Pipeliners Podcast

  • Your Host
    Russel Treat
  • Dan Sensel
    Our Guest
    Dan Sensel
  • Kyle Miller
    Our Guest
    Kyle Miller
  • Jason Dalton
    Our Guest
    Jason Dalton


In the second of a series of episodes, Pipeliners Podcast host Russel Treat welcomes Dan Sensel, Jason Dalton, and Kyle Miller from Marathon Pipe Line for an important discussion of how to manage slack line in a liquid pipeline.

In this episode, you will learn about what causes a slack line, the risks associated with a slack line, what should be included in your operations plan to manage a slack line, the impact of a slack line on leak detection, and various calculations to help identify the appropriate steps to take.

Show Notes, Links, and Insider Terms

  • Marathon Pipe Line (MPL) is a subsidiary of Marathon Petroleum Corporation that owns, operates, and develops midstream energy infrastructure assets. MPL operates pipelines, storage tanks, and barge dock facilities.
    • Dan Sensel is the Leak Detection Project Lead at Marathon. Find and connect with Dan on LinkedIn.
    • John Dalton is the leak detection and hydraulics supervisor at Marathon. Find and connect with John on LinkedIn.
    • Kyle Miller is the pressure control process lead at Marathon. Find and connect with Kyle on LinkedIn.
  • Slack line is a condition when both liquid and vapor exist in a liquid pipeline at the same time. A similar term is column separation.
    • Open Flow Channel refers to when the liquid and vapor are flowing at the same time.
    • High Point refers to the highest point on the line between the initial receipt location and the delivery point of the product.
    • Vapor Pressure is the boiling point of a liquid or the given temperature when a fluid will begin to boil or transition from a liquid to a vapor. In a liquid pipeline, vapor pressure is dependent on the product on the line, the temperature, and the pressure.
      • A Vapor Pocket results when the local pressure at the high point is lower than the vapor pressure of the fluid in the line.
    • Back Pressure is the pressure that a pipeline delivers at its termination point, which is typically a storage tank.
      • PSI is the measurement of pressure to determine the desired flow rate.
        • Example: If the pressure at the high point is lower than the vapor pressure, then you should increase back pressure.
  • Leak Detection is the process of monitoring, diagnosing, and addressing a leak in a pipeline to mitigate risks.
    • The Real-Time Transient Model (RTTM) for leak detection simulates the behavior of a pipeline using computational algorithms. The model, which is driven by the field instrumentation, monitors discrepancy between the measured and calculated values potential caused by a leak. RTTM uses flow, pressure, temperature, and density among many other variables.
  • MOP (Maximum Operating Pressure) is the maximum pressure that a pipeline can withstand based on its design, function, and strength.
  • The Joukowsky Spike (water hammer) is an equation that captures the maximum pressure change in a pipeline due to sudden valve closure.
  • API 521 (Pressure-relieving and Depressuring Systems) is an industry standard that provides guidance on how to examine the principal causes of overpressure, the individual relieving rates, and appropriate disposal systems.
    • Section 4 focuses on the causes of overpressure and their relieving rates.
  • Normalization Deviance is a method to help pipeline controllers understand how to identify a leak event by looking for the signature of an event.

Full Episode Transcript

Russel Treat:  Welcome to the Pipeliners Podcast, episode 57. This episode of the Pipeliners Podcast is sponsored by EnerSys Corporation, providers of POEMS, the Pipeline Operations Excellence Management System, industry leading software for the pipeline control center. To find out more about EnerSys control room solutions, go to enersyscorp.com/podcast.

[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. We appreciate you taking the time and to show that appreciation, we’re giving away a customized YETI tumbler to one listener each episode. This week our winner is John Castille with Northwest Natural. Congratulations, John, your YETI is on its way. To learn how you can win this signature prize pack, stick around to the end of the episode.

This week we have the fab three. That’s what I’m choosing to call them at this point. From Marathon Pipe Line: Jason Dalton, Dan Sensel, and Kyle Miller are returning to talk about managing slack line in the liquid pipeline. Jason, Kyle, Dan, welcome back to the Pipeliners Podcast!

Dan:  It’s good to be here.

Kyle:  Pleasure to be back.

Russel:  We’re going to talk about slack line today. I know there’s gas guys that listen to this. I think most people in the liquid world would know what slack line is, but maybe you could give us a definition for those in the business that might not have heard the term slack line. Just tell us what is slack line and why is it so important for us to talk about it?

Kyle:  Sure. A slack line is simply a line in which liquid and vapor phases both exist at the same time. That’s typically at a local high point. If it’s flowing we’d call it two-face flow or open channel flow. Of course, it’s technically not an open channel because you have vapor, but the vapor space does rob you of capacity.

Russel:  It also creates other types of operating problems, right?

Kyle:  Right. Yeah, we’ve got a couple, whether if it’s running, there are some leak detection implications with that. If it’s down, if we’re shut down and we don’t have a tight line, tight being the opposite of slack, we’ve got leak detection implications there, too. We can get into that.

Russel:  Cool. Maybe we’ll just start and you guys can talk a little bit about what are some of the things that could cause a line to go slack.

Jason:  Sure, Russel. There’s a couple main things that cause a line to go slack while you’re running. Before we start, a quick terminology thing. For Marathon Pipe Line, slack line is pretty much any condition where there’s vapor present in the line, whether the line is flowing or the line is shut down and not moving product.

Some operators choose to use the term column separation. We don’t use that term. We just use slack line. When a line is running, the things that typically cause a slack line is if there’s insufficient back pressure.

Sometimes like when we’re coming down into a river valley, you clip the high point right before you come into a terminal that’s going to load barges. In that kind of case, you need more than just the delivery back pressure to make sure that you’re clearing the high points and that your local pressure at the high point is above the vapor pressure.

On the flip side, if you’re discharging at too low of a pressure and you’ve got a big hill on the middle of your line, you could also cause the local pressure at that high point to go to vapor.

The last one is fluid properties. We do a batched operation and we’ll carry many different kinds of fluids on a pipeline. The same lines carrying diesel fuel with a very low vapor pressure can also be carrying propane. Pressure controls have to be designed to use both of those fluids with very different vapor pressures.

One day when you’re carrying a line full of all diesel fuel, everything is fine. The next day you switch over and you’ve got propane in the line and you’re cutting the high point, which means your local pressure at the high point is less than the vapor pressure of the fluid. It boils off and creates a vapor pocket.

Russel:  Right. First definition, high point. That sounds straightforward enough. What’s a high point?

Jason:  The high point, there are actually a couple different ones. There’s the local high point. If you’ve got a pump station feeding another pump station, it could be the local high point in that segment of the line. The other one is the highest point on the line between the initial receipt location and delivery point.

Russel:  It’s just the hill top basically in simple terms. Yeah. Then, what is vapor pressure?

Jason:  You can think of it as the boiling point of a liquid. We want to talk about water. At normal atmospheric temperature, water wants to boil at 212 degrees Fahrenheit. As you increase that pressure, your boiling point continues to go up. It works in reverse. If your pressure drops below atmospheric pressure, it will start to boil off at lower and lower temperatures.

The vapor pressure, we just save vapor pressure in the industry and we don’t tie it to a specific temperature on the liquid side. We just say, “Oh, the vapor pressure of propane is typically 147 ESIA.” It’s actually the vapor pressure at the temperature that the line is flowing at. You need to know both things to manage slack lines.

If you’ve got a line that’s warming as it goes up flowing along the terrain, that vapor pressure is changing. The vapor pressure, to circle back to the original question, is just the pressure at a given temperature at which a fluid will begin to boil or transition from a liquid to a vapor.

Russel:  Right. That sounds relatively straightforward, but I know that operationally on a liquid line, particularly if you’re using multiple products in that line, that’s a big deal. Of course, the reverse is true, if the product gets too cold, it will congeal, become a solid.

Jason:  Yeah, I think they have a lot of trouble with that in the gas world with the gas hydrates. We don’t see a lot of that in the liquids world.

Russel:  It depends. If you’re using like bunker fuel or some of the fuel oils used for power generation, those have to be heated because they solidify at like 90 degrees Fahrenheit, 80 degrees Fahrenheit. There are some exceptions to that.

I think the thing that’s interesting in this whole conversation and why this becomes complex is that vapor pressure is dependent on the product that’s in the line and on temperature and pressure. It’s a combination of those three things that determines what your vapor pressure is.

Jason:  Absolutely.

Russel:  Then you also used the term back pressure, so we ought to unpack that a little bit. What’s a back pressure?

Jason:  Back pressure is the pressure that the line delivers at its termination point. Typically, when we’re doing a first design, we’ll start at a tank. It’s delivering from one atmospheric storage tank. It goes potentially hundreds of miles through multiple pumps and it ends up delivering to another atmospheric storage tank.

At the delivery end we’ll start with atmospheric pressure and work though the elevation change and the tank. You add in your static head of whatever that fluid is and the height of the tank.

Then, you look at the rate the system is flowing and calculate all your friction losses from where the pipeline ties into your delivery facility and add all those together. You’ve got 5 PSI through a meter run that you’re going to lose. You’ve got a strainer that could potentially lose 10 PSI. Then you’ve got a tank that’s got another 10 PSI.

You sum those together and you end up with 25 PSI back pressure at the flow rate you’re running. If your back pressure at that point and you put a hydraulic gradient in it, which is your friction drop per mile, and you look back the line to where the gradient goes over the high point.

If the pressure there is lower than vapor pressure, then you have to increase back pressure in your delivery facility. You can look at, “Do I want to do something like decrease the size of my tank farm piping so that I’ve got more back pressure in the facility?”

Or, the real common thing and the things that we do a lot of is you put a back pressure valve in which creates variable friction loss, so that you can run the line always in its optimal way. You don’t want to always be burning off 100 PSI through your facility. Sometimes you need 25 and sometimes you need 100.

Dan:  We have some lines that keep about 40 to 50 pounds of back pressure. We’ve had lines that keep 400 pounds of back pressure in areas where the elevation profile is treacherous.

Russel:  Right. This is one of those things. I say this all the time, but everything is easy until you know enough about it. This is one of those conversations where on the surface of it it seems pretty easy, but it gets real complex real quick as you start throwing in all the variables.

A simple illustration of this for listeners that might not have ever been exposed to this stuff that we’re talking about — this hydrology. If you think about a straw and I’m trying to move liquid through a straw. If I’ve got a mouthful of water and I’m blowing through a straw, if I put my finger over the end of the straw, I can blow real hard and I’m not going to get any fluid to come out.

If I remove it, it’s just going to come right out. If I take it and I just leave a little bit, I’m going to get a stream of water, and that water, if it’s a high enough pressure stream, is going to go to vapor pretty quickly or small droplets pretty quickly.

All of this conversation about slack line is managing that reality in hundreds of miles of pipe. It sounds simple if you say it quickly so don’t slow down.


Dan:  Right, and we’ll say that nothing is ever empty. We talk about an empty pipe, well, that pipe may be full of air. We say a line goes slack. Okay, it’s empty at the top. Really, that’s a vapor pocket. All of those things have to be managed. It’s never empty, and air and vapor have to be managed very differently. It just makes our job fun.

Russel:  Yeah, well, that’s right. Fun, that’s the perfect word. Let’s talk a little bit about normal startup and shutdown because those are the three key operating conditions you get concerned about around slack line.

Let’s give the listeners a definition of normal startup and shutdown and why slack line is a different consideration in those cases.

Dan:  Honestly, it might be easier to start with shutdown because we don’t like to start up a line having been slack for any length of time due to leak detection purposes, and we can talk about that in a bit.

When we shut our lines down, we’ve actually taken great efforts to do studies on all of our pipelines to model their particular shutdown characteristics for those elevation profiles, the pumps, as Jason mentioned. We have back pressure valves on several lines.

We will tune transient models to accurate line conditions. We will develop shutdown procedures to try to trap as much pressure as possible on the line. You don’t want to trap so much that you pop a relief valve or you exceed 110 percent of MOP. We also don’t want to trap nothing and shut the line down in the slack condition.

On all of our systems, that’s a big thing that our group does, is we will model those shutdowns. We work with the operations center on procedures to shut it down and trap as much pressure as possible. That way, when we start up, we know that we’ve had good leak detection the whole time.

On the startup of a line that has gone slack, one thing that’s difficult about that is when you have that vapor cavity collapse at one or several high points along your line, you’re going to have a pressure spike there that’s not going to be seen at the transmitters on either end.

Industry literature indicates that that would be typically about half of a typical Joukowsky spike, which is the instantaneous surge spike you get on a valve closure. There’s a bit of a peril there. We’re currently working through the phases to manage that.

Russel:  Yeah. We probably ought to talk a little bit about Joukowsky, and what’s otherwise called water hammer, and why that’s such an important consideration in this. If we could, maybe we’ll just depart on that a little bit.

When you think about normal operation or what’s sometimes called static operation, that’s just the line is flowing and it’s flowing around its designed pressure — around its designed parameters. That would be kind of considered normal operation, another way to refer to it.

When I’m shutting down, I’ve got to close valves to hold that pressure, and the rate at which I close those valves is material. If I close them too fast, I’m going to get water hammer. I’m going to get that pressure spike, and I’m going to move it up and down the line. That could be very damaging.

It’s a non-trivial thing, right, to shut a line down and hold pressure but without causing pressure transients.

Dan:  Right. That’s where we have…all of our lines either have a mainline surge relief valve at the delivery location to manage this or we’ll put a flow control on the line to keep the shutdown at an intrinsically-safe pressure.

If we did have a valve closure, we would be able to withstand that spike and that pressure build without exceeding 110 percent of our MOP.

Russel:  Right. Do you guys have any situations where you would automatically close a valve on some kind of an emergency or upset or is that all done by…when I say automatically, I mean through the automation versus by a controller operating a valve?

Jason:  Russel, we don’t have that particular piece of equipment. We’re taking the stance that actually induces more risk on the system than having an operator or human in the loop.

Anytime you have something that’s automated, it’s really hard to model every potential scenario and make the programming applicable across the board. There’s stories in industry of a lightning strike that hit a valve actuator and made the actuator go closed all by itself. It adds a level of risk that we’re really not willing to adopt.

All of our valves are commanded closed by an operator in our operations center when they feel the need to close that valve following our normal procedures.

Russel:  Right. That’s why I asked the question. I think it’s important for listeners to understand that because that’s a question for someone who doesn’t understand this issue of managing pressure and pressure transients that they often don’t understand. You’ve got to be very deliberate about how you close these valves.

Jason:  Absolutely.

Russel:  They can cause problems and they don’t…Some of these valves can take several minutes just in normal actuation to close, and you do that on purpose to not have these water hammer issues. Anybody that’s got valves at the house and has closed them quick and has heard the lines shake in the old house, you don’t want that happening on a pipeline.

Jason:  No. There’s a significant amount of energy that’s in these lines, and the way that they’re being managed, they’re safe and the energy is handled in an appropriate manner but every time you start to try to force the systems to make fast actions, you start to erode some of the safety that’s built into these systems.

Russel:  Yeah, very well said. Let’s circle back and talk a little bit about what can cause a line to go slack during a shutdown.

Kyle:  The most common cause of slack for us is cooling during shutdowns. Our product, whether it’s crude or refined products, is typically at least several degrees warmer coming in than the ground, and it doesn’t take much, between 50 and 100 PSI. It’s not uncommon to lose for just one degree Fahrenheit, especially if you’re coming out of a refinery processing unit. Maybe it’s 100, 105, 110 degrees and the earth is at 60 or at 70. It’s not going to take you too long to drop 300 or 400 pounds and to go slack.

There is an equation in API 521, specifically Section 4, that we use that relates the pipe properties with the fluid properties and then pressure loss for temperature loss. When we shut down and we know we’re going to go slack or if the line’s going to be down for three or four days, our operations center analysts have a tool that we have put together in the knowledge management database.

It’s called the slack line tool, and we’ve got profiles of lines. We’ve got screenshots of what the leak detection system is going to look like. We’ve got notes on which segments go slack right away, which ones never go slack to try to give them a point of reference for what’s normal and what’s abnormal.

If it’s only temperature loss and not a leak, there’s a nice slow and steady decay that they’ve been trained to recognize. If you had a rupture while the line is down and static, it’s going to get a little bit steeper.

There’s many times that we’ll get called in from our group to log into the system maybe after-hours and look at a pressure trend, then look at the ground temperature probes near the line, and look at ambient weather conditions and try to figure out if it’s just due to cooling or if it might be something else going on.

Russel:  As you’re giving that description, I’ve got pictures going through my mind of what that looks like. I was quiet there for a second as I was thinking about it. That’s a huge amount of work to build that up in a way that the operators can use it effectively.

I think that’s because these situations are normal but not usual. Knowing what the line’s going to do given various environmental and fluid conditions other than just pump the fluid is not really what you’re doing as 80 percent of your job. 80 percent of your job is move the product.

Jason:  Correct. One of the things that I think the whole industry as a whole is trying to do is to ask the controllers or whoever is sitting in the operations center running the line to think through their head what do I expect this line to do when I make this change and compare what they’re thinking should happen with what they’re actually seeing.

What Kyle was talking about, one of the things that we’ve done, and I know there’s some other companies that are doing the same thing, is to save pressure trends of what a typical slack line event looks like following shutdown and provide historical pressure trends and historical leak detection system output for those events.

They can look back and say, “You know, I’m pumping product out of a tank that’s at 90 degrees and it’s winter, so I’m going to expect the line to lose pressure pretty quick,” and then they watch what’s happening in the line and it loses pressure a lot faster than they would expect. It goes slack a lot sooner.

They can pull up a historic trend and go through and take a look at what has the line done in the past and compare that with what it’s doing right now and give themselves a level of safety that, yes, I’ve seen it do this before. “This is typical for this line” or “No, this is not typical. This is something that I’ve never seen before.”

Russel:  Right, and that’s a really important point here is that this kind of work you’re talking about in terms of doing the hydraulic analysis and the calculations versus what’s actually going on in the control room and the nature of that and the need to collaborate.

I think you’re making some really important points to be thinking about as a liquids operator is what should the controller expect to see and how do they know. Where do they go to get something factual to compare to? It’s a big deal.

Jason:  Absolutely.

Russel:  Let’s shift a little bit. We’ll talk a little bit about startup. What would cause a line to go slack on startup?

Kyle:  Typically, if you’re getting ready to start up a pipeline, one of the ways that you can end up with a slack event during startup is that you would open the head gate at your delivery location. That allows the pressure to drain off into that delivery station.

If you don’t have sufficient pressure on the line prior to opening that or if there’s any type of delay that happens in your starting up the receipt end of the pipe, you can have sections of that pipe go slack during that startup event. Again, that’s where we would look for what is a normal trend that we would expect to see in that case.

Russel:  Right. I think, too, that starting up with some of my lines slack is fairly common. It’s certainly not desirable but it’s fairly common for the kind of reasons we were talking about earlier here.

What are the kinds of things I’m going to do if I know I have a line that’s slack and I’ve got to start it up? What am I going to be looking to do?

Kyle:  At least here, we want to understand how slack is the line and how much vapor space do we anticipate is in that line so that when we start it up, we have some understanding of how much fluid we’re going to be pushing into the line before we expect to see a response at the other end.

That’s an area where, at least with our leak detection system, there’s a time period where you’re putting more fluid into the pipe than you’re taking out, and that can be a cause for concern if it’s more than you’ve anticipated.

Jason:  The scariest feeling is when you know you’ve got a line that’s slack and it’s several hundred miles long, and you’re going to start to see meter counts on your receipt end and you don’t see any corresponding meter counts on your delivery end.

Just looking at it, you’re putting a lot of barrels of hydrocarbon into the line and you’re not getting any out. If you haven’t done the research prior to that to know that you’ve got 1,000 barrel vapor space in the middle of the line, that has every signature of a leak event.

Being able to tell the controllers that they need to expect that is something that’s very important. You don’t want to expect them to every time they turn it on think I’m going to add 2,000 barrels to this before I start to see my pressure drives.

That’s the easiest way to…We throw around the term normalization deviance. That’s one of the ways you cause that is the controllers get used to the fact that they’re going to turn a pump on and run the pump for 30 minutes and not see anything. The next week, they’re going to run the pumps for 31 minutes and not see anything.

Before it’s over with, you’ve got a pipeline that’s pushing hydrocarbon out onto the ground where you don’t want it.

Russel:  Right. Again, that’s kind of what I was driving at in the question. There is a need to understand where I’m starting from.

Jason:  Absolutely.

Russel:  I have a vapor space and how big is that vapor space and what should I anticipate once I start closing that vapor space. It’s interesting, and for people that aren’t familiar with liquid operations, kind of having a knowledge of, “Oh, that’s why this is hard, that’s why this is complex,” is important, even if you’re not really working in that domain.

Kyle:  Yeah, I know here we actually monitor…We track all of our slack lines. If any segment of a system goes slack, we receive an alarm in our OC, and we track each of those alarms.

We won’t restart that pipeline until the trends have been reviewed and that we’ve verified that it was a slack line that followed the normal or anticipated trends and that’s been reviewed by someone other than just the console. Then we’ll approve it for restart of that line. It’s something we take very seriously here in our operations center.

Russel:  Right. I think that’s probably the most risky thing that you do in liquid pipelining is start up a pipeline. It’s where you tend to know the least about what’s happening and where you need to know the most competitively.

Kyle:  Right, I would agree.

Russel:  I think that’s really important that you guys are laying out kind of a process of we’ve got to have some seniority, someone experienced look at it. There needs to be more than one set of eyeballs, and then we need to have a plan for restart. That’s pretty much every time we start up pipeline, we’ve got to have a plan for that particular restart.

Let’s talk a little bit about how slack line impacts leak detection.

Dan:  Sure. Which part of it do you want to talk about?


Dan:  There’s a lot of ways that slack line impacts our leak detection systems.

Russel:  I would say let’s just start at a high level because we could probably do a whole 30 minutes, 45 minutes just on that subject alone. Maybe we’ll just address it from a high level.

Dan:  In a leak detection system, the meters and the pressure transmitters are our eyes on the pipe. Pressure transmitters can see, at least in a liquid line, as far as there is positive pressure on that pipeline.

Our model can interpret pressure changes as volume changes anywhere along that line as long as there’s positive pressure. When you get to a vapor pocket, that vapor pocket — the pressure in that segment of the line doesn’t care what elevation changes happen in that pocket. We get effectively a blind spot for our pressure transmitter. Now, we try to compensate for that by having the model calculating the amount of vapor space, doing the heat transfer calculations to determine how much vapor we would anticipate, and establish where that vapor liquid barrier would be.

At that point, we’re really calculating a lot more unknowns than we have knowns. There’s a great deal of uncertainty that comes into our leak detection models when we have those vapor pockets in the line.

Russel:  Yeah. That’s a hard thing to speak about in simplistic terms, I think, because the nature of a fluid and how it carries pressure versus the nature of vapor and how it carries pressure are very different. That causes confusion, if you will, for the leak detection model.

Dan:  Yep.

Russel:  Yeah, it’s very, very problematic. I know that some operators have policy where they will not run with slack line. If they get slack line, they shut down and close it up before they run.

Some operators do because they have to. Because of the nature of the elevation changes in their system, it’s just impossible to not have slack line. They tend to be very on what’s going on with slack line and how that’s affecting things. It’s a big deal. What are the key things you need to do to mitigate slack line?

Dan:  Well, some of the methods we talked about earlier are backpressure valves, then detailed shutdown procedures to keep us from going slack.

Another thing is if you are going to experience a slack line is understanding the limitations of your leak detection system and looking for any way that you can to supplement that in the areas where you’re concerned about being slack.

Kyle:  Then the running condition we have station suction set points that will keep the gradient. There’s only one or two places on all of our lines I can think of where we ever go slack and it’s extremely, extremely rare because we have suction set points high enough to keep the gradient tight, as Dan was saying, to keep positive pressure at all times while we’re running.

Russel:  A lot of times I like to try and do kind of a how do you wrap all this up. What are your three key takeaways? I think this one is a little bit more challenging maybe than some of the other conversations we’ve had on the podcast because there’s really a lot that goes into this.

  1. I think the first thing is slack line increases operating risk. That’s a key takeaway. It increases that risk for a myriad of reasons. It impacts the effectiveness of my leak detection and it creates the opportunity if not properly managed to create pressure spikes that I don’t want to have. I think the first thing is just slack line represents risk.
  2. I think the second takeaway is from an operational standpoint, you’ve got to have a plan each time you have a slack line and that’s got to involve the appropriate people and experience to evaluate it and address it.
  3. Then, lastly, that when you have slack line, your leak detection is impacted and you may need to think about alternative means of leak detection when you have slack line.

There’s one other thing I’d like to get your comment on. That is for periods of time when a pipeline is shut in and you’ve got cooling operating. Do you think it would be appropriate to have a completely different type of computational leak detection for that situation? What might that look like?

Jason:  I’m not the leak detection expert in the room. That’s Mr. Sensel, but I’ll answer it from the hydraulics perspective. If you’ve got a line that’s going to be slack for a while, one of the things that I think is important to look at is what your historical pressures are.

The changes in elevation create static pressures on your pressure transmitters. If you’re looking at a transmitter and for the past five years of its life it has never dropped below 17 PSI and today it’s at 10, that’s a pretty big smoking gun that something abnormal is going on.

Controller monitoring is a recognized leak detection method. One method I would say would be an alternative way to look at the line. Dan’s got some experience with other systems he might weigh in on.

Dan:  Yeah, from taking the look at what’s historical, you can also apply some type of statistical model to your pressure trends to show that it’s within the statistically normal range for what your trends have done during that shut in time period.

Some other things we can do is if you know that a line is going to be down on slack for some period of time, as Jason was saying, the pressure shouldn’t change on that line. You can tighten up some deviation alarming or some limit alarming on your SCADA system so that instead of a large pressure change signaling an alarm it might be a couple PSI will give me an alarm.

You also can increase aerial patrols or you can increase ground patrol of your line to supplement that leak detection system while you’re in that slack condition.

Russel:  I think, Dan, the kind of thing I was driving at in my question is the idea of watching your pressure, because theoretically if I have a line isolated, I should hold my pressure.

Dan:  Within some bounds based on how much cooling you have.

Russel:  Right. Exactly. Which is a number you can mathematically model. It’s interesting. I think as I look at leak detection in our industry, a lot of our focus in the last 10, 15 years has been on the ability to find the smaller leaks in normal operating conditions.

I think that focus is beginning to change or at least be extended where we’re starting to look how do I find leaks when I’m not during startup or shutdown or shut in. How do I find leaks in those situations? Because actually that math is a little different.

Dan:  Absolutely.

Russel:  Guys, look. This was great as always. I really appreciate you coming on board. I’m looking forward to having some more of these conversations.

Jason:  Us too, Russel.

Kyle:  Thank you, Russel.

Dan:  Thanks, Russel.

Russel:  I hope you enjoyed this week’s episode of the Pipeliners Podcast and our conversation with Dan, Kyle, and Jason with Marathon Pipe Line. I certainly did and I certainly learned some new stuff.

Just a reminder before you go, you should register to win our customized Pipeliners Podcast YETI tumbler. Simply visit pipelinerspodcast.com/win to enter yourself in the drawing.

If you would like to support this podcast, please leave a review on Apple Podcast, Google Play, or whatever smart device podcast app you use to listen to the Pipeliners Podcast. You can find instructions on the website.

[background music]

Russel:  Finally, if you have ideas, questions, or topics you’d be interested in, please let us know on the Contact Us page at pipelinerspodcast.com or reach out to me on LinkedIn. Thanks for listening. I’ll talk to you next week.

Transcription by CastingWords

Pipeliners Podcast © 2019