Can MFL Capture River Bottom, and What Exactly Is A Hot Take? Featuring Bruce Nestleroth

0:00 Rhett All right. Welcome to today's edition of Pipeline Things. Two topics on everybody's mind. Who's your ILI vendor, and how is your fantasy football team doing? And maybe there's some correlation between the two of those, similarities. Maybe we can draw a metaphor between fantasy football and ILI vendors. You're going to have to pay attention to this episode to find out lots of hot takes from Bruce Nestleroth, a returning guest, and a little discussion on fantasy football. Thanks for joining us. 0:45 Rhett: You're such a giraffe, an animal. All right. Welcome to this edition of Pipeline Things. That's actually take two on the intro, because in the first intro, Chris fell onto the table and broke multiple things. Literal, literal proof of broken things. Evidence of carnage on Pipeline Things. So, I am excited to be back this week. But before we get into this, this week, you're going to date the episode, and I know enough about you to know that I can't talk you out of something once you have it on your head, Chris, or on your mind. So- Chris You're successful sometimes? Rhett No. Not really. No, no. Not really. So I know what's on your mind because I'm the one that's probably partly responsible for introducing you to it, like four years ago. Chris I think it's more like five. Rhett And that's fantasy football. Chris Yeah. Rhett Chris, this is the opening week of fantasy football. What's on your mind? Chris Yeah. So it's a great, it's a great question because it feels a lot like the topic for today's episode. So, I'm in a situation, right? Where we-- I have to pick three players to be in our starting lineup. And, two of them play tonight. And so out of the three that I have to pick, I have four legitimate contenders and all within like a point and a half of each other, two of them are established players that we know their talent and they're- Rhett Veterans. Christopher Yeah. Well, I guess you could call them veterans. Yeah. And two are rookies, but they're really good, really good matchups and in really good situations, Rhett High potential? Christopher High potential. And so, I'm stuck. I'm not sure who I want to pick. And I have a timeline like two of them play tonight. So, I have to make my choice of, "Who am I starting?" Lineup is going to be here pretty quick. And I just I felt like often we feel like these pressures, pressures are analogous to what we deal with. And what we're gonna talk about today on the episode. 2:41 Rhett Is it really? You really feel like that's uh, that's analogous to MFL or- Chris ILI? Of course it is. Which vendor do you pick? You know, I have, I have to get a run by this date. Which tools are available? 2:51 Rhett Now that maybe, but I was thinking more like usually, you know, you have, you have the vendor that you want to use, but it's like they're coming out with something new, right? Or maybe it's using the technology in a different way. Chris Is it a vendor or the technology? It's like, hey, we know we have to do a metal loss assessment. The line has had these two vendors in it before, and we're comfortable with either of them. But there's this other vendor that seems to be doing really good things, and I feel like I want to give them a try because I feel like I understand the integrity of my pipeline right now. And we know that if we look at KPIs, you could find a legitimate argue to bring them in. Either it's a first-run success or on time reporting. You're like, well, this is where I get to with the fantasy football. It's like my two veterans should perform, but sometimes things happen that they didn't that day, and it's perfectly explainable sometimes. Rhett Is it? You were supposed to get a first-round success. It doesn't happen. They perfectly explain it. Rhett But like let's flesh this out. 3:50 Rhett So, I'm all about killing this analogy. Chris Oh- Rhett Yeah. So, I could see it like you also see, but you already answered the question, which is when you have a proven veteran, you run the rookie against the proven veteran and the veteran is your baseline. Right? And then you evaluate the rookie as to whether or not he matches up against the veteran. But you can't do that in fantasy football. Run one against the other. Chris You can't. Yeah. Because once the week is gone, it's done. Rhett Let's go ahead and take this analogy step further. My grizzly veteran Amari Cooper just announced he's retiring. Chris He's retiring! Rhett So, will you please extend that analogy to ILI? I mean, does this happen where it's like you contact the vendor and the like, no. Peace out, dude. We're out of the market. Chris Actually, I recently found out that there is a fleet of tools that recently got retired. That wasn't in the last 24 hours, but it was within the last, call it, 18 months. 4:39 Rhett Yeah. Well hopefully that doesn’t happen to clients where they're like, yeah, I know we have that run scheduled for you next week, but we're actually closing shop. We're done. Chris So maybe not the whole show, but anyways, yeah, I just, it's fantasy football is a fun time. There's wins and losses, and I feel again, that's a whole lot like fantasy football from week to week sometimes. 4:59 Rhett Yeah. Well, mine's starting off. We'll see. I guess I'll lose our league. Chris So best of luck to all of you. Best ballers. Redraft, dynasty leagues, all that fun stuff. Rhett So, on that note, we're bringing back a guest who, was probably one of my favorite episodes. And I think the reason that I like this particular guest is I mean, obviously, our guest is really well established in the industry. Really recognized. You know, I think I actually sent an email to a client the other day who said, hey, somebody said this, and I literally copied and pasted the transcript from our previous episode into the email and said, yeah, this is what Bruce says. And I take what Bruce says against anyone else, hands down. But you can choose however you'd like. And so, without further ado, that is our guest, Bruce. Welcome back to the show. Really glad to have you on. 5:50 Bruce All right. Thanks for having me again. I'm looking forward to this. Would have rather been, you know, live in the same room as you guys. I think it's a lot more energy that way, but not sure I want to be in Houston at this time of year. It's, seventy degrees and sunny up here. I don't know what it's like down there. Rhett We are in the middle of a cold front, and it's 96. Chris Yeah. Rhett So, I mean, you know, Christopher And what he means by that is this morning when I was taking my kids to school, it was actually really nice out. There's a cool little breeze for us Southerners. It was, that nice humidity, not the sticky humidity. It was awesome. And then you walk outside at 3:00 and it's like, what happened? Rhett Yeah. Right. Well, so, Bruce, clearly you did not get enough punishment from us last time or you enjoyed it so much that you, you couldn't resist. But the audience that doesn't, or if you’re new to this episode or new to this guest, I would highly encourage you to go back and listen to the first episode to get a little bit of background on Bruce, hear where he came from. But since the last episode and today, Bruce, can you tell us a little bit about what have you been up to? What's, what's Bruce doing these days? Bruce Yeah, kinda what I tell people is I'm easing in retirement, I probably only work a day or day and a half a week anymore and kind of, kind of don't want to work that much anymore. I'm getting in shape, trying to, trying to, you know, play with the grandkids and going on vacations, just did a three week vacation in Europe. There's just a lot of a lot of fun stuff. Rhett So, is this work or vacation? Chris Or fun? Bruce Oh, this is this is this is fun. I don't count. Chris YEAHHHHH. Rhett There we go. Bruce I’m not... I'm not billing. 7:29 Rhett We need to double check that, miss producer. Chris Make sure just in case. Rhett So. Excellent. Well, then. Hey, I'm excited to talk to you. I know Chris is so, audience, get ready. This is probably going to be more of Chris talking than it is me talking just because Chris gets really excited about this. So, Chris, you know what? For once, I'm going to turn it over to you and let you guide our return guest through the journey that is pipeline things. Chris Well, we will see how long that lasts before you take the reins. So, Bruce, you know, obviously with your background, if anybody who didn't listen to, to your first time on the podcast, your background, has a heavy magnetic based ILI technology component to your experience. And so, recently, if, you know, if you scroll through LinkedIn for longer than 10 seconds, you're, you're likely going to see some kind of post related to either an MFL tool or any other magnetic base technology. And so, we thought it'd be fun to bring you back on and just ask you a series of questions where we'd like to get your hot takes. Really, That's the idea. Rhett Wait, does Bruce know what a hot take is? Bruce, that's a new term. Like I just learned what hot take means. So, I feel like Chris is throwing this term. Do you know what hot take means? 8:41 Bruce Maybe not in your context. Chris Go for it. See you couldn’t be quiet, go for it. Rhett So. Yeah. No. So, hot take is with the new kids on the block are using like, my son Bruce Ah, nope Rhett And my high school students, right. So hot take means I'm going to take a take that's usually opposite or likely to draw a lot of fire or criticism, for what’s said. Right. So, I don't want to give a hot take on this episode because I'd inevitably like, insult somebody. But, yeah, I'm trying to think Chris Yeah, don't do that. It will come; it will come. Don't worry. Bruce I'm not afraid to get in trouble. Rhett Well, there we go. Right. Bruce Yeah. And I think my, my very first paper I gave back in, I think it was ‘91 on, on MFL technology. They asked me what is your definition of a high-resolution tool was? and I thought for about three seconds, and it was after paper and I go, well, that depends on the smallest defect you need to size. Chris Yeah. Bruce And it is a perfectly accurate answer. Chris Yes, it is. 34 years later. Bruce Joe Caldwell, I don't know if you remember him. Joe Caldwell is an old timer. He goes, Chris No, sir. Bruce That is an absolutely unacceptable answer. Rhett But okay. Yeah. Okay. Bruce So, so that's a hot take? Rhett Yeah. I mean, it could be if what they needed was an absolute size, for sure. Chris So the idea is, Bruce, what I'd like to achieve with this episode is, is to get your, your, your opinions on things and, and share some of your experiences such that our, our audience as they're flipping through LinkedIn or they're looking at a proposal or they're sitting in, a technology presentation, they have a basis to, to determine if they feel like they agree with you or if they disagree. And I think that's important. Right? Because that's when people start investigating what they believe, and they start learning. Right? And so that's why I feel like hot takes are important. Bruce Yeah. Chris Not so much to convince people to agree with you, but to rather to draw a line in the sand and see where people fall, if that makes sense. And so, to get into it, if you wouldn't mind, we talked about this a little bit last time. What? How would you describe the ILI systems that leverage high field and low field technologies? Not necessarily residuals. Right? But this idea of I have a calibrated high field technology, you have a calibrated low field technology. How would you describe ways that those technologies could be used and, and any personal experiences that you have with them? 11:20 Bruce Sure. I mean, when I started in the business, what, what is considered low field now was actually high field that was as good as you could, the magnets were not that strong and so but what, what the industry has done and for good reason is that they went to higher and higher magnetic fields because they got the magnets which kept getting stronger. Because what that does is make all pipe look the same. In other words, you know, it makes the magnetic properties of the steel all the same. And then, but if you go to low fields now, all these magnetic, the magnetic properties are different. And if you have some kind of metallurgical variation in your steel, you're going to be able to see that if that can come from mechanical damage, that can come from, you know, hard spots or, you know, any, you know, all that kind of thing can change the magnetic properties of the steel. And so, when you have these so-called low field or even a residual field, you know, the metallurgical changes give you a different signal. And so, you can actually see things that, you know, aren't metal loss. You can see mechanical damage. You can see hard spots and those types of things. So, you know, it's, it's getting back to, you know what you could see. Old MFL tools saw everything. And that was part of the problem. And so, I used to get a lot of, a lot of false calls when they, when they're 1970s and 1980s, they get a lot of false calls because they were seeing things that were just metallurgical changes. Chris Yeah. Because they weren't saturating the wall thickness. Bruce They weren't saturating it to the level they do it now so and so. But on the other hand, you know, some of that, some of that, you know, old signals were important. You might it, might be mechanical damage. It might be a hard spot. So, you know, those are things that you got to find. Chris Yeah. So, it's fair to say that part of the evolution of, I guess MFL technology, to be clear, right? Not just a magnetic base tool, but MFL technology is that the biggest change is possibly towards POI, right? Bruce Yeah. Chris Because they weren't properly saturating lines there. You'd get indication, but you weren't sure what it was. Now, by using these permanent magnets that we have saturating the material, we're saying, hey, it gives us a lot better, view of when we see leakage because there's so much saturation. It's most likely due to some kind of volumetric metal loss. And so that's what's driving the indication. Bruce Yeah. Do you ever get a call-- now when you look at a log, do you ever get a call-- it's called metal in close proximity or metal in the backfill? Chris Yeah. Rhett Yes. Bruce That doesn't show up anymore. You know, in the in the old days that was a common call. Chris Yeah. Bruce It's just one of those things, you know, they've gotten rid of those things. But, you know, you know, sometimes, you know, that could be important. Chris Yeah. So, I'm curious if we could take advantage of your, of your experience a little bit. So, you know, as part of the failure files, we went over an incident, from 2019, the Danville, Kentucky failure, the hard spot failure. Bruce Ok. 14:32 Chris And just a little bit of a recap, right. So, at the time Tuboscope had a hard spot tool and, they, they ran they deployed that tool, call it circa 2004. Bruce Ok. Chris And it only it only called a certain number of features. And then they reran it in in 2011. And after that run, again, only a nominal amount of features called. Fast forward failure happens in 2019. The data is reevaluated. And now instead of finding something on the order of magnitude of a couple of tens of features now, in the hundreds of features. What's your take on that? Right? I mean, the data was there. That doesn't seem like it was a magnetic improvement, right? Because they're still looking at historical, magnetic data now it's more, it seems more algorithmic or experience. Bruce Well, what was what was the tools that changed by then, didn't it? Chris That's the question. Right. So again, I wasn't around back then. So from 2004 to 2011. Bruce So, from 2004 to 2011, it did change. Well, I was working on a project for, and that was when we were developing the dual field MFL technology. And I'll take, you know, and so we started doing the technology transfer in 2005 and Tuboscope and, Rosen and TDW were, participants in that project. 16:02 Bruce And we started providing information in there. So, at that point, it's very well good chance that the 2004 tool, now I'm going to take a step way, way back is the, the best hard spot detection tool. And time you mentioned Tuboscope. And that company isn’t around anymore like, and you know and so, but they were. they were the first pigging company. They were from in the United States, I know. 16:10 And everybody's going to say they had a couple of them, but they were really pretty big, big news in the 1960s and 70s. And they had, they had a very odd implementation. Also, electromagnet versus a permanent magnet tool. And the odd thing about that tool was it was great at hard spots. It would run through the they run with the magnet the first time through with the magnets on and then run it through a second times with the magnets off and the way it left the residual field, it left it at a really high level compared to a permanent magnet tool. I've got a paper on this on why that's true. We'll get into that. But just trust me, it does give you a very strong field. And what it did is it was really good at hard spots. But running a tool, the electromagnet, just wasn’t practical. So, when the permanent magnets got strong, that tool, you know, got shelved. That's where it all went. So. But the new tools, really don't produce as good an information as that old Tuboscope tool. And then you have to have now, going forward, what everybody did is they still used a basic concept that Tuboscope did that it was basically look at the high field. You know, it magnetized a pipe and then look with the trailing set of sensors for a residual field. And that's what they did in 2004. And yeah, they found it. It will find some features. And my usual statement is if you find a few, a few features with a residual field tool, I think you should run a dual field tool because you'll find more. And that's basically what you came up with, I think. 18:12 Chris Yeah. What's the from, from a tool perspective. Can you help us understand the difference between I mean, what is the actual difference in the residual field and a, a calibrated low field. Right? And maybe the advantages there. Bruce So, a residual field just means you magnetize the pipe and then oh maybe you know, you because you have a multi body pig. And so maybe 3 to 6ft behind the behind the magnetizer, you know, you run another set of sensors and everybody knows especially your welders know that there is magnetic field left in the pipe after you do an MFL run, you know, they once they try to run when they try to weld that can be a mess of a time. And so yeah there's always magnetic field left and with hard spots or cold working if you, you know, have some kind of mechanical damage or cold works it what it does is it just changes the amount of flux that that the pipe holds. It's called retentivity. It's a big word. And it just tells you how much residual magnetism is left in there. And so hard, hardness. If it's hard, it's magnetically hard. It holds more material. A magnet is a very hard, magnetically hard material. It holds a lot of magnetism. Harder, harder steels hold more magnetism than soft steels. You want soft steels. You don't want, you know, you don't want something brittle and hard. So that's what you get. So, when you go through the residual, it measures it, you magnetize it. And then at the end of this thing it's left with a little bit. But it's usually something that's on the order of 10 to 20% of the original field. Not much. And of the magnetizing field and you're left with the signal is small. If you have an active field going behind, but it's in the low area, you're still in the area where you're sensitive to those magnetic changes. Why you had all those false indications, why you had those metals at close proximities you know why? You know, you had all these problems you had in the 1970s and 80s. That's, that's where you're operating. But you go, oh, there is something there. And it is repeatable. So that's what you're doing. You're just coming back with something where you got a good control of it. Whereas the other one? It's kind of an open loop. Yeah. You'll see a few things, but it's going to be such a small signal that it's, you know, it's going to be hard to detect all of. 20:41 Chris Yeah. So, then it sounds like the, the take is if you were using a residual field, then you're probably finding the more, the more significant or gross features because they're retaining more magnetism. So, you might miss other relevant ones. But that's where calibrated low field came in. So, it gives you visibility to some of the, the, the complimentary features that maybe the residual field couldn't see. Bruce Yeah. Chris So it's, it's a push towards a POD of 90% since that's our, our benchmark in the industry. Bruce So, what you're trying for, you know, 90% of what. I gotta ask you something. And it's, it's always a hard question. What is a hard spot? I mean, once you get down there, the outer I get to the, you know, I mean, I’m the ILI guy and you're always going to be picking on ILI. And I think they do a pretty good job is that, you know, because once you get down to mechanics guys, they can't decide, you know, what's a hard spot that I need to find? I mean, the API definition of a hard spot, if you had that in your pipeline, it would already have found you. Chris Yeah. Well, fortunately for us, PHMSA has drawn that line in the sand with their advisory bulletin. So, so that's helpful, right? So, if anything, it gives us, a line in the sand for reporting thresholds. Right? So, you know, if you've ever talked to an ILI vendor a lot of times when you ask them what their capabilities are, they'll reply with, don't worry about my capabilities. Tell me what the ask is. Bruce Yeah. Chris And I'll tell you if I can do it, right? So at least now we have a line in the sand that we can say what is our, our probability detection for features that have these attributes. And so fortunately PHMSA has kind of helped nudge us in- Bruce Oh good. Chris: What that might be. Bruce I guess I have to I have to go read up on that one. I'm, I'm a little behind the times there. Chris No, no, it's all good. Rhett anything on, on high field. Low field, hard spots? Rhett No, I do have questions on some things that Bruce said, but I think the second part of this episode, we're going to take a break real quick audience, because when we come back, we're going to shift directions and talk a little bit more about, complex corrosion and maybe different burst pressure prediction models and get Bruce's take on those. And I got some follow up questions on that. But hang around and we'll be right back in just a moment. 23:01 Chris Da na nuh. Da na nuh. Rhett That was actually terrible. Nobody knows what you were trying to do, but Chris Everybody knows that's the NFL theme song! Like theme! Rhett Chris, it's probably, the first part of this episode is probably the least I've ever talked on any episode. How did it feel to lead the show? Audience you judge! Chris It was a lot less stressful, that's for sure. Rhett Was it really? I don't know how to take that. I find that mildly offensive, but the audience can judge. We get enough feedback. Maybe we'll just. We'll turn you over until the lead host. Chris No, we're not doing that. Rhett Hey, Bruce, I had one question that I wanted to follow up because you made a statement, earlier, which is, is sensors just keep getting smaller and smaller, and you were in your hot take thing, you guys, you asked that you were asked the question in a publication, which is how small is high resolution? It got me thinking. It feels like the move in MFL tools, certainly over the last decade has been smaller and smaller sensors and it seems like, Chris Ultra! Rhett What’s that? Chris Ultra technology. Rhett Yeah, I mean the ultra I don't know what we'll call the next one. I mean, as I said, I feel like eventually it'll be nanotechnology is what it'll be. Chris Well, there's a there's a micron right now. Rhett Well, I know and only thing smaller than that's nano. And I don't know if we have anything smaller than nano yet. I mean maybe quantum, but it doesn't feel right. So, Bruce, I just, it got me thinking, since, you know, it is, is there a practical limit, like is there a ceiling on or floor I guess, Chris Yeah. Rhett Below which you'd say like, look, once you get below a sensor size of X, practically it's impossible for some reason or other. Or there are theoretical limits. Chris And specifically, haulofect sensors right. So not just any magnetic base technology. We're talking about haulofect sensors. Bruce Well, you know, haulofect sensors are the only ones that can keep getting smaller and smaller. You, the only other technology used on MFL tools are coils. And you can't, you know, winding a coil on something that's, you know, getting it down to, you know, so that does because I had a wire gauge and all that kind of stuff. That's just impossible. But, but you know, haulofect sensors, that’s a semiconductor device. And, you know, semiconductors, you can keep making smaller and smaller and you get finer and finer resolution. And I do think that the ultra-tools do give a step, advantage over the traditional, you know, tools that, that they were before that which were, you know, on the order of probably three millimeters or an eighth of an inch or quarter of an inch type of thing. Those were, those are pretty good resolution. But I think the, the, you know, the finer resolution that you get with these new tools is a function, but you're starting to get down to, to the, to things where, you know, corrosion is still stings the size of your fingers. You know, even a, even a and a pinhole, a pinhole is something that's 4/10 of an inch in diameter. I mean, your pinky could fit in that. And so, if you have, you know, let's say, you know, ten measurements across there, 40 thousandths of an inch, that's a millimeter. That's perfectly fine. You know, I, you know, even, you know, five measurements across that you're that's about where you're getting to. So, you're getting close to you need on the order of between 5 and 10 measurements for the dimension of a defect that you need the size. Otherwise, it- Chris Yeah. Which sounds great, right? I mean to say, seeing an ultra-tool with two millimeter sensor spacing is pretty reasonable these days. Bruce Yeah, Chris 80 thousandths Bruce And that and, you know, two millimeters is, you know, across a defect that's, you know, a pinhole which is ten millimeters. That's, that's okay. But if you're trying to say something that's two millimeters cross, you're not going to size it. I still stick with that. It's the it's on the smallest effect you need to find Chris And today our standard is POF, right? Which gets down to 1A or sub 1A1A, which would be ten millimeters by ten millimeters. For pin. Rhett So- Bruce Right? Now. Rhett Oh no I was going to say that that brings us to our next conversation. And I didn't want to transition without you, but. Bruce Oh, well the other the other side of things is and I guess before you get there, I did want to say one more thing. If we're going to change something before we get to the end. The end goal, though, is that over general corrosion, you're still not going to get much. You're not going to get better than a depth and a light. You're not going to get the river bottom. And as you go through the— Chris Oh- Rhett And you see, you went, you went to the next, that was where I was going next. Bruce, is I feel like, oh no, it's fine. I mean, let's, let's go ahead and just open up the can of worms. You're probably going to give hot takes here. I don't think you have to go further than LinkedIn to look at, honestly. Almost. There are multiple vendors out there and I think the claim persist that it's , it's sensor size or other things that lead to the ability to handle complex corrosion and produce river bottom profiles. And I feel like, again, one of the central maybe it's maybe it's just clever marketing, but one of the central claims is that the sensor sizes now are small enough. So yes, 20 years ago you couldn't get a river bottom, but now you can get a river bottom profile because the haulofect sensors are small enough. The resolution is great enough, Chris Sampling rate. Rhett Right? It's like we went from a, traditional TV with four by three ratio to now we're an HDR, and now we're an eight K, right. And with eight K, I can actually see the stitching on the football as the wide receiver drops it on Chris’ fantasy league. So, what say you Bruce? 28:41 Bruce You know you're still, you're still stuck with something that is pretty foggy with an MFL signal. it's not going to give you, you know, it's not a direct measurement. It's an indirect measurement. It's more of a shadow technique. And so, what you're going to get is you're going to get piecewise river bottoms. No matter how you do that, you're going to get, you know, if you have a pit and in one area and another pit in another area, you're going to get those peaks and you're going to get, but you're not going to know how deep the, what the what the area is in the in between those pits. You're not going to tell you if that did that go to full wall thickness or did that go to 20% or 40%? You're not going to know the valleys, how deep the valleys are. It's not going to be a true river bottom. It might tell you where some peaks are. And it can tell you, you know, where the overall length is and where those peaks are located. But it's not it's not. That's not river bottom. Chris So, let me ask this a different way. Rhett Wait, wait, why is that not river bottom? That's because you're saying that you can't. You can get the individual peaks, but you can't get what's in between the peaks? Bruce Right! Well, that's what RSTRENG did. Then, when RSTRENG, you know, took every whatever decided you want to make your measurements, it was the highest amplitude. And let's say every quarter of an inch down the pipe or every half an inch down the pipe, and you found the biggest, you know, the deepest point on every one of those rows in an Excel spreadsheet or, excuse me, a Lotus 123 spreadsheet. Rhett Lotus! Bruce Notice that was what it was originally written in, and so when you were going through there and, and then it just, you know, some, you know, and then it sums up those steps or now, you know, okay. You're just saying, well, I'm going to give you the peak here and then the peak here and the peak there and the peak as you go. But that has no information about in between. So that's not river bottom. 30:37 Chris Yeah. So another way to think about it is comparing these ultra-high resolution MFL tools with, let's just call it for now, two millimeter sensor sizing and really high sampling rates, that if I were to take a laser scan of a feature, a complex corrosion feature, not this nice pretty manufactured flaw or series of manufactured flaws, but let's call something or we'd actually want to use our stream, right? Let's call it it's, you know, four feet long by 50% of circumference of the pipe. You know, I mean, it's this is substantial. So you're saying if I laser scan that and I have, let's call it two millimeter by two millimeter grids from the laser scan, that it's not practical to say that I would be able to extract the same location and number of true deep pits from the MFL data that I would find in the laser scan data because of the, the physics behind the measurement principle of how the sensors are right along the ID, even at the resolution and sampling rate? Bruce It's not going to be the same picture. As you know, it's not going to, it's not going to be the same picture to get exactly the same measurements that you have out of a true RSTRENG, micrometer depth measurement. It's just not going to be. Rhett Well, let me ask you, I want to ask you two questions. Opportunities for hot takes, Bruce. If you so, choose. Why do you believe there is so much confusion? Right. Maybe I'm assuming. No no no no. Maybe it's. Maybe I am presuming. Chris Yeah. Rhett And maybe Bruce would reject the hypothesis. My, my confusion. So let me start over, Bruce. I believe our industry is confused about this. Chris Confused or just don't understand. Rhett No, I think confused. I think confused. And if Bruce rejects my hypothesis and he can choose not to answer the question, I think most vendors, by client request have a column in their spreadsheet that has p burst effect of area. That's what it's titled. Bruce Yes, it does. Everyone. Everyone, everyone. Everyone has that. Rhett If what you're saying is true, why, number 1, why are operators asking for it? Why are vendors providing it? Bruce Well, vendors provide it because operators ask for it. 32:55 Chris Hey, how many times you heard this, Rhett? How many times? We're in a meeting and we say we're are capabilities. And they say, tell me what you want and we'll figure it out. Rhett I feel like- Bruce It's a don't I mean, they're trying to, they're trying their best. And I think the vendors try to do that, and they're trying to give you a value out there and their competitor say no. And they and their competitors say the same thing. But, but they all know that it's not true river bottom. And, you know, I mean, and it's never been shown or actually proven that it has it's always there's a few papers that demonstrate it, but it's never been, you know, hard core shown that it that it gives you the same value. It's probably conservative. Rhett And I agree with you. I'm not casting shade on the vendors. And to be quite clear to the vendors that are listening to this episode, I do believe that the vendors default response is yes, and I can understand that. I so maybe I'll just ask it in a different way. Why or why is we why as an industry, is that understanding not more clear among operators as to the limitations of using MFL with RSTRENG? Have we not done a good job of education? Do you think it got muddied somewhere? Is it? Do you think we dropped the ball? I'm just curious because. Chris Is it a dirty secret? We just don't like talking about it because it gives us, a larger burst pressure, and we like that. And we know there's other conservatives and built in, so we just don't talk about it? Rhett Or is it because, like, UT in and UT does have the ability to do it if it's taken point by point measurements. It was just curious. I wasn't looking to cast shade, just curious. Bruce I've been I don't understand the problem either, right? I still teach this, and I've had people challenge me and I'm sticking to my guns. And if somebody can show me that they are actually getting to R\fiver bottom and show me how that's being done, I'm all for it. But you know, if you know the other way to look at these things is you look at the papers where they're trying to do magnetic inversion, where they actually take a 3D field, take the all three magnetic sensors and try to invert that into a, into a geometry. And they're not all that successful doing that. Chris Yeah. Bruce Every paper I've ever seen on that never does that. And if you can't get the picture, you can't get the river bottom. Chris So, I, I did a little bit of research on this in my earlier days of being integrity engineer, like when I first started out. And I have assumptions and I'm calling them assumptions. So, audience, if you have an I, please share. And I truly mean this. If there is literature that is published, that that contradicts what I'm about to say or fills the gap, please share with us. So if I'm not mistaken, I may get the year off or so I remember when I was doing homework back in like 2008-2009, I found an IPC paper by PII call it 2004 was around when it was published, and they introduced the concept of LAPA. And so, and my take on it then was that basically it was they would take the box feature dimensions, and they would cascade them down into river bottom profile. And that was strictly based on interaction criteria and what the ILI tool is able to see. And they collaborated with a couple of operators and showed that that approach could be conservative like it could be used. And so, they would use the sequence of analysis that effective area deploys against the profiles generated through the boxes that the ILI features called, and that process was not called effective area or RSTRENG according to B31G. It was a process that I understand was branded as LAPA. And so, I'm not saying that that was validated, and I am not saying that that is it is a is a technically feasible approach. All I'm saying is that that was a paper that I found that was intentional about saying, this is how we are deploying the, the, this is our application of an effective area type approach using MFL data. 36:57 Bruce And that's the that's the paper I remember, and I when I present my, explanation of can MFL do RSTRENG? I say that paper demonstrated that it's that it you know, this technique is where you basically are just using stepwise over each individual pits and going that way. I kind of used to, it looks like a Minecraft kind of boxy thing. It's, you know, that's kind of how it looks. It's not smooth. It's step, step, step, step, step, step all the way through the whole thing. But it's not been validated. It's been demonstrated. Demonstration is not validation. Chris Yeah, I agree with you on that take. Now what I would say is what we have seen though is it's so once these boxes are in place, I have seen how these boxes are then sliced into different, axial dimensions. Right. So, let's say they'll take that and slice it into one millimeter by one-millimeter slices and run the effective calculations on that so that more is it models effective area. But again, the river bottom profile itself is based on, on boxes. 38:08 Bruce But it's still but it's still Chris Yeah. Bruce Plateaus. Chris That's exactly right. Bruce Going through the entire complex and it's, it's that's not my understanding of river bottom Chris Yeah. Well, I guess what I would ask the audience is again, guys, we're not trying to be, you know, confrontational or contrarian. I think the ask is the ask specifically is not just work that's been done or, or examples of how it's, it's implemented or its application. I think what we're curious about is , is there any literature that validates the approach of using MFL technology for computing effective area versus pressure calculations, and that it's conservative. And I'm aware of the Advantaca report, and the Advantaca report clearly states that B31 and modified B31 were looked at for MFL data, but it clearly does not say RSTRENG based calculations. And I think that was left out on purpose. Rhett I hate to be the the bearer of circular logic, but you guys, those were with older sensors, and the new sensors are smaller Chris and they're better! And have better algorithms. Rhett Well, Chris, I do want to challenge you a bit on that. And, Bruce, maybe you have an opinion. Chris Yeah, Rhett Because you were asking about literature. There's been a lot of published literature on the p squared approach, which uses RSTRENG as its basis and was developed, as I understand, with ILI as a basis. So, I think the first thing that, hopefully a listener would come back with is, hey, look at all the publications on the p squared approach, right? Which does, as I understand, show that it has some conservatism. And as I understand, Chris Yeah, Rhett Was developed with ILI in mind. Bruce, you probably are closer to this than either of us. Do you have some opinions or, thoughts that you can share on that? Bruce Well, I just think the p squared is just a more practical implementation of RSTRENG. And it, you know, when you know, when you only have a, you know, IBM PC with a five and a quarter floppy disk to be able to run your, your calculations, you're not going to be able to find the optimal path through all these boxes. You're just going to say, well, I'm just going to take the deepest one. And my row and say, that's the best I can do. It's not going to be able to do those kinds of calculations. What you can do today is you can do better calculations, but still, when you look at the MFL data that they have for when they when they did the validation and again, it was still plateaus through the deepest pits in that area. It wasn't it wasn't looking at the, you know, the laser data or, you know, high resolution images. Chris Yeah. So, I was kind of excited when I saw the p squared work happening back in the day. Obviously, I don't think either one of us were involved. And obviously, Bruce, I thought I saw you on a couple of papers, but my Bruce I was fortunate that I was the there's a lot of mechanical guys on that committee that was doing review. I was the ILI guy. So, you know, I mean, and mechanical, you know, they were looking at it, but I was trying to evaluate it there, and I just, I thought it looked like I mean, it was good. It was better than that RSTRENG because it, you know, it did find a plausible profile. I always thought it was kind of ,kind of bad that, you know, if you have a, you know, corrosion pit, you know, at the 11:00 position, another one at 1 back at 11 and back at 1, it goes, you know, that's the deepest point. Go like that. You know that that fracture surface isn't going to run back and forth. It's going to find a path through there. It's got to be plausible. And, and makes a whole lot of sense to me. But getting in, but you still were doing the plateaus for, for doing the MFL data. Now they had what they had a, you know, a couple dozen burst systems. I remember, or something that they did, but it wasn't, which is great, but it wasn't, you know, and, you know, and it wasn't a lot of a lot of proof, and it was just a few different diameters. And it I mean, it's expensive to, to, to validate something. That's the, you know, the difficult. Chris Yeah. One of the papers that I really appreciated that gave me hope of P square and I mean that I'm very, I'm very high on, on p square-based approach because and again it's, it's one piece of evidence, not enough to say it's validated for sure, but it's going in the right direction in my opinion. So, I think it was 2020. TC energy published, a paper at IPC where they basically took, I think it was 154 immediate calls that were a result of RSTRENG in an MFL based ILI report, and of those 154 immediates, they were able to get, I think it was 84 laser scan data for 84 features that were called as immediates by RSTRENG. And of those 84, when they then ran the LIL data through P square, only 15 of them required immediates. And so, what I'm getting to here is saying that type of analysis, that type of data where they're very transparent about the process. Right? I had an ILI report. It had this many features that were called as immediates through RSTRENG. We went in, we did field excavations, we collected laser data, and then we ran p squared. And we do this transparent comparison of them. And they walk through how they're doing p square and how explicitly how ILI data was used. Specifically, the MFL data. I think that starts raising the bar of confidence. Right. And all in all, my position has always just been I'm not saying that IMFL data can't be used for RSTRENG. I'm just saying show me how it's done and show me why we should trust it. And I think that some of the modern literature that we've seen around p squared starts raising that bar of confidence, and it starts presenting information in a way that I think we can start buying into it. And back to Rhett's point. I think that you're right. You know, they do speak to RSTRENGs performance in those papers. Right? And so, by proxy, you also begin to gain confidence in the approach of R string. What I would say is though, that's that is unique to potentially either that operators process or that's vendors’ process. Right. So just be real careful about taking someone else's work and adopting it in your processes without understanding what those essential variables were in in said process. 44:40 Bruce I agree 100%, but it is probably some of the best validation of, of MFL being able to use the piecewise, plateau method of analyzing MFL data into a RSTRENG type calculation, that it does actually work and it does. Well, it does. That's probably the closest we've gotten to validation for our strain. And it takes it the next step and says, and we can do a little better if we do it with LAPA. And I agree, it's good stuff. It's just, you know, again, what you said, make sure your central variables are the same. Make sure your corrosion looks like their corrosion a bit. Yeah. You know, there's a very, you know, corrosion on one person's pipeline and, and person on another one. There are often different morphologies with those. And that can make a difference on whether this technique will work until you've checked it on a, on a good variety. Chris For me, it's also the I like system, right? Like MFL is not MFL, right? I mean, one vendor's tool is not the other vendor’s tool. I mean, and I think ILI vendors do a really good job of staying competitive and operators need to keep running them so that they keep innovating and keep challenging each other for market share, because that's how we get better tools and better solutions. But I do think people need to be careful about how one MFL service provider will translate haulofect sensor voltages into some kind of characteristic data of a metal loss feature. You know, saying that you meet a template doesn't necessarily mean that the data analysis and the export of that data into your template, are the same. They're not always apples to apples. 46:20 Bruce Yeah. And you have to I don't think that any of the Trans Canada work that you were talking about there was done with any of the ultra-tools. It'd be really interesting to see where that goes. Chris Yeah, that's a good that's a good point. You know. Yeah. I'm not aware of that either. Bruce I wasn't around then. You know, it's just that their, their data collections in the late, late, 2010s. And that ultra tools that work. Chris So, if we think cup half full, that means it's only it only could be better. Now. Bruce Yeah. Rhett So, Bruce, Bruce Maybe, Rhett I want to say thanks for joining us again. It's been a lot of fun. We have more topics we could cover, which also makes me hope that, you know, again, in your retirement and your one and a half days a week of working that, we manage to find time to, to chat with you again in the future, but definitely enjoy having you on, and, Bruce And it's always great. Rhett Yeah. And, for our audience, you know, pay attention to this season, because you will get updates on how Chris and I fared in fantasy football. And hopefully it's better than the broken pieces of miss producers equipment that I see on the floor over there. Chris I can say this. Rhett and I are in a basketball fantasy football league together, and, I would be curious who y'all think is going to win out of the both of us. Not against not just against each other, but who's who's going to end up higher ranked at the end of the year? I think everybody knows the answer. Rhett Stay tuned to find out. Thank you, guys, for joining us on this edition of Pipeline Things. We'll be back in a couple of weeks to talk with you guys again. Cheers. 47:58 Rhett All right, credit for this episode. Thank you for Sarah Etier for being executive producer. My glorious co-host Christopher de Leon. Our returning guest Bruce Nestle Roth and to the Work Lodge for letting us use this space for the production. This episode was written by no one. It was made up on the spot. Thank you.