Welcome to the Pipe Liners Podcast, episode 189, sponsored by Rosen, the global leader in cutting-edge solutions across all areas of the integrity process chain, providing operators the data they need to make the best integrity management decisions. To find out more about Rosen, go to rosen-group.com. The Pipe Liners Podcast, where professionals, bubblegeeks, and industry insiders share their knowledge and experience about technology, projects, and pipeline operations. And now your host, Russell Tree.
Thanks for listening to the Pipe Liners Podcast. I appreciate that you're taking the time and to show the appreciation we give away a customized Yeti Tumblr to one listener every episode. This week, our winner is John Lester with Marathon Pipeline. Congrats, John, your Yeti is on its way.
To learn how you can win this signature prize, stick around until the end of the episode. This week, Karen Meyer of Atmos International joins us to talk about the use of API 1130 to create an Auditable League Detection System. Karen, welcome to your first episode of the Pipe Liners Podcast. Thank you so much.
I'm happy to be here. So if you would, would you do a little introduction for the listeners and tell a little bit about your background and how you got into League Detection? So it's actually kind of interesting because when I started my career in my education, I was actually an accountant for public education for school districts in California and Colorado and kind of fell into the oil and gas business. It's just a crazy phenomenon.
And I actually started with BP Pipe Liners and Tulsa. And I became, I started as a Skated Engineer and became the Skated Team Lead. And right then is obviously when we started working with League Detection and I worked with Atmos League Detection at that time and then moved to Mark West in Denver and became the Senior Skated Manager over seven regulated control rooms about the time that CRM was rolling out. And obviously we started getting into, it's a mid-stream company and we started getting into liquids and had to add League Detection onto that.
So I've really been digging into it since the beginning of my oil and gas career. And you're with Atmos now, right? I am. I'm currently with Atmos.
Actually on the 15th, it'll be a year. Well, so for the benefit of the listeners, Karen and I have been knowing each other for some time, probably over 10 years now. And we did some work on one of those control rooms she had at Mark West. And if you ever meet her at a conference, you need to ask her about a wedding at a bowling alley in West Virginia.
Oh my gosh. How's that? That was, I sit on top of the cake for the day we had it. We had a really rough day that day working with OPC and you're right, I've known you over 10 years now.
And that was a really rough day. And so we decided to burn some steam off and go bowling. And there literally was that wedding proposal and they had Journey Plane in the background. I mean, it was a night to remember.
Well, that's what I think of every time I see your talk. I always go back to my book. I tell you, that has nothing to do with asking you to come on and talk about ever. I asked you to come on and talk about API 1130 and specifically, what does an operator need to do so that their leak detection program would pass an audit.
So maybe the best way to start is I'll just ask, what is API 1130? So API 1130 came into play for liquid pipelines specifically. And it's actually if you have a CPM, a computerized program for your leak detection. This doesn't, that's why it's including liquids because there's not necessarily that requirement or the same type of technology for all of the gas.
You can do flyovers and a few other things. But for liquids, they want a form of leak detection. And if you have a CPM form, then you must follow API 1130. So what is in API 1130 that an operator needs to follow?
There's multiple sections in API 1130. And so they want to know obviously, I mean, they talk about the robustness and the reliability and the sensitivity, you know, with that type of information. But to break it down, I'll tell you more in detail what they're looking for. So what they're looking for is they want to make sure that you've done an analysis or a risk analysis on your pipelines to determine which is the best process or which is the best leak detection application that you can put on that pipeline.
And once you do that, they want to ensure that you have thought, basically, it's the idea that you've thought everything through, that you've actually done everything as well as you can test it and it's following all the performance indicators that you want and all the different pieces. So there's multiple different sections to it. So once you select one and you put in a CPM, they want you to detail and think about how are we going to test this? When are we going to test this?
What are we going to do to test this? So there's a big testing program in there which incorporates potentially maintenance documents and different other things that can affect your leak detection system. Because as you remember, approximately, I want to say, was in September two years ago, they added that group training regulation, which is incorporated with this because a lot of times if someone went to the field to a meter or valve and to calibrations or did different things, they didn't understand the impact that it had potentially either in the control room on the SCADA and or the robustness of the leak detection system. So there's a lot of aspects to it.
There's different pieces in there regarding what kind of training are you going to do? How often are you going to train? So it's definitely incorporated with your OQ and different pieces of that. Then, of course, you have your SCADA side, which my background is I worked more SCADA and kind of fell into leak detection, but I actually started on the SCADA side directly.
So you've got to think about the different things that incorporate with control management, the API 1168, or even SCADA screens, API 1165, or even alarm management, API 1167. So it really encompasses a bigger piece than just we have leak detection. This is how often we test it and these are our results that we've tested. Yeah.
And Karen, you just said a mouthful there. I know. I realize that. Any operator that's sitting listening to this and says, well, I have leak detection is probably quaking a little bit after listening to all that.
And I don't disagree with anything you said. I think I ever bit it. That's right on point. But it raises the question as an operator, well, what do I need to do to have a leak detection approach that's going to stand up to the audit?
So I think what I would break it down to, and I'll ask for your feedback, is I need to have some kind of deliberate process that I use to select my approach. I need, based on that process I use to select and the risk I think I have, I need a way to test and assure performance. Correct. And then I need a pretty rigorous on-im program to make sure that it continues to operate.
And then what do I do when I lose leak detection? Exactly. So like I said, I know I said I'm out full and I threw out a lot of APIs and a lot of different programs and stuff that are out there that we have to comply with for pipelines. But one thing that I can tell you from experience is going through an audit and we got away with saying we have this form of leak detection and here's the documentation of how that leak detection works and here's our testing records.
At that time, it worked. Now this was a few years ago. So things are changing now and the audits are changing, especially being virtual audits or documentation audits and not face-to-face audits. So you can't just talk your way through it so much anymore.
You just have to send them all the information and now they're realizing and I'm hearing from a lot of different operators that PHMSA or whomever that's audited on if it's a state audit or a federal audit are coming back saying we want more documentation, this is not sufficient. And so we're finding that there's a lot more focus on leak detection than there has been in the last, well, since I've been involved in the last 16 years. I think it warrants asking the question, why is that true? Do you have an opinion about that?
I do have an opinion. You always have an opinion. I always have an opinion. As do I.
That's why we do long so well. So I would say in the last maybe four years, we've had, especially since the pandemic and I'm actually going to use that as my market date. So let's go back a year and a half. Even in a year and a half ago on leak detection specifically, we have had more leaks and it breaks my heart, but it's happened.
We've had more leaks in the last year and a half than we've had in the last 10 years. And so I think that they're looking at it different. It's kind of like the same idea as some of the other regulations. And like even, I was on a call to supporting about API 1164, cyber security is now a big thing.
I think that put in, put a push on leak detection. Yeah, I, I believe that to be true. I think there's another factor at play. And the other thing that's happening is that there's a lot more attention on pipelines.
When I started in the pipeline business 20 years ago, nobody ever talked about pipelines. Right. So that's all the time now. Right.
So anything that happens gets a lot of air play. So that creates two things. It creates operating risk for operators, but it also creates regulatory compliance risk for many others other than just other stakeholders and just the operators. So I think that's kind of pushing us.
And then with the new administration and their focus on minimizing releases of all types. Yes. That's also putting some focus on leak detection. That's really that focus is really outside of liquid pipeline leak detection through a system.
But it does what, and I would also say too, kind of just, as I'm thinking through and unpacking this, is that we've kind of been under this, this idea of, well, if I have a piece of leak detection software and it's running, well, then I'm covered. True. But that's not actually true. That's a true statement.
And I didn't want to get too political. So that's why I didn't go down that. I don't know. I know.
I know. I know. But it is a reality, right? Agreed.
And I totally 100% agree with you. And I think that's a big push on it too, especially with all the new emissions and things coming out. But that's more on the gas side and the new regulations coming down for that. So leak detection industry as a whole is going to have to adjust and modify to be able to accommodate for that.
But I do agree with you that you made a statement that we are definitely in the limelight. You know, like you said, 20 years ago, you didn't hear about pipelines. Now you hear about them consistently for multiple different reasons. Well, I think where I would go with this is that you were talking about the pandemic.
The other thing that's the reality is the, you know, fems has continued to pretty much sustain its targeted auto pace. They're a little behind, but not a lot. That's true. And they've been doing these audits remotely, which means there's been a lot more focus on giving me the document, let me analyze the document versus I'm going to comment, we're going to sit down for a week and have conversation.
Exactly. So that's where the document needs to be. You need to ensure that your policy's procedures document program, whatever it may be, is solid, very solid because you're right, because before you could sit in a room and you could say, well, even if it's maybe not hyperlinked or however you have your documentation designed, you can pull up a document on the screen and say, yeah, but we have this covers that requirement. And now you have to send them everything.
So you can't just say, oh, we do have that. You know, they want to, they're going to deem you and then you go back. They're going to thank you. That's the word.
They are. Okay. That's what you say you're doing. Show me the policies and procedures that demonstrate how you do it.
Show me the records that demonstrate you're doing it. Show me the analysis that demonstrates doing all those things creates the safety benefit we're looking for. Exactly. Exactly.
And I do agree with you. I remember what the comment was that you made was about you put in a CPM system and you think you're done. That was the comment I wanted to piggyback off of a few minutes ago. And that's not a true statement.
Some people, some customers do that. Some operators do that. You know, I've been in the industry 16 years now and I've seen it and they're just like, well, we have went in there. But like you stated, that's not necessarily the case.
And that's what API 1130 is trying to get people to understand and document and recognize that there's certain testing requirements. There's certain pieces that go above and beyond your OEM or even your, that not necessarily above and beyond control room management, but they coincide together and how things are related to each other and that you really need to keep an eye on it. I mean, definitely, especially your documentation and your program, your leak detection program. You know, what do you do if you calibrate a meter?
Do you go back and do retesting? Do you just say, I no longer have a false alarm, you know, whatever, you know, different things for the robustness and the accuracy of your leak detection system. They want you to think about that and continually be improving it to make it obviously just that, you know, better and catching a leak quicker and sooner. Right.
I want to catch a smaller leak. I want to catch it faster and I want to know where to go. Exactly. In location, yes, that's definitely a big one too.
And stuff. So, yeah. So let's unpack that a little bit because I think if people understood a little bit more about 1130 in the details and whenever I'm looking at a, you know, a leak detection program around CPM, I'm looking to try and answer how does the operator answer the four key questions that API 1130 says you need to look at, sensitivity, reliability, accuracy and robustness. So what is sensitivity?
Sensitivity is how I see that is how like the span of the, I'm doing visual things here. So you have to ignore me. I'm looking away. That's so effective on the podcast.
So do that all the time. Making end motions and faces all kinds of stuff. All sorts of stuff. So that's how far down can you get your percentage span to be able to determine a leak.
So as you know, probably there's depending on the pipeline, depending on the terrain, depending on the size, depending on the metering, all that information, you know, can depend on the reliability of your data. So basically for a CPM, you really want repeatable to be able to repeat stuff, to keep the reliability and not get the false alarms and that type of thing. And then how much can you tune it down to? Can you, because of the communications, and there are also, I forgot communications, that's a big one.
How far can you tune it down to? Can I say I can get a 5% leak without tuning it too much to get too many false alarms or can I go down to a 1% leak or a 0.05, you know, whatever it may be. And so that would be the accuracy is looking at your data. And one thing that you'll find or I've seen in the past is implementing a CPM system.
We found a lot of issues in the field that needed to be corrected and or equipment updated or communications modified or something to be able to get this data or this, you know, accuracy data that's repeatable. Yeah. I think hearing is such an important thing for people to understand is that the sensitivity you can drive out of a leak detection system is constrained primarily by the quality of your metering and the rapidity how quickly you can pull and get the data in. Exactly.
Those two things are the primary drivers. And if I've got metering that's in and out and, you know, meters are going offline and they've got bad meter factors or whatever or I'm having problems with density consistency or any of that kind of stuff is going to impact sensitivity. Oh, definitely. Software can only do as good a job as the data is fed.
Exactly. And I think that's sometimes sometimes because you've got your field operations and leak detections more typically out of the control room or potentially an engineering group or depending on how your organization set up. And so those two sometimes don't always understand what the other one does or the importance. I think that's where that team training regulation come in, which is a totally different podcast, but you know that they need to work together and these actually go hand in hand and we're not pointing fingers or, you know, yeah, it's a team.
It's a team effort. We need to understand that when I'm out calibrating a meter or servicing a meter, what impact did that have to the operations and the safety of the overall pipeline. So let's transition. What's reliability?
Reliability is, you know, basically it's kind of falls a little bit under the same thing. It's just, you know, as far as I'm how I see it, I'm going to rephrase that is, you know, kind of like the reliability of the data, false alarms, that type of thing. Can you actually trust the CPM system that you chose? You know, there's a big thing about eliminating false alarms and finding a CPM system that doesn't give a lot of false alarms.
You may get a few for whatever reason, but you want to be able to tune it down as much as you can. And it's obviously just that reliable and it gets the information that it needs and it's keeping it. Yeah, exactly. Reliability is just kind of simply stated as false alarms, but the thing that I think a lot of operators don't understand is there's two kinds of false alarms.
There's false alarms related to instrumentation problems or maybe problems with the software, the servers, you know, something gets corrupted or it's not running quick enough or whatever. That's kind of one thing, but there's also false alarms that are related to normal operating conditions that occur rarely like slackline or whatever else that can generate false alarms. So really reliability is what are you doing to manage and deal with false alarms and what are you doing to make sure that the best your ability every alarm is valid? Exactly.
So that actually, like I said, there's different pieces of this and I brought up the alarm management, you know, control, and this kind of goes hand in hand with that. You also, from experience, I have found where we had a bunch of, as you call, operational false alarms, let's say, and that was because other field operations was running the line and doing all these different things and they didn't realize the impact it had and then we actually got together and decided or worked together to come up with a better work process that actually stopped surges or stopped, you know, slack in the line or different things and smooth out operations. You tighten up your operations. I know operations.
Yeah. Exactly. Yeah. I think too that I've had some really interesting conversations with some of the operators I work with about false alarm versus valid alarm, meaning yes, the assistant should have generated an alarm, but it's not a leak.
Yes. Versus a leak alarm and like, how do they manage the distinctions between those things and how does that relate to procedures in the control room and in the field? You know, that's interesting that you say that because a lot of these like that were not operational issue. That was a valid alarm.
It was not a false alarm, but when you talk to a lot of people in the oil and gas, they only have those two acronyms false alarm, leak alarm, but a valid alarm that's not a leak should really be in conversation more because I think that made it builds confidence. Well, and that's where that's where your alarm management program and your alarm response should be focused is on what do I do about these valid alarms that are not that are not leaks. Exactly. That might not be leaks.
How do I get some really strong fencing around that definition? Exactly. And that is part of the section of the API 1130 is they talk about the alarms and what do you do? Do you have meetings to kind of circle the loop basically to if you have a valid alarm that's not a leak, what do you do to mitigate that?
And you know, how are you handling those? And that's actually part of the 1130 process also in the program. Yeah. And for people that are interested, the Marshall incident, Marshall Michigan incident in the Calumency River, if you look at that one, then very much tied to what we're talking about now.
They had they had valid alarms that were actually leak alarms, but they didn't interpret them as leak alarms. They interpret them as operational alarms. And you know, that you can there's some real if you read the timelines and you read the analysis that the NTSB did as reported around Marshall. It really kind of puts a box around this problem or this challenge.
Exactly. Exactly. So let's shift again. So accuracy as it relates to a valid 30.
Well, you're only as good as your data. Right? I mean, in the big picture, you're only as good as your data. So accuracy would be, you know, just that that, you know, from personal experience, you know, we had a couple meters that were calibrated off just a little bit, but it was on a long pipeline.
So it affected the readings from the meter before and the meter after. And so, I mean, obviously accuracy, you need to, we had to do a work order and get that corrected, but it took us a while to figure out what was happening because we kept asking for the program and this and the other. It's another one of those working together, you know, the team training idea. But you know, you're only as good as your data and.
Yeah, it's interesting. I kind of focus on accuracy. This is kind of a different way to say what you're saying probably, but I kind of focus on accuracy as when you get in the warm. Can it tell you a leak rate and can it tell you a leak location and can it tell you that with some level of accuracy?
Yes. And that would depend on the CPM system you have. And so some of them are offer location. Some don't some offer a area.
I mean, it really depends on what CPM system you have on that. A lot of them do sizes, but location seems to be one of the bigger challenges or a lot of the CPM systems out there. But if you look into your research and that's part of what actually it's more API 11 or RP 1175, but to go through the analysis to determine why you pick the system that you picked that one, but that one falls a little bit more under API 11 RP 1175. Okay.
Interesting. You know, it's interesting to me because I'm there are a number of systems out there that do not really give you any indication of location of this between these two metering points. Yep. Right.
And the other thing you mentioned is how I'm sorry, I'm, you know, I'm closing my eyes and my wheels spinning in my head here. I'm trying to figure out, I'm trying to frame this up. I'm trying to say you made the point that about metering, and I think you're absolutely right. Part of accuracy is being very clear about what is the performance standard for my metering and for my instrumentation?
Correct. And how do I maintain that performance standard for metering and instrumentation? And then how does that relate to my CPM performance? So and like I said, as part of API 1130, if you read through it, there is a section obviously on this and that's where it can incorporate your maintenance records or maybe something from INE or your measurement.
I mean, however, your organization is designed and set up. So it all comes back to it's more than just saying I have a leak detection system. What are you doing to be able to make sure that your leak detection system is doing what it's supposed to be doing? Yeah, exactly.
Exactly. And then lastly, what about robustness? What is robustness? So, um, oh gosh, I just linked on that one.
Well, I'll tell you what my, how I read 1130 robustness is how well does your leak detection system work when you have communications problems, server problems, metering problems, et cetera? How well is it able to operate through a range of operating conditions and through a range of abnormal operating conditions? And then again, how do you get a fence around that? Yeah.
So there's different processes in there and procedures and different things that they want to see to handle those kind of conditions. You know, the AOC program or the AOC, you know, matrix and different things along those lines and training. And when do we call the field out? When do we close or shut up pipeline in?
When do we do certain things when different abnormal operating conditions? I guess it's the best way for me to put that occur. And are you addressing the ones that affect your leak detection equipment? Because in the control room, you may have an AOC matrix that is for all metering or whatever you have on your pipeline.
But then only maybe a third of that is actually somehow related to your leak detection system. I'm just pulling numbers out of the air. So I'm not, you know, trying to say only. Yeah.
I think you're right on it, Karen. It's understanding your AOCs and how those AOCs and the specific AOCs that affect leak detection and understanding what your policy is about addressing those. Exactly. And so, so when we talk about the API 1130 program and even with the alarm management, I'll use that one as my example.
So you have your safety related points that you have identified along your pipeline. All of those may not be part of your leak detection system. They could be, it depends on how you're set up. But they want to see for the API 1130, what have you earmarked the ones that do affect the leak detection system?
And are you doing things to ensure that those points are handled appropriately for the leak detection application that you have? Right. Right. Yeah.
You know, it's interesting at 1130 pretty much just it, the first time you read it, right? You think it's about the CPM software. And then about the fourth or fifth time you read it, you start realizing it's not really about the CPM software as much as you thought. Exactly.
Yeah. It's more about all the other things that have to be in place for the CPM software to do his job. And that's exactly it. That's why, like I stated earlier, that, you know, the audits are getting harder because before you could talk your way through it, you know, and just say we have a CPM system, here's the documentation on how they do their algorithms, statistics, whatever it is that application does.
And then, you know, here's our test records, here's, you know, our training records. And it's more than that. It's so much more than that. So a lot of people are discovering, especially now that everything is done more virtual and you just send them all the documentation thinking that you have all the pieces and then you find out you may have them, but they're not all together in the same place or at your mark in the fashion that they want to see them.
I mean, everybody, we do it all. Nine times out of 10, we're already doing it, but we just may not have it organized and put together appropriately. Yeah. I think that's absolutely right.
I do think though that this is one of those areas, particularly for the smaller operators where they might have risk, they don't know they have. True. Talk about regulatory risk more than operating risk. Agreed.
I definitely agree with you because, you know, we, we all have the same goal as far as operations go, safety, safety, safety, safety. But you're right, regulatory, you just may not have it put together or documented or even thought about some of the stuff to realize the impact of everything of it, bringing it until you bring it all together for like a program for API 1130. Yeah. And we are in the process.
Sorry, I mean, catch up, but we are in the process right now. The API is get reviewed every five years and API 1130 is actually on the marker to it has been rewritten and it's actually going to vote in Congress sometime this year. I don't know when. And so the regulation is going to be changed and or added to.
So, you know, which is going to cause another potential gap for clients or customers or, you know, companies out there because it's going to be a little bit different and, you know, but I don't have a date of when any of that's going to happen. Yeah, I think the API has been shifting its priorities around both the change priorities from the administration and related to the colonial incidents kind of shifting their priorities about what standards are getting primary focus. Agreed. Agreed.
So, and for just five more reasons, right? I totally agree. So, what would you, what would you say the key takeaways are for an operator? So, so the key, I would say, you know, get familiar with API 1130, you know, there's companies out there that can help you pull it together and do with this different kind of, you know, making sure you have everything, but it is bigger than just saying I have a CPM system.
There are so many different aspects to it. And understanding that, you know, that there's other components that are involved and potentially could impact your CPM system. It's just more than just saying I have a CPMs early detection. Yeah, you know, I think again, very well said, the other thing I would just illustrate is that an operator could, you know, with a simple, you know, point to point pipeline could do a simple mass balance in their SCADA system.
And as long as they did all the API 1130 around it, they could be compliant. Yeah, they, so the mass balance is, you know, if they did it like with a spreadsheet or that type of thing, it would be, well, it depends on how they do it. So, if they do it computerized, it would be 1130. If they just did it on paper visually or a different type of way, it would fall more under API RP 1175.
So it just kind of depends. Well, I'm, yeah, I agree. I'm referring to somebody that had a SCADA system could build something of their own that's pretty simple. They could.
And as long as they had 1130 wrapped around it, that could be compliant. That is a true statement. Well, not that's best practice is a different question, right? Okay, good point.
That's exactly a true statement. Yeah. Yeah. Yeah.
Yeah. You know, I mean, and you and I both know enough to know what the operational reality of that is and how how actually effective it is. I'm just making a point that the key thing about 1130 is not really what software you pick. It's all the things that you're doing to operate and maintain that software.
That is a true statement. Like I said, it's so much bigger. Yeah. And then on the flip side of that, even if you've picked a well-known, you know, highly used software, if you're not doing 1130, it's kind of the flip side.
Even if I have good software, I still have to do the 1130. That is a true statement also. So yes, agreed. Agreed.
Yeah. That's kind of be would be the key takeaway I would want people to take is it really doesn't matter what the software you have is you still got to do all this other stuff. Exactly. And so you can have a lot of different kind of lead detection software on your running in your company.
It doesn't matter. That's not uncommon. No, not at all. Some of the bigger operators, they're running multiple tools as a way to mitigate risk.
Yeah. And they also, I mean, the same thing with data. There's multiple tools for everything and they can run multiple ones, but your API 1130 should encompass the program for your CPM system, not the technology. Well, listen, this has been great.
And I will tell the listeners that we plan on having you back. We've got some other things we want to talk about specific to. You've mentioned several times 1175. I think it'd be great to do a conversational 1175.
And I think another thing would be good to do is talk about all of the regulations incorporated by reference that are impacted by lead detection. That would be wonderful. I would love to do that. Well, we'll definitely have you back.
Listen, thanks for coming home. It's been fun. Thank you so much. I appreciate it.
Hope you enjoyed this week's episode of the Pipe Liners Podcast and our conversation with Karen. Just a reminder before you go, you should register to win for our customized Pipe Liners podcast Yeti Tumblr. Simply visit pipelinerspodcast.com slash win and any yourself in the drawing. If you'd like to support the podcast, the best way to do that is to leave us a review and you could do that on Apple Podcasts Google Play or whatever app you use on your smartphone to listen.
If you have ideas, questions or topics you'd be interested in hearing us talk about, please let me know either on the contact us page at pipelinerspodcast.com or reach out to me directly on LinkedIn. Thanks for listening. I'll talk to you next week. Thanks for listening.
I'll talk to you next week. Thanks for listening. I'll talk to you next week. Thanks for listening.
I'll talk to you next week. Thanks for listening.