, you guys are providing characteristic data that can now be used in different ways. Right, that differentiation between gallage and metal loss, the fact that you have a spec, again, you can begin to draw a line and say, this is mechanical damage, I can address this this one way, and this is co -located, not interacting, but co -located features, so.
Rhett:
And you know, actually, that’s where I’m gonna insert a brake, cause I wanna pick up talking about 712C and how this is gonna tie into that, cause I think that that specification, and particularly your ability to provide confidence and size and it’s going to play an important component but we’re going to take a break real quick where we hear from Ben and BJ our sponsors on Pipeline Things at PPIM 23.
BJ Lowe here at the Clarion Technical Conferences and if it’s February it must be PPIM. We’re here that’s right for the 35th time in Houston, Texas starting in 1989 Here we are in 2023 with another record breaking turnout for the world’s largest pipeline technology related event. We’ve got almost 3 ,500 people coming for four or five days, almost 200 companies exhibiting on the trade show floor, another sell out. It’s not too early to begin thinking about 2024. We sell out around June or July, so clarion.org is where you go to sign up and we hope you can join us for the next conference if you’re not here this year.
Another big part of the conference is our training and education program, which takes place on the Monday and Tuesday of the week, and this year we have eight really really great courses, world -class instructors, and almost 300 people showing up for the specialized courses on different aspects of pipeline integrity technology. I’m here with Ben Strowman, my colleague at Clarion. Hey, Ben. What’s up?
Hey, doing good. I’m always happy to be here because it’s, as we say, PPIM season. It’s February. So I would like to remind everybody that we also have a very specialized technical conference that’s associated with PPIM and this year I’m proud to say that we have the largest conference program that we’ve ever had. It has 94 technical papers that were chosen from a large selection of submissions that were very high quality, so we’re very pleased to be putting on these 94 papers during the week. I’d also like to mention that we do have online training if you aren’t able to make it to PPIM each year. We Present online training throughout the year. You can check more of that out at clarion .org and We hope to see you in one way or another online or in person and if you’re not here at the 35th version of PPIM I Hope you’re here at the 36th. Definitely
Rhett:
Welcome back. We are here with our guest Matt Romney from TDW Picking up where we left off talking about the Dentwood gouge specifications. It’s been a great conversation so far. So right before the break, Christopher, you alluded to how operators would use this data. I want to tie it very directly to the updated gas rule. So in 192.712C, they released the new ECA requirements for dents. I’m not sure how familiar you are with them. I assume you’re probably somewhat familiar?
Matt:
Yeah.
Rhett:
That had been a terrible question, by the way. If I’d asked You’d have been like I have no idea yeah, yeah, I’d have been like oh I failed as a host –
Chris:
That’s your job.
Rhett:
Yeah, I should have cleared this question beforehand. You supposed to ask the question
Matt:
192. What’s that?
Rhett:
I don’t know
Chris:
192 galges no no no no no no no Change the threshold
Rhett:
So that 192 712 that new you see a process. How do you see this fitting in there? You see a lot of demand coming you see operators asking for it because of 712 or what do you say?
Matt:
Yeah, I hope so right so 712 talks about the operator’s responsibility to understand not just the dent anomaly, but the features that are in proximity, right? I think it specifically talks corrosion crack cracking those sort of things so I think understanding, you know that metal loss feature that’s in co-located with your dent, whether a gouge or not a gouge, it’s going to become very valuable, right? As operators put together their plans on how they’re going to manage dents, the ability to say look I’ve got a dent with a gouge and I’ve got this mitigative action that I’m going to take, or I’ve got a dent where I have confidence that it’s not a gouge. I’ve got some other metal loss feature, Maybe I have a dent with metal loss, but I’ve got confidence it’s not actually a gouge, and so I’m gonna take a different action. I think the specification enables that as a defendable position.
Rhett:
And I completely agree. I think one of, in there, I think the thing that really catches on to me is when you, in 712C, I think it’s like paragraph eight, if I am not mistake, and says that the operator has to account for tool sizing tolerances and other things. I think as operators try to employ that ECA process from 712 and you need some real meat for the sizing tollerances that is when that specificication is going to become so important.
Chris:
It is an enabler to do it right? It is an enabler.
Rhett:
Right, its an enabler and I would even say that there is a chance of even setting presidence. Again, I don’t have enough positive things to say. I really hope that is when you guys continue to push industry forward and other people catch on. I do have –
Chris:
I know exactly what some of the integrity managers are thinking. They’re like, where was this during baseline assessments?
Matt:
Yeah, yeah.
Chris:
It takes time, right?
Matt:
It takes time.
Chris:
And they’ve done it. You’ve got the feedback you’ve needed. My perspective is it’s, you know, industry’s giving you guys the feedback and you know the operators that you’ve been able to work with to not just do that small scale testing, you know to establish a spec but to me it sounds like on the magnitude that you’re talking about I mean that’s really more like validation right I mean you can use historical information that these guys have given you to say this is why we feel comfortable so it almost feels like it’s been validated you know it’s an operator should be able to use that under certain circumstances so just you guys just keep bringing value to this game
Matt:
Yeah I mean you know one end of the big the benefits that I see is often in laboratory, you’re focused on the performance of, let’s say, the science of the classifier. The way that we approach this, the way that it really played out, we leveraged the entire system. So we’re measuring our performance, not just on how this algorithm works and doesn’t work in a laboratory environment, but how it’s applied and utilized through the data science organizations. So the full system’s really being validated, tested, and measured, and that’s what we’re able to—
Rhett:
I got a question.
Chris:
Make it real for us.
Rhet:
I’m gonna do what I know a lot of operators are gonna do. — Hold on. I didn’t run your MDS tool. I ran something else. Can you give me the baby specification? Do you guys see trying to push that bound or you just don’t think that’s ever really gonna happen?
Matt:
– No, so the—
Rhett:
– Cause you know, the operator’s always trying to push it. I know you haven’t already gotten that. I know it’s coming.
Matt:
Yeah, no, we get the question quite a bit.
Rhett:
I just asked it for all the viewers on the thing. I’m like, I know somebody is asking this right now.
Matt:
But, no. And the reason being is we’re, you know, the development of the algorithm, we leveraged the machine learning modeling, you know. It’s not, you know, it’s not super extensive, but we’re feeding in due to it from the data that we have, from the MDS data, in that process you identify what’s critical. Turns out a lot of the MDS technologies and the amplitudes and the way that those waves are produced are pretty critical. The SMFL data is leveraged pretty heavily, the axial MFL data, and actually the low -fill technology. It plays a big role in really into that classifier is or is not a gap. So not able to apply it to something that we don’t have that data set.
Rhett:
Thanks for the explanation.
Chris:
I feel like we should address that a little bit and everybody’s technology is different so I’ll be very broad right so for the listeners is it’s if you’re saying well hold on well I’ve got two of those three right so maybe I don’t have the SMFL or maybe I don’t have the low field why does that matter again what you want to think about is not just the integration but also differences right when you’re thinking about classifying you’re trying to say it’s this and not that right and so those other technologies are really what are allowing you guys to say this is why we have confidence that it’s this i.e. gouging versus not so it’s not just a one system not just one technology but rather the system as a whole yeah so if you guys if they figured it out that way use them that way that’s what we would say right you have a solution get the solution
Matt:
So what I will say is we didn’t make a hardware change with this so the MDS tool we ran last year we can’t apply it so
Chris:
They brought back the residuals so no no no
Rhett:
You tell me I can go back to like 2010 and take my MDS data
Matt:
If you believe that data is still relevant to your pipe condition today the tool the tool It didn’t change. This is not a hardware change.
Chris:
Didn’t the first MDS have a residual field versus a low?
Matt:
Yeah.
Chris:
-So you kept residual, but now it’s value.
Matt:
So we still have residual on two of our technologies. And it applies the same way. The way that that feeds in, whether it’s our low field technology or the residual technology, they both apply-
Chris:
– So actually, that’s a good technical point. Does the spec apply to both the residual and the low -field or just the low -field?
Matt:
It applies to both.
Rhett:
He got really excited about that match. He wanted to make that point.
Chris:
That’s so good.
Rhett:
You need to settle down.
Chris:
Will you make this practical for us? So I got a question for you.
Rhett:
Make this practical.
Chris:
Make this practical. So if an operator ran the MDS and they find that under the prescriptive measure they may have to respond to a feature, they did with co -located metal loss and they ran the MDS. How do you see an operator using this as it relates to ECA?
Rhett:
Right, so as you walk down the process, I mean, first step, you’re gonna run the strains and find out whether or not you’re gonna be able to fall within that process, right? So if you pass the strain limits, then you’re gonna kick into the ECA process and you’re gonna have to redefine your reassessment interval. And in doing that, that’s where it’s gonna come really critical, so you’re gonna have to identify, it says to review all of your high resolution data, you have to identify the threats that are associated with it, then you’re going to have to assess them. And it’s really when you get into that assessment part where I think it’s going to matter the most because if you’re going to assess a dent with metal loss, that’s actually going to be a little bit different than a dent with gouging. But in both of them, you’re going to need a sizing specification because then in the back half you’re going to have to assess more than likely conservative estimates of the metal loss sizing. I think where TDW really set themselves apart with this and again I keep and I feel like I’m just like — I’m excited because this is what this is what happens when you don’t I have a dent with metal loss and the operator’s like okay I want to analyze that what’s the size of the metal loss I’m sorry I can’t tell you what it is well wait I have a dent with metal yeah I just can’t tell you what size it is well what do you mean and I’ve run into situations where it’s a below spec metal loss and a below spec dent I run into situations where if the dent box hadn’t been drawn, the metal off signature wouldn’t have been called. And I’m like, wait, so the only reason you called this metal off is because there was another box here? Well, yeah, I wanted to help the operator. You did not help the operator. You might have just cost them a whole lot of money for something that’s not significant.
Chris:
Benign, yeah.
Rhett:
Benign, yes. This, for me, really starts to turn that on its head because now it’s dent with metal loss. And here’s the information you need to complete the picture and confidently complete the picture. That’s big.
Chris:
I feel like we get it.
Rhett:
We do. Yeah, let’s talk about the future, right? So we’ve got a couple minutes left. Where do you guys see this going? I mean, is this it? You’ve done with it or how do you see this going? No, I wish. It’ll be continuous improvement, right? So we’ll continue. I hope operators leverage it, continue to leverage it. We’ll continue to gather data, re -evaluate the specification, and continue to improve it. We’ll look for other opportunities, other ways to apply other critical dimensions or data sets to further refine the spec. Obviously, dent with metal loss isn’t the only challenge out there, so there’s other ways, other processes. One of the things I’ll be talking about at PPIM this week is our Selective Seamworld Corrosion classifier.
Rhett:
So are you going to be meeting with us again next year to do the same thing about Selective Seamworld Corrosion?
Matt:
We’ll have to see, but you know, it’s the same story, right?
Rhett:
It all depends on how many ratings the show gets.
Chris:
No, no, you know what I was thinking, how many people in the next three weeks from when this airs, they get called and say, “Hey, can you pull that 2019 date and do this analysis for me?”
Matt:
Yeah, that’s right.
Rhett:
For our audience, we get absolutely no returns from anything. We have not monetized the podcast. There’s nothing here for us. This is just for fun. TDW paid us no money, and we are paying TDW no money to appear in the podcast.
Matt:
All those disclaimers, no. So it’s continuing to leverage the technologies we have to solve problems, and that’s always fun.
Rhett:
And you’re not done.
Matt:
We’re not done. That’s fun.
Chris:
We need to go play disc golf again over there.
Rhett:
Are you gonna play with us next time we come?
Matt:
If you guys come out there, I’ll see if I can find a way.
Rhett:
No, no, I need to know if you’ll come throw a hole with us.
Matt:
I will throw a hole.
Chris:
We’re gonna get the one over there by the intersection.
Rhett:
The one that’s right next to the street so you can throw into the hole.
Matt:
I’ll hit cars. I know I’ll hit cars. I haven’t played that course I have played disc golf I’m not that good.
Chris:
It’s okay it’s about having fun. If you haven’t figured this out it’s about having fun.
Matt:
I get it.
Rhett:
It’s fantastic. All right Matt I want to say thanks for joining us on the show today we really appreciate it.
Matt:
Appreciate the invitation this has been fun.
Rhett:
It’s been fun I hope we do it again. To our audience out there I hope you guys enjoyed this episode in the ARC where we’re talking to vendors at PPIM23. I’m your host is my co -host, Christopher De Leon, and we will see you again in two weeks.
Thank you!