This week’s Pipeliners Podcast episode features first-time guest Tod Barker of T.D. Williamson discussing how to use dig feedback to improve your pipeline integrity program.

In this episode, you will learn about what dig feedback is, a new technology called Coherent Light Profilometry and how it improves pipe measurement accuracy, and the difference between data gathering and data verification. You will also learn about the benefits of effectively gathering and verifying data from a dig.

Pipeline Integrity: Show Notes, Links, and Insider Terms

• Tod Barker is a Senior Product Manager at T.D. Williamson. Connect with Tod on LinkedIn.
  • T.D. Williamson (TDW) serves the gathering, transmission, and distribution sectors of the pipeline industry with a global portfolio of products and services, including advanced isolation, integrated pigging, and integrity assessment solutions.
    • The MDS tool developed by T.D. Williamson detects, characterizes, and sizes interacting threats to pipeline integrity, such as mechanical damage. The tool integrates multiple ILI technologies on a single platform, overcomes the threat detection gaps of single magnetic flux leakage (MFL) based tools, reduces risk, increases the accuracy of results, and provides comprehensive information about features previously undetected by other ILI tools.
• Integrity Management (Pipeline Integrity Management) is a systematic approach to operate and manage pipelines in a safe manner that complies with PHMSA regulations.
• Metallurgist is a materials scientist who specializes in metals such as steel, aluminum, iron, and copper.
• ILI (Inline Inspection) is a method to assess the integrity and condition of a pipe by determining the existence of cracks, deformities, or other structural issues that could cause a leak.
  • Listen to the Pipeliners Podcast series on ILI with Marc Lamontagne.
• Dig Feedback refers to the results captured at a dig site where pipe is exposed for review by an NDE evaluation crew.
  • NDE (non-destructive evaluation) uses quantitative measurements to identify a defect in a pipe. Measurements focus on the size, shape, and orientation of the defect and take into account the physical characteristics of the pipe.
• Ultrasonic inline inspection uses sound waves to send a signal into a steel pipe to detect the presence of corrosion or cracks within the pipe.
• Coherent Light Profilometry is the acquisition of high-resolution, three-dimensional surface data of complex entities such as pipelines using reflected light to measure the profile of a surface.  
• SCADA (Supervisory Control and Data Acquisition) is a system of software and technology that allows pipeliners to control processes locally or at remote location. 
• PHMSA (Pipeline and Hazardous Materials Safety Administration) is responsible for providing pipeline safety oversight through regulatory rulemaking, NTSB recommendations, and other important functions to protect people and the environment through the safe transportation of energy and other hazardous materials.
• PRCI (Pipeline Research Council International) is a community of the world’s leading pipeline companies, vendors, service providers, equipment manufacturers, and other organizations supporting the oil and gas industry.
  • Listen to Pipeliners Podcast episode 54 with PRCI president Cliff Johnson on how the PRCI has developed a data hub to store information from across the pipeline industry.

Pipeline Integrity: Full Episode Transcript

Russel Treat:  Welcome to the Pipeliners Podcast, episode 143, sponsored by Burns & McDonnell, delivering pipeline projects with an integrated construction and design mindset, connecting all the project elements, design, procurement, sequencing at the site. Burns & McDonnell uses its vast knowledge and the latest technology with an ownership commitment to safely deliver innovative, quality projects. Learn how Burns & McDonnell is on-site through it all at burnsmcd.com.

[background music]

Announcer:  The Pipeliners Podcast, where professionals, Bubba geeks, and industry insiders share their knowledge and experience about technology, projects, and pipeline operations. Now your host, Russel Treat.

Russel:  Thanks for listening to the Pipeliners Podcast. I appreciate you taking the time. To show that appreciation, we give away a customized YETI tumbler to one listener each episode. This week, our winner is Michial Poteet with NGL Energy Partners. Congratulations, your YETI is on its way. To learn how you can learn this signature prize pack, stick around until the end of the episode.

This week we have Tod Barker with T.D. Williamson to talk about using dig feedback to improve your pipeline integrity program. Tod, welcome to the Pipeliners Podcast.

Tod Barker:  Thanks for having me on, Russel. I’m excited to be here.

Russel:  I asked you to come on to talk about using dig feedback to improve pipeline integrity, but before we jump in, maybe you could tell the listeners a bit about yourself, and your background, and how long you’ve been working in pipelining.

Tod:  Great, yeah. I’ve been working for T.D. Williamson for 15 years. That’s the length of my pipeline career. I got into it in engineering designing inline inspection tools. Prior to that, I was working in engineering in the airline industry.

I’ve transitioned through mechanical engineering into product management. Now, I’m responsible for the business side of it for T.D. Williamson’s corrosion, geometry, mapping, and customer-facing software technologies.

Russel:  Cool. Did you start out as a metallurgist or did you start out as a mechanical engineer?

Tod:  A mechanical engineer.

Russel:  Most of the folks I know that work around the ILI tools, they tend to have more of a metallurgy background, at least in my experience. My experience’s very limited so I don’t know that that means anything.

Tod:  That’s a great point. I worked in sheet metal for a number of years. It wasn’t actually metallurgist, but I did work with a lot of different types of metal and properties.

Russel:  My degree’s in civil structural, so I got a lot of…It’s interesting when you talk about this because we all are working around the metal but we all learn about it from a different perspective.

Tod:  Yeah, exactly.

Russel:  Just talking about all the different things you could know about metal would probably be a pretty good podcast. That one right there, that would probably peg the geek meter. I should probably figure out how to organize that.

Tod:  That’s one I’d definitely listen to. If I could add, I’d participate.

Russel:  There you go. Let’s dive in. First question for you is what is dig feedback?

Tod:  Dig feedback are results that are captured at a dig site where they expose the pipe. They’ll dig it up. They’ll take off the coating and measure an anomaly or some type of feature and provide that information to another party. Usually this is done by an NDE or a non-destructive evaluation crew. They’re providing it to an ILI vendor and usually the pipeline operator as well.

Russel:  Man, I only know about this stuff notionally. I’m familiar, of course, with what NDE is. I would think that there’s a fairly wide variety of the equipment that’s used to gather this data from a ditch. Would that be right?

Tod:  That’s exactly right, Russel. Years ago, way before my time, they were using, and maybe even well into my time, they were using manual pit gauges, which is just basically a card with a projection and an indicator that sweeps through an arc to tell you how deep it is. It was basically just analog type.

They’ve gone to digital pit depth gauges. There’s also handheld UT, ultrasonic technology, to measure wall thickness. More recently, they’ve gone to the laser scanning of the pipe, which is very, very accurate.

Those others all have operator error in tolerances associated with them. The other thing is are you in the right spot? The laser scan captures everything so it’s much easier to correlate the two.

Russel:  That’s the thing that’s really interesting when you start talking about this and you start really unpacking the complexity. My experience with NDE is on flat, level surfaces. Most of the equipment, again at least in my experience, is designed to work on a flat, level surface.

You start talking about I’ve got to dig around a pipe, and a pipe is round, and I’ve got to hold it up against the pipe versus sit it on a flat surface, that starts adding a lot of potential error and difficulty in the actual measurement itself.

Tod:  Exactly. A lot of times when you’re talking about a flat plate, you’re talking about in a manufacturing environment, or a mill, or something like that. You can actually see and touch what you’re looking at. Knowing where that pit is, let’s say, is really easy because you can see the pit. I can see the pit. We measured the pit.

In the pipe when you expose a pipe, not only do you have the complexities associated with the curve, but you also have many different pits, and are you in exactly the right place. You could be off by two inches in a couple of pits, and that’s also a challenge.

Russel:  My understanding of dig practice is, again it’s notional at best, but what understanding I have is typically in this day and age when somebody does a dig, they’ve identified something by running an ILI tool, and they’re going to expose a piece of pipe from weld to weld so one string of pipe.

They’re going to run this kind of equipment on that pipe, but they’re trying to true up and get very accurate measurements.

You start thinking about a piece of pipe is going to be 40 or 80 foot long and I’m trying to know within a tenth of an inch where I am on that piece of pipe, and then I also got to know where I am in terms of up and down, that starts getting, notionally that seems straightforward, but in practice that’s very much non-trivial.

Tod:  It is. There’s some great crews out there in the industry that work on the NDE stuff. They spend years becoming experts, but it’s not trivial as you’re saying. The laser scan has really helped because you can see an image of it when you get out. It’s not just you measured a pit or you measured a couple of pits, and you were six feet from the weld.

Russel:  Exactly. The question that, we already talked about this a little bit, but a lot of the equipment that’s being used is different. I would expect the nature of the raw data that you get following measurement is somewhat dependent on both the equipment that’s being used and the crew that’s doing the measurement. How do you pull all that together?

Tod:  There’s a difference between dig feedback and dig verification. Dig verification program or dig verification is once the data’s been verified so you’ve correlated it between the two, and there’s been some agreement that yes, we’re talking about this pit right here because corrosion’s easy to discuss, right?

We’ve got this pit here, and this is the one that we’re seeing in the ILI data, and this is the one that you’re measuring in the NDE data, and now we’ve agreed so now we can see how accurate both sets of data are.

Until then, once you have the measurements that they gather at the ditch, that, you know when they dig up the pipe, that ends up being the dig feedback, but then you need the dig verification. You need some kind of correlation between the two.

Russel:  I’m thinking about what you’re saying, the distinction between data gathering and data verification. Tell me a little bit about how — the mechanics.

How do you actually do that? What’s the nature of what you get off of this equipment and then what’s the nature of what are you doing as you’re going through the process to take this raw data and get it to verified data.

Tod:  The information that you’re getting off of the equipment is much, hopefully as much information as you can. They’ll measure corrosion pits. They’ll measure, and when I talk about pits, I’m oversimplifying because they’re usually a group of pits.

You’ll measure, there’s dents that are being measured. They’re measuring from a weld as you indicated. They’ll usually be where the pipe was tied to the other piece of pipe. You’re measuring from a weld. There’s fittings, and tees, and all kinds of stuff they’ll use to locate this in the pipeline system and provide that information.

They’ll provide it to the pipeline operator themselves and typically to the ILI vendor as well.

Russel:  My background’s in SCADA and automation measurement, all that kind of stuff. I can talk in great detail about how an instrument puts out an electrical signal, and that electrical signal gets translated into a block, and that block gets scaled to an engineering unit, right?

Tod:  Right.

Russel:  I would assume with a raw instruments you’ve got a similar process, but the output is going to be just raw numbers. What I’m getting at is this. Let me unpack the question a little bit different, Tod. I’m trying to understand in the details. I’ve got this NDE, this nondestructive evaluation, piece of equipment, and I’m holding it up to the next pipe, up to the pipe, and I’m getting readings.

Am I writing those readings down? Am I taking that as a data stream into a laptop? That reading, does it have location data? I’ve got to add location data to it. How does all that happen?

Tod:  That’s a great question. It depends. All of the above. Depending on the type of equipment you’re using to measure this, you could be just not hooked to anything, just the manual type pit gauge all the way to the laser scan.

The laser scan would typically be hooked to a laptop, and they’d be scanning a larger area. That’s one of the reasons why us, as ILI vendors, love the laser scan is we get a lot of information about the area instead of just whatever the NDE team wants to provide. That’s recorded.

They do have to add the location information, but they usually can add that before they go down to the ditch, add the location information, because they’ve gone there for a specific reason as you alluded to earlier in your podcast. Then they can provide that laser scan, and you end up with a lot of features in that laser scan that are all useful.

Russel:  One of the things I try to do in this podcast, Tod, is I try to help people who have never worked a dig, for example, or who have never run an ILI tool understand what’s actually going on in the field.

Those of us that are familiar with a particular topic, we tend to, it’s easy for us to just make a whole bunch of assumptions because there’s a lot of it that’s just well understood within our community of people.

When you start talking about a dig in particular, just getting the dig done, that’s a whole topic just to itself. Getting the dig done, and getting the pipe exposed, and doing that in a way that it’s safe, and all that is, all that’s non-trivial.

Once you have all that, you’ve got this crew of people that are using various kinds of measurement techniques from a tape measure, and a chalk marker, and a camera to advanced laser scanning. I would assume there’s even people, if they’re not doing it yet, there’s probably people looking at building robots that can sit on the pipe and gather all this data and sweep it.

Tod:  The laser scan’s very similar to that. That’s a good point. They used to refer to the tape measure. They’d refer to chaining it in for the distance when they exposed the pipe.

Russel:  I ran chains when I was doing surveying in college. I know exactly what that is.

Tod:  I’m sure you do if you did that.

Russel:  Then you take the outdoor temperature and you correct your chain for the outdoor temperature. [laughs]

Tod:  Right. Thermal expansion.

Russel:  Again, I’ve been saying this a lot lately as I’ve been talking about some of these technical things, but all this is easy until you know enough about it.

Tod:  I love that expression.

Russel:  Once this has been done, so the verification process is basically taking all the data regardless of how it’s been collected and putting it together in a way that pretty much anybody that works in integrity management could look at it, understand it, and believe it. Is that a fair comment?

Tod:  Yeah. HighEye level, that’s exactly what you’re looking for. You want enough information.

Along those lines, one thing that you alluded to earlier that I’d like to just touch on real briefly, you talked about a camera and pictures of, they’ll put an arrow. Sometimes it’s just an 8.5 x 11 sheet of paper they draw an arrow on, and they’ll put that on the pipe, and it’ll show the direction of the flow so you know up and downstream.

You’ll have the date, and they do capture the temperature, to your point. That’s all relevant information, especially when you’re talking about calibrating the equipment, and then getting the complete information as we’ve been talking about for the last few minutes.

One thing that, again not to keep beating or harping on this laser scan, but one thing that ILI vendors really need, and this is all ILI vendors, is we need all dig feedback.

A lot of times in the past pipeline operators have liked to tell us you missed this one pit or this one was a little bit out of size. If we changed our model to just the wrong stuff, the model wouldn’t fit anymore. We need all of it to improve the model.

Russel:  There’s two things about data analytics that are generally not well understood. The first is I need organized, consistent, accurate, verified data.

Tod:  I agree completely.

Russel:  The second thing is I need a lot of it because I can only find things that I’ve seen before.

Tod:  Exactly. That’s something that I really wanted to touch on too, so I’m glad that you brought that up is somebody’ll bring, let’s say you bring 100 anomalies at a 100 pits sized back to us for dig feedback, and you say now can your tool be that much better, this ILI tool.

There are pipelines out there that we find 100,000 pits in one pipeline. You bring me 100 pieces of dig feedback, that’s a really small bucket in the whole global scheme of things.

Russel:  That’s right. I need a very large sample set, and I need to know what it is I’m looking for.

Tod:  Yes.

Russel:  That’s what’s required for these big analytical engines to work.

Tod:  Yeah.

Russel:  If I do a really good job of gathering and verifying data from the dig, what are some of the benefits of that?

Tod:  Some of the benefits are, let’s say your dig was based on preliminary data. You ran an inline inspection tool. You got the information, the data out of it, analysts started looking at it, and they said we’ve got some stuff here that maybe, to your point, you need to have seen what you’re looking at before.

Maybe there’s something we haven’t seen before, or the analysts, it just isn’t making sense. It’s out of context. It’s like this is normally here, but it’s looking different so I’m not sure what that is.

For whatever reason, they can provide the pipeline operator a list of some preliminary places to go dig and say we’ve got this, let’s say we thought we found narrow axial anomalies or narrow axial corrosion on the long seam. It’s like okay, is this something that’s really bad, or is it something that’s not?

We can give you four or five locations. Go dig. Come back. There’ll be pictures. There will be measurements and more information from the NDE team, probably UT wall thickness information. Like you said, there’s the temperature and all the other information. They’ll even take the pH of the soil.

We look at that information. We say okay, maybe in this circumstance what we’re talking about here isn’t axial corrosion, you know narrow corrosion in the long seam, but let’sit’s, say, preferential. It’s just a grind mark from the mill or something from when they welded it at the mill. It’s been hydro tested, and it’s now considered stable.

Instead of having 500 axial anomalies in your long seam, you don’t have any, and you’ve got a little bit of corrosion.

Russel:  That’s basically what I would call truthing the data.

Tod:  Yeah. I like that term.

Russel:  I’m using my, what I’m doing in the dig as establishing truth. Then I’m correlating that to what the ILI tool is seeing, and seeing if the ILI tool is telling me truth. Is that a fair way to frame it?

Tod:  Yeah, it is. It’s the truth data. You just have to understand that all data’s valuable, but some data’s even useful.

Along those lines, the truth data, that’s why we need to validate the dig feedback to make sure that it really is the truth. Once we have that, we’ve trained, if you will, the system, and we can now give you a lot better, you being the pipeline operator, a lot better final report because we did that work at the preliminary.

Russel:  Exactly. This has come up on some of the other podcasts that we’ve done where people are talking about the need for information sharing. I’ve talked a little bit about some of the things that are going on with PHMSA and PRCI around this kind of subject.

Certainly, there’s value in doing laboratory kind of work, but what you find in the field in real-world practice is just different. There’s more variations and more complexity.

Tod:  Definitely. That’s actually one of the biggest challenges with improving the accuracy or getting highly accurate inline inspection tools is we need that direct correlation which is really easy if you have, in the laboratory, you have a pull test facility or something like that, and you can take these anomalies, and measure them, and run the tool across them, maybe multiple times, get all kinds of positions, and you end up having really good information.

The problem is you usually create these anomalies in the laboratory, and that doesn’t look like what you see in the field. You’ve got to make sure that what you’re training your model on is actual what you’re going to see in the field or it becomes less valuable. Maybe you have to extrapolate. No one really wants to do that. That would be going backwards.

We’ve got a lot of dig data. We’ve got a lot of inline inspection data. Using that is the key to everyone getting better.

Russel:  To the extent we as an industry can more broadly share this information, we’re going to all get better as an industry.

Tod:  I agree. That’s a huge challenge. We’ve talked about several of the challenges, but there’s probably more. Some of them are you’ve got to take the dig feedback and have it verified so you have some dig verification type data. One ILI vendor may use it in a different way than another. Then there’s the proprietary systems that the data collected from each ILI system has.

It’s something that’s interesting, and it could be valuable if we find a way to meld all that down and make it all equal.

Russel:  Equitable is probably a better word than equal, but yes. The challenge is there’s various parties that have an interest. Those interests all have their own unique special interests. While there’s agreement in principal that there’s value in data sharing, there’s also a lot of risk associated with that. You got to figure out a way to do that and to the extent possible remove or mitigate the risk.

We’re in the early days of all of this. Certainly, particularly the idea of getting, I could see where we as an industry get to a day where there is a generally accepted practice for how you gather data where that’s done through robotics package and a whole bunch of sensors, and it gives you a standard raw data set. Then everybody gets to use that to determine what they need to determine from it.

Tod:  That would be this robot in the ditch?

Russel:  Yeah.

Tod:  We’re getting closer to that. Like I said, these laser scan systems are, and it’s really come on the last few years, coherent light profilometry.

Russel:  What now? Say that again. [laughs]

Tod:  Coherent light profilometry. Easy for me to say, huh?

Russel:  Profilometry. I got coherent light. I didn’t get profilometry. You’re going to have to explain that to me.

Tod:  It’s just like what it says. You’re profiling. In this case, you’re looking at, I’m trying to think of the easiest way.

Russel:  I’ll make a stab, Tod. I’m sorry. I would say I don’t mean to put you on the spot, but I kind of do. [laughs] This is my inner nerd coming out. Coherent light means I’ve got a light source that is a very controlled kind of outbound light, which allows me to look at how that light is reflected back. By looking at how the light’s reflected back, I can get a profile of a surface.

If it comes straight back to me at 100 percent, that tells me something. If it comes straight back in 50 percent versus, tells me it’s reflected at an angle of 30 degrees and such. Profilometry is basically metering the profile of the surface of the pipe.

Tod:  Yeah, exactly. It was initially used for surface roughness. Where we’re talking about such small features here, it’s been found to be very useful for corrosion, and small dents, and pitting, and that type of stuff.

Russel:  There’s a lot of robots already that we run inside the pipe, right? We call those ILI tools.

Tod:  Right.

Russel:  I could see where we create robots that run on the outside of the pipe and use some of the same technologies. Then you could really correlate the data more effectively. Again, that’s completely notional in my mind, and I realize there’s a lot of complexities taking data from the inside surface versus taking data from the outside surface.

I could certainly see where the work we currently do in a ditch with a team and a lot of different kinds of handheld equipment ends up on a package that’s run as a tool on the outside of the pipe. If people are working on that and are ready to share and talk about it, man, call me because I want to know about it.

Tod:  I do, too. That would be very interesting.

Russel:  There’s probably some super-smart engineer in a garage someplace that’s trying this already.

Tod:  I think there are several that are working towards that direction. Like I said, just the improvements that they’ve made in the last 10 plus years has been amazing. 10 more years, it wouldn’t surprise me.

Russel:  I’ve had this conversation with several people. It’s going to be very interesting to see what our business looks like in another 10 years. We have a lot of technologies that are evolving very quickly.

That’s probably a good way to transition into the next question I wanted to ask, and that is what’s happening with this technology. What are people working on? What are the things that are either brand new? We’ve already talked about the laser device. The, what did you say, you called that consistent light…

Tod:  Coherent light profilometry.

Russel:  Coherent light profilometry. I’m going to have to say that 10 times fast after a couple of cocktails. That’ll be fun.

Tod:  There you go.

Russel:  I guess one of the things you do when you’re a podcast host in this domain, you collect really big, complicated words.

Tod:  [laughs]

Russel:  What’s happening with the technology? What’s some of the new stuff that’s out there that we haven’t talked about and what’s coming?

Tod:  Some of the things that we haven’t talked about that are exciting are, and these are recent things we’ve worked on, when we get dig feedback, sometimes it’s things, and we’ve touched, danced around this if you will, but it’s things that we haven’t seen before that there really isn’t a standard model out there anywhere in the industry for.

Some of the things I’m talking about, and these do further ILI development and the data development, is wrinkle bends with cracks.

A wrinkle bend is in the ’40s, and I think it was totally done away with by the ’50s, but in the ’30s and ’40s you’re putting in a pipe, you get there, and you need a little more angle than you really had, they’d put chains around the pipe, and bend it around a tree, and put it down in there til it fit. You had these wrinkles in there.

They’re seeing cracks in these wrinkles. They’re not smooth on the inside of the pipe. You run that ILI tool across there. It was challenging to find these cracks. They didn’t realize until more recently. They were doing digs and found, “Oh, we have cracks in our wrinkle bends. Can somebody find that?”

They took that dig feedback and we said, “Yeah, we think we have a tool that’s capable of doing that.” TDW’s MDS tool does really well with that. Then, the last couple of years there’s been…

Puddle welds have been brought up where this is, again, from the ’50s, ’40s, ’30s. They would find corrosion in the pipe and they’d just go out and take an oxyacetylene welder and fill that corrosion in. At that point they said, “Okay, it’s stable. It’s not corrosion anymore.”

Again, I think that practice was abandoned in, certainly before 1970. There’s pipelines out there that have these puddle welds in them. A lot of records have been lost or maybe they didn’t even record they were putting puddle welds in. I was recently made aware of a pipeline that they found more than a hundred puddle welds in. They excavated it. They found all these puddle welds. They said, “We’d like to find more of them. We’d like to map these and know where they all are.”

When they bring that information to us, we’re able to find that with MFL technology. We can find these puddle welds and identify them.

Russel:  That’s awesome.

Listen, Tod. I really appreciate you coming on. I think we barely scratched the surface on this topic. There’s probably a lot more here to talk about. We’ll see about having you come back. We’ll see if we can come up with some more big words for Russel to try and pronounce.

Tod:  Yeah. I like big words myself. I’m sure that’s a possibility.

Russel:  All right. Thanks so much, Tod.

Tod:  Thanks, Russel.

Russel:  I hope you enjoyed this week’s episode of The Pipeliners Podcast and our conversation with Tod Barker. Just a reminder, before you go you should register to win our customized Pipeliners Podcast YETI tumbler. Simply visit pipelinepodcastnetwork.com/win to enter yourself in the drawing.

If you would like to support the podcast, please leave us a review on iTunes/Apple Podcast, Google Play, Stitcher, or whatever smart device application you happen to use. You can find instructions at pipelinepodcastnetwork.com.

[background music]

Russel:  If you have ideas, questions, or topics you’d be interested in, please let me know on the Contact Us page at pipelinepodcastnetwork.com or reach out to me on LinkedIn.

Thanks for listening. I’ll talk to you next week.

Transcription by CastingWords