Episode Transcript
Rhett:
If I were to ask you, what do iPhones and the gas industry have in common, your answer would be?
Chris:
They both have cracks.
Rhett:
That's why I use an Android. Hey, on this episode, we tackle a new gas crack detection technology with NDT. Think you'll enjoy it.
Rhett:
All right, welcome to today's edition of the podcast, Pipeline Things. I am your host,
Rhett Dotson, my co-host, Christopher DeLeon. So as we get started today, Chris, and I was thinking about the segue for this particular topic. I know you're like a big iPhone person, and I'm not. Yeah, yeah, yeah, yeah. Curious, do you remember when the iPhone was first introduced?
Chriss:
Oh, I'm totally guessing here. I would say –
Rhett:
No, I would just – when do you remember? Let me ask you. When do you remember first hearing about the iPhone?
Chris:
Yeah, it was probably – I'm going to call it 2004, 2005-ish, and I'm totally guessing here.
Rhett: Man, I remember its predecessor, which was the iPod, right? And everybody came out with this iPhone. And I remember my first thought when the iPhone came out was, why do you need that? Nobody had any idea how transformational the iPhone would be.
Chris:
I see where this is going.
Rhett: I think it's interesting when we think about technology developments and how transformation – I mean, because think about it now. Nobody – I mean, even my parents don't have flip phones anymore, right? Which is kind of crazy. And I can – when my first son was born, Caleb was born in 2009. I know I did not have a smartphone because I remember having to text everybody. I think I had a juke, one of the ones that slides up. Oh, yeah. And you had to like press the button three times. Like the was T-T-T-H-H-E and then like space, right? And texting was exceedingly difficult. You didn't have RCS texting for sure or any of that. And I mean now everything's smartphone. It's everywhere, right? Like you almost can't even remember a time when you had that. And I think like every year it's like, the new launch of the iPhone.
Chris; I mean, if we're going to get nostalgic, I'll even say at least my recollection of it was... the why for the iphone was interesting because back then what was really popular were the ipods and it was like oh why do you need that much music available to you and then they married phone with music and you were like why are we doing this right and so that was at least the way i rendered it and it was either that and if you wanted something smarter then you got a blackberry and that was just having a qwerty board where you could tech stuff we're
Rhett: Talking about blackberries went by the wayside man that died
Chris: The whole point was it was for me it was that whole idea of how two things that you traditionally wouldn't bring together Or the way I consume the iPhone. I was like, yeah, I want that. I have an iPod. Why am I going to carry that device around and a phone when I can have one?
Rhett: And when I first saw it, I was like, y'all are idiots. Make a phone a phone. And if you want to listen to your music, listen to your music. And you never moved away from that.
Chris: You still don't have an iPhone. Get on the bus, buddy.
Rhett: No. So I'm actually waiting for that. I'm actually kind of interested in the Galaxy S26 drop. And so, yeah, I'm kind of moved into the Samsung.
Chris: And I'm assuming you don't listen to music on it, right? Since you didn't think it made any sense.
Rhett: Oh, I do. I mean, but music, look how much music has changed, man. I mean, like... Is iTunes still around? Does that exist in any form? Yeah. Oh, people still use it?
Chris: Now they call it Apple Music.
Rhett: Oh, okay. I mean, Spotify is pretty much where I do everything now where I've kind of been pushed.
Chris: But you now are one of the adopters. You listen to music on your phone.
Rhett: Oh, I mean, 100%.
Chris: So you've changed your perspective.
Rhett: Podcasts and news.
Chris: So you were biased, and then in time, you finally said, you know what, this makes sense.
Rhett: I would not have known. no way had the foresight to see how transformational the phone would become, which is cool, right? And now it's kind of cool because it's just interesting to see how a new technology comes on the market and changes things.
Chris: And you're going to have people that just won't buy into it and some people who will. Well, you know, that's... And thanks to the people that did buy into it. Now you get to have a Samsung phone with Spotify on it. You're welcome.
Rhett: That absolutely has applications to the day. And today, I mean, audience, we're going to be talking about the development of a new technology. That was a non-covert segue. And so I'm going to go ahead. This is based on a PPIM publication and bring on our two guests while I introduce the background. Our two guests are Michael Haas and Rogelio Guajaro from NDT. I think I butchered your last name. Again, Rogelio, as I was ruling it off, I was like, I totally messed it up.
Chriss: But you tried.
Rhett: I tried. I already agreed with Michael before the podcast that I wasn't trying the German pronunciation. We were just going with straight Michael. So welcome, guys. Thanks for joining us today. We are going to be talking about the PPIM paper, which is the development and validation of a novel gas crack detection tool using guided waves. the driver for this a little bit before I ask you guys. Because the paper specifically mentions two drivers. So this was a paper co-written between Enbridge and NDT. And the drivers for this new technology were founded in a 2018 rupture that we have not yet covered on Failure Files, Chris, but maybe we need to.
Chris: Maybe now this is a plug for it.
Rhett: Which is the Shelley British Columbia failure. And apparently that failure prompted Enbridge to begin significantly using EMAT technology, which they then went and started running in all of their lines. After several years of running the EMAT technology, they identified a shortcoming. While they were having successes running the technology, they had a significant shortcoming, which was documented in an IPC paper. And Chris, that shortcoming, as quoted in the publication, is historically about 80% of the excavations triggered by EMAT, revealed either no detectable features or anomalies unrelated to cracking, resulting in a high false positive rate.
Chris: Are you going to put a bunch of caveats on this now?
Rhett: Oh, you can. I'm just reading straight from the thing and I was surprised, but that was the driver. I mean, image's driver was failure. We need to get in a gas crack detection using technology on hand. Technology on hand has an inadequate POI and is driving us to dig more than we want to. We need something else. enter today's technology yeah
Chris: So here's the very quick caveat to the audience right is it's let's assume there's about four emat technologies on the market available in north america and of those four none of them operate the same way so it's not like say one mfl is like this mfl these are completely different ili systems so we're not generalizing emats performance with that statement continue
Rhett: Oh, the publication did. I mean, we don't have to, but yeah, the publication did.
Chris: We are not saying that because I actually don't believe that to be true as a point of opinion.
Rhett: It's right here. It's written.
Chris: It's okay. That's fine. That's their opinion.
Rhett: All right. So with that, I'm going to go ahead and, Rogelio, so that's the precursor, I assume, how you guys got contacted, right, and asked to co-develop this technology. From you guys' perspective, is that what the driver felt like? I mean, was your first thing like, hey. we need a new technology or was it you guys were already working on something in the back burner if you don't mind can you shed a little bit behind the curtain on how you guys got pulled into the development of this technology
Rogelio: yeah well safety is is part of ndt that's why we're always pushing uh the the boundaries of technology and the gas market is something new we've always been known from the liquid side so as part of our strategy of growth the gas market was the the place to go so that's why we we already had started some development some previous iterations to enter the crack gas detection market. But then we got this partnership with Enbridge to pair up and invest in the development of new technology to address what they mentioned over there in the paper, right? It's the part of the efficiency of their program. So that's how we got cold and we embarked with them in this development in the past couple of years. to develop this new technology for gas crack detection
Rhett: and it's and it's new just so the i mean i think the audience knows this already but it's new because you can't just take your ultrasonic sensors which are fantastic and drag those into a gas line right because Gas has such a low acoustic impedance that everybody knows you're used to having ultrasonic sensors, the gel, the baby analogy, right? We're always used to having something to send the ultrasonic waves through. And a gas technology, you don't have that, which is why EMAT works readily so well. So I assume that this technology we're talking about is not based on your conventional ultrasonic sensors, but must be based on something different.
Rogelio: Correct. But for that one, I'll leave the expert to talk about it.
Rhett: Oh, well, please, by all means, let's hear from the expert now.
Chris: So, Michael, if you don't mind, this is your first time on the show. Will you introduce yourself a little bit so everybody can get to know who this new cool person is?
Michael: Okay, yeah, sure. Education-wise, I have a degree in physics, but after university, I right away got into the ILI business. So since almost 20 years now, I've been working in R&D for ILI. I have been working on EMAT, conventional ultrasonics, mostly ultrasonic crack detection, also phased array. And now for a couple of years, I've been working on NET's crack inspection service.
Rhett: So we can, without reservation, say that the physicist is the smartest person on the podcast now. I think you're going to get that title for this one.
Chris: Isn't physics where everything begins? So I would say, yeah, that still makes sense too. Well, welcome, Michael. Yes, welcome.
Rhett: So what can you tell us about this new technology? Can you give us a little bit, help me understand, or help the audience understand, how does this technology work? What's different about it?
Michael: Yeah. So conventional ultrasound crack detection, it's dealing with shear waves bouncing 45 degrees between the inner and outer surface. And we have to maintain a certain angle of incidence to get those 45 degrees. And gas, because of the low impedance, the robustness regarding that angle of incident is really really low it's tens of degrees that you would need to maintain an accuracy and if we go to what we call land waves guided waves that robustness increases to the same level we have for conventional ultrasonics and liquids and we have also an increased efficiency in getting the ultrasonic energy into the pipe wall that's why we go from shear waves to birded waves, more specifically lamb waves.
Rhett: So maybe you can help me. I need to ask ignorant questions now. These shear waves and lamb waves, can you help me understand them physically? What do they represent? What's the difference between a shear wave and a lamb wave? Is that an easy question or is that a hard question?
Michael: Medium.
Rhett: Medium. Yeah.
Michael: It's a bit like if you imagine a pipe wall is a room and then you have a torch. Is that the right torch?
Rhett: The flashlight? Yes, torch. I knew. British thing.
Chirs: Got it. We caught on.
Michael: Good company, I guess. You can point your flashlight. right and it travels through room and it's some parts are getting the light and some don't um and lamp waves are what we call guided waves they fill out the whole room so to say the whole available space in our case then uh the pipe wall um i think a good analogy would be also to look at uh laser light uh guided by optical fibers the laser light then being the analogy analogy to to guided waves and waves so they fill out the whole pipe wall And travel very far in the pipe wall. That's the difference.
Rhett: Okay. And it wasn't just that. Maybe this is too early to ask. But that wasn't the only thing that was unique about how you were using the technology, right? So it's a certain type of sound wave. And that sound wave permits you to use an angle of incidence that's... I'm going to use the word more workable if I can. But you also had some other adoptions too, right? Which... come which you did grab as i understand from your ultrasonic background and conventional right and that's you're not using just one mode what's the word it's it's pitch and catch and you're also using pulse echo correct that's what i was going for okay
Michael: yeah sorry um a bit slow here um well it's actually another analogy to um our history in ultrasonics um because if if you remember you see eclipse right it's it does use the shear waves 45 degrees but it also has a pulse echo and a pitch catch or through transmission signal and those two signal modes or measurement modes give us the superior performance and we have actually tried to maintain as much as possible from from this approach so we have a similar sensor arrangement it's a different angle of incidence on the but the robustness is the same as i said But we are also then recording a pulse echo signal and a pitch catch signal. That's actually intentional because then we can basically feed off our existing procedures, algorithms, and also data analysis strategies.
Chris: So Michael, I want to try to see if I can get excited about this. So it uses LAM waves.
You're doing pulse echo pitch catch, which is very similar to the Eclipse technology. So that's fantastic. Let's get into some of the little weeds so people can understand this and try to see what you're saying. So it sounds to us like the transducer does not need to be in contact with the pipe is what we're saying, right? So there is a standoff.
MichaeL
Yes, there's a finite standoff. That's perfect.
Chris: Go ahead, finish.
Michael: Yeah, several tens of millimeters standoff.
Chris: Tens of millimeters or 10 millimeters?
Michael: No, tens, like above 50.
Chris: Oh, wow. Okay. And then the next one is, would you consider this technology to be high resolution or is it just a couple of sensors with like a larger resolution? How would you describe the resolution of the technology?
Michael: Oh, it's definitely high resolution. We do have a higher redundancy. We do see each crack with many sensors at the same time. um and uh with that we get to get our robustness and reliability
Chris: i'm going to go ahead and ask because we already talked about y'all's y 'all's y'all's crack technology on the ut side are y'all leveraging some of the cool findings of um the eclipse technology on this technology too or is that maybe something we'll talk about later on maybe on another show
Michael: yeah i guess it's a bit better equipped the analogies with ucx eclipse
Rogelio: so i can i what we were talking about the high resolution yeah so we are we're detecting the the features with the same or more number of sensors as we would do with liquid high resolution ucx eclipse technology or even proto So that's what we're looking at. We're looking at the repeatability of the measurement that is sub-millimeter. So we're talking about plus minus 20 mil or 0.5 of a millimeter repeatability, which is something that in the liquid world we weren't able to achieve until we went in 2016-17 into the high-resolution UCX Eclipse technologies. Just from the get-go, what we're seeing is that with this technology, we're matching some of those cool advancements that we've had from the liquid side, but already just from the first iteration of the development.
Chris: So even the analysis technique, so if someone were, say, Eclipse, does that approach translate over to Sigma? Is that something they could expect?
Rogelio: There are similarities. So the strategy we're following with the analysis teams is that we're cross-training the liquid crack analyst team into this new technology. In fact, we're using the same analysis software database in the background. Visually, the data looks quite similar for them up to the point that for any of the crack analysts, just to highlight an area of interest without having any training with a new technology, they can easily say, well, that doesn't look right because of the similarity.
Chris: I think you guys are pretty excited about that.
Rogelio: Oh, yeah. Once we saw the data, yes, there we were.
Rhett: So it's great. So we went right through the whole measurement principle and the development, but everybody likes a good story. So I want to make sure that we set the stage, right? You guys have this measurement principle. You have this technology. And this is what the paper is really about. Can you take us on the development journey of how you guys proved it up? Because I thought it was really cool, right? So we start with a measurement technology, which is integrated with pitch catch and the lamb and shear waves, like you mentioned. And then you take the sensors and you put them on a tool. If you don't mind, tell me a little bit about the validation that was presented in this paper.
Michael: Yeah. So. um basically um I think it's well represented by by a cycle of tests and validations that we go um through and in each step or each stage we increase the complexity and getting closer to the reality of an inspection um and then the main piece in the whole chain with uh sensors and signal processing data analysis it all begins with the sensor we did have a first generation with which we were able to uh get to a level of proof of concept that we are insensitive to corrosion for example less sensitive to laminations and things like that but the sensitivities of the signal strength in essence wasn't good enough so the whole piece starts out with being able to test single transducers in a static setting a small pressure pot we then move on to an scanner so we can scan plates with notches cracks corrosion artificial laminations and whatnot and then whole assemblies of transducers which we call a skid that can be scanned inside a pipe it's all happening in nitrogen so far because it's easy to handle it's not dangerous to deal with it's inert and it goes up to the point where we are able in a really large chamber to put a complete sensor module into a six meter spool and move it back and forth. That's then the validation of the measurement principle, but all the pieces afterwards, including data analysis. So that's already...
Rhett: I want to pause you there, Michael, because I did appreciate this. Because you always talk about sensors, right? Like tools come down to the sensors. And one of the things that you guys did is you actually took the whole sensor package and put it in like a hyperbaric pressure chamber just to make sure... that it could survive the pressures it would see in a gas pipeline. I don't know why that struck me, but I was like, oh, that's so cool to actually have to take the sensor package itself and make sure that it'll survive the conditions of a pipeline. Like it seems elementary, but then it got me thinking like, oh my God, how bad would the first test have been if the whole sensor package would have collapsed? And I thought it was, again, I just want the audience to appreciate that the publication takes us through the details of... know, you mentioned small scale to integrate it on a slightly bigger scale to integrate it on the hyperbaric chamber before you get to the next one. And I interrupted you right before you dropped the next one. The next test is probably the apex of it all, the climax.
Michael: Yeah. So this is all up to the point that you talked about that we put a whole sensor module in a pressurized chamber with a six meter spool. It's all pull tests. So there's some mechanics that moves the tool at relatively low speed. and it's all happening in nitrogen just safety so what we then did in the next step is really to design and build a test loop for our 30 inch tool so this is was done in the UK and the whole test rig is as i said 30 inch pipe i think it's roughly 270 feet long We've put the 30 -inch pipe into 36-inch pipes to be able to go to high pressures of, well, PSI. 900-something PSI. I think 940 PSI. And even still have cracks that are like 60, 70, 80% deep without blowing the system up. So we did hydrostatic tests before. and all these things. There are a couple of fans that generate the flow of up to one meter per second. We do use natural gas, but we did use natural gas in that system. And the whole tool train moves as a real pipeline driven by the gas flow.
Rhett: It's so cool. So Chris, I want to rephrase that for the audience. We've all heard of liquid flow loops. There's one at PRCI at the TDC right down the road. I don't think I had ever heard of a... A live gas flow loop. Have you seen a live gas flow loop? I actually thought that was so cool.
Chris: We haven't used one, at least in any of our experience. And maybe you'll give us some background, Michael. Was that flow loop set up for this project or was it existing and it was available?
Michael No, no, no.
Rhett: They built it.
Chris: That is awesome.
Rhett: I looked for this project. Because I told him, I was like, I didn't know there was a gas flow loop. And Rogelio was like, yeah, we built it.
Chris: And it's a 30 inch, right? This is a big boy.
Rhett: It's a big flow loop to drive a free swimming tool. All right. So before we go on break, I want to describe. So you guys built a flow loop. And what's really cool is I'm going to set the stage and leave the audience with a cliffhanger. They built a test loop that had over 300 features, but it wasn't all 300 detectable features. I'm sorry. I actually got the wrong. It's 173. That's right here. 120 of the features were in scope. Right. So think things like notches and synthetic cracks. And then they also had 53 features that they considered in scope. So think high pH, SCC, synthetic colonies of SCC, things that they were going for. So they had features that were both out of scope, in scope, kind of doing a full-blown test on the POD and POI. When we get back, we're going to talk with Michael and Rogelio about the performance of that tool. Hang on and we'll be right back.
Rhett: All right, well. We are back with our guests from NDT talking about the development of their novel gas crack detection technology. And yet we didn't know when we were going to release the name. We had this whole conversation before Chris came on about, do we say the name? And I was like, oh, maybe we're going to save it at the end and do something cool. And then Chris just blurted it out in the middle of the last segment, didn't even realize it, right? Which the technology we're talking about, if you haven't heard the word on the street is Sigma, right? No one knows where the name Sigma came from. Right. And it is not to be confused with 6-7 or any of the other trendy TikTok things that are currently out there.
Chris: At least in North America. North America, these kids have all these random things they say. And I'll say this. So I teach fifth graders on Tuesdays for some extracurricular stuff. It's faith-based. And they were all blurting around this word sigma, sigma, sigma. So I did this on purpose. I put them aside. And I said, each of you are going to go through and tell me what Sigma means. And I had nine kids in the room.
Rhett: And one of them was like, that's a novel gas crack detection technology from NDT. Right? Just one of them?
Chris: We are in Houston, so it's very possible one of them works for NDT. But anyways, none of them said the same thing. So I'm sure you guys can define for us what Sigma means because none of the kids on my Tuesday class were able to.
Rhett: Okay. So I want to take the audience back. We talked a little bit about that big 30-inch flow loop, which, again, I kind of want to see this at some point. That thing sounds really cool. I want to give you guys the parameters before I turn it back over to Argelio. The inspection speed in that flow loop was approximately 1.8 miles per hour at a pressure of 652 PSI. The pipe that was used was obviously 30 inches with a nominal wall thickness of 374. I mentioned to you before that it had 173 total samples with 120 that were considered out of scope. So think metal loss, synthetic cracks, things like that. And then 53 actual or synthetic SCC features that were considered in scope. But that 173 is a bit misleading because the paper says that they ran it at least five times to produce 865 unique indications, from which they then developed the specs. And we're going to go through that. So if you don't mind, Rogelio, I don't want to say your specs for you. So will you tell us, how did the tool perform in those five assessments?
Rogelio: So before that, just a comment. What it's written in that paper is just the tip of the iceberg of the tests that we did. We didn't build a test rig for just five. pull through tests or let's say pump tests. We actually recorded around 12,000 or inspected 12,000 features. We varied the pressure, not only the pressure that you have there. So we went down to below 700 PSI. We changed the pink pattern. We changed the actual resolution. We included acoustic protector shields. or not to the sensors, all of them to retrieve data. This was the fastest way for us to try to mature that technology as much as possible, because in comparison to what we would do in the liquid side, we will run it in a water loop. This one's tricky. So we built the rig and then we changed all of those variables. So in total, it was eight tests. where we varied pressures we even varied the medium composition the methane content that we have in there to retrieve all of that data so what we have there in the paper just the tip of the iceberg actually we have way more tests up to this point we've processed around 8 000 of the 12 000 tests and that's what i'm going to be talking uh to you about of what we've seen from those results so
Rhett: So the paper mentioned a level two, API 1163 level two validation, or you're building on that database for your level two validation at this point, I assume.
Rogelio: Yeah, correct. So with this partnership with the operator that we had, we had some target dimensions for POD, POI, depth sizing that we were planning on achieving. So for example, for the POD, our target was... of 157 inches in length by 60 mil in depth at 90 percent pod confidence interval at 95. that was our target our poi target was of 80 percent and for depth sizing actually it was bucket sizing because the main focus was the poi so we said okay let's see if we get the proper amplitude dynamics
Rhett: Bucket sizing So not like, so UT tools, this is where UT tools were like, what a decade ago, it was like 20 to 30% or 20 to 40, 40 to 60, 60 to 80. You started with bucket sizing on the Sigma technology as well.
Rogelio: That was our target.
Rhett: Okay. And what was the minimum detectable feet? The paper says one and a half millimeter depth for up to five millimeters. Is that still where it's at? Or is it, that's the 60 millimeters. Gotcha.
Rogelio: Okay. yeah the minimum the minimum pod is is yeah 1.57 inches long by 60 mil deep and then the accuracy of the depth sizing well that's what we were able to achieve based on the sensor development and the technology itself it was tricky to actually do bucket sizing so we derived a sizing curve that allow us to go up to that plus minus 60 mil tolerance up to 200 mil.
Rhett: So you skip bucket sizing and went straight for discrete
Rogelio: Directly Yeah. So so we got so comparing this with with the liquids technology it we we have a head start of 15 years just just going avoiding the bucket sizing. And this is where the first iteration coming back and comparing it to the liquid side this is equivalent to where the liquid sizing was the first time we actually got a discrete depth. That's where we're starting at the moment. So what we saw throughout the different tests in terms of POD, all of the features that we implemented, you mentioned it before, we have features above the minimum POD, below, synthetic cracks, in-service, machined, all of them. the lot so all those different combinations the the features that simulate sec with or without corrosion we detected 98% of them but not not only the ones within those minimum dimensions we actually detected the ones that were below those minimum dimensions even at half of the of the length with the same depth we still detected with above 90%. So detection for linear features, our single cracks or our colonies, we achieved the target. For the nonlinear features, because we also had corrosion, we had general corrosion, we had pits. We had pinholes, sharp edge corrosion, simulating actual slottings, parallel laminations and surface breaking laminations. This is where the technology comes into play. And we have really cool things here. So for example, corrosion, general corrosion, they were not detected. None of them. Across all different tests and variable changes, none of them were Detected.
Rhett: Wait, wait. Did you have any sharp edged corrosion?
Rogelio: we did have the sharp edge and those ones those ones were the let's let's put it this way from the metalos features those were the ones that were detected at a 70 pod uh and then during the poi phase Only 30% were called out as notches or cracks. So the rest, there was sufficient information in the data to discriminate that it was a corrosion. But we only struggled in terms of POI and false calls with the shark patch corrosions, which if you start thinking about it from the integrity side, how are you going to assess those features? They're right at that borderline between being assessed as a crack or not. But those were the only ones that we detected. We have some of those sharp-edged corrosion that live within a general corrosion. We don't see the general corrosion. We only see the thin part of the actual sloping. That would be equivalent to your gouge or a groove inside the pipeline. So those were the ones that actually showed up in the data. But the ones that are a little bit more volumetric, bigger, general corrosion, the pits, the pinholes, they're not in the data.
Which brings us to one of the cool things that we have where the technology, the POI, is being linked to the POD. Our analysis team cannot misclassify something if they don't see it in the data. So we cannot call out a corrosion and misclassify it as a crack because it's not there. It's simply not in the data. And that's the advantage of using the pulse echo. Because pulse echo is being used for the detection piece, and then the transmitted data is used for the sizing. So in case we have a crack within a corrosion, the depth sizing is going to be the full penetration. So it's going to be your corrosion plus your crack. In the pulse echo data, you won't see that there is a corrosion. In the transmitted data, you'll get the full penetration. But based on how the procedure is built for detection, you won't see it. If there was only a corrosion, it won't show up.
Rhett: So the precursor to this whole technology was a POI-driven problem. And I was going to ask you about which one's more challenging, POD or POI, but you got to it before I could even get there, which is the two of them are so closely linked. I just want to repeat what I heard because I thought that that's pretty cool. It's not that an analyst has to see and classify corrosion correctly. The way that your technology is working right now is you're not seeing the corrosion at all, such that there is no need to classify the corrosion based on the measurement principle. And that's definitely novel. I look forward to seeing how that behaves out in the wild once it gets out in the field, which really logically leads to our next question. So I'm sure a lot of the audience... who are familiar with gas crack detection and the challenges it presents, wants to know, where are you guys at right now? What should we expect to hear from NDT in the coming year? I mean, obviously, clearly you have a 30-inch tool, right? At least one 30-inch tool, which has been in a gas flow loop. I assume you have that tool. What else? Where do you see it on the market spectrum right now?
Rogelio: Our strategy as a company, we need to maximize the fleet, right? There's no service just with one tool. So by the end of the year, we're going to have four tools available. Well, we brought back a concept that we had in the past that we call the tool factory when we actually doubled our tool fleet for liquids some years ago. So we brought that one back. We're maximizing the tool build so we can offer a service and we don't have issues with booking because at this point today, getting the tool or the couple of tools that are going to come out in the next month, it's tricky. There's some slots here and there, but pretty much they're already commissioned. So at least for the next years, we're going to be doubling. the fleet kind of like year over year and uh and yeah we're going to be going to to the to the diameters where the market is the big the biggest chunk of the markets so it's going to be it's it's
Rhett: a client should expect to have tools and certain diameters available by the end of the year and more diameters by the end of 2027 that's what i'm hearing yeah cool maybe
Chris: Just to kind of link that a little bit together, right? Because you talked about validation and you talked about all the inputs and now you're talking about tool reproduction. So there's two words that I always like to bring up. here recently, and I'll just ask you what you think y'all's status is with it. So it sounds like you've got a good grasp on the essential variables is what it sounds like. Can we just hear your thoughts on that? Again, how do you tie what you found from the flow loop into now we're able to go in and say we're ready to start producing tools? You've talked about PODs, POIs, and sizing, but we haven't used the word essential variables. Where do you guys feel like you stand with understanding the essential variables?
Rogelio: So we have established an operational range, an operational envelope for the tool based on what we know today. Many of the variables are still going to be expanded as we gain experience in the market. What we've seen is that, for example, you could have a lower pressure within our operational range, and that will have an impact in terms of the signal-to-noise ratio but that could be offset by your medium composition that could offset and then something that might be outside of that standard operational range then you could say well you know what we should be able to inspect and get this type of of results so we have this list of essential variables We understand the impact, what they're doing individually based on what we saw in that test loop. We know that when one goes towards an extreme, the other one, how we can tweak the other one to compensate what's happening. There's still some variables to consider. One of those is the coding. For example, the only variable we didn't have in the test rig was coding. And we know that different types of coding are going to provide different types of attenuation to the signal. So we have some experience. We've actually ran in service pipelines already. We've processed the results. We've seen as well that the technology is robust enough that even in overspeed going up to 70%, the performance specification that we observed in the test rig can hold. We still can get the same PODs, POIs and depth sizing, and that's already in non-optimal conditions. So it has been, even for us, a surprise that the technology itself is more robust. than what we originally thought. We thought being gas, it's going to be more black and white kind of thing. But we've seen that the combination of all these essential variables actually provides different shades of gray that is helping us address more pipelines.
Chris: So what I heard, Rogelio, was it's kind of this whole podcast we've heard is it's NDT. Obviously, with Enbridge being a catalyst, has really put the organization forward to say, we're going to make this gas crack tool happen. And you have enough of it ready to where now we need the industry to step in and say, let's start understanding, help you guys understand what the essential variables on this tool is.
Rogelio: Yeah, this is where the operators that use this technology for the first time, they're going to have to walk with us. Because even from their side, we're going to have to understand what this technology can provide to them to close the gap in their integrity management program, how it complements the data they already have. So it's not that Sigma is going to come and solve all of their problems. They also need to learn how to use it, what to expect from it, when it's good to use it, when not. And all of that just... depends on how I actually want to use it.
Rhett: Wait, you mean if my gas composition changes, the tool works different?
Chris: No, he didn't say that.
Rhett: He did say the gas composition affects it. I could see a whole new learning curve on that.
Chris: I think the whole point of that was to say, again, you guys have done a great job of putting
something out there because...
Rhett: I mean, I'm excited to see where it goes.
Chris: There's obviously demand for it. You guys have done a great job. You've put it in a diameter size. It's very attractive. And you're out there. Now it's time for the market to respond to that and say, hey, guys.
Rhett: I'm going to bring this full circle, Chris. If ultrasonic technology was the iPhone,
this is like the iPad. Think about how much the iPad leveraged the learnings from the iPhone when it was adopted. Oh, that's where you're going with it. Yeah, I mean, the iPad scaled up. much quicker compared to the iPhone. And I thought that was the cool thing is hearing how they actually leverage some of their, their learnings and things like, just like, Hey, we'll start with buckets. Cause that's where we need to start. And then by the end of it, they're like, throw the buckets out. We're good. Let's go straight to screen. So, um, I'm really excited to see where you guys take this. And, um, you know, I also want to give a shout out. Michael, Rogelio, this is the first time we've had four guests on a podcast. And I thought it went extremely well. Thank you guys so much for joining us and taking the time to share this information with the industry. Really appreciate it. Cool. Thank you, guys. Thank you. To our audience, I want to say thanks for joining us on Pipeline Things. And again, we look forward to talking to you again in the next two weeks. I'm your host, Rhett Dotson, and we'll see you again later.
This episode was executively produced by Sarah Etier. Thank you to our guests, Michael Haas and Rogelio Guajaro from NDT Technology, PPIM, for the paper development and validation of a novel gas crack detection tool using guided waves and for form marketing, allowing us to use their venue to film.
