Episode 350 | Performance Benchmarks

0:00Drilling fluids touch just about everything in the drilling process. We're here to deconstruct the drilling process and drilling fluid concepts to provide a deeper understanding of our industry. In each episode we'll share information, talk to interesting people and maybe share a few stories along the way. Welcome to the Flow Line, a production of AES Drilling Fluids brought to you by Matt Offenbacher and Justin Goethe-A. Welcome back to another episode of the Flow Line. We got Matt here and myself, Justin. We've got an exciting episode today, Matt. This is one that we came up with within the last, maybe six minutes, but it goes deep and the fact that you were able to rip something together, some show notes and a guideline for us is pretty impressive. I was telling Matt he's going to look really smart after this and so hopefully you hang on to the bitter end. But Matt, before we get into that, how was the weekend?

1:00Anything fun on the personal front that we can share with the lovely audience? I mean, I had, I took some extra days, did a church retreat to Mo Ranch in the hill country and beautiful along the Guadalupe River. My oldest caught his first fish, rode in a canoe for the first time, went down a big water slide. Oh man. I missed, you sticks in the dirt and had the time of his life. So I think, I guess, as an aside, I think, you know, the hill country is one of these gems of Texas that I think almost everybody feels connected to if you grew up in Texas, going to summer camps or just what have you. And so I was really excited to see my kids kind of get their first real taste of that. And certainly after kind of the devastation that happened last year, see that that area is starting to recover, although I think you can sense there's still some wounds that

2:00are, have yet to heal. But all that being said, it's just such a wonderful place. And so I was really neat to relax and spend a lot of time with friends and all that kind of thing. No, that's great. We were actually, I guess you could call it someone hill country. It's just southwest of Fort Worth Lake Granberry. Yeah. Beautiful area. Yeah, it was really nice. A friend of ours has a place out there, has the boat. And so same thing, my kids got to do something for the first very first time. I get told on a tube behind a boat. That was fun. My daughter tried wake surfing. She didn't quite get up, but she kept trying. And my wife was on a tube. She took my kids about 10 minutes to convince her to do it. I don't know why she was scared to do it, but she was. And so that was fun. Yeah, but just that part of the, part of Texas is beautiful. And yeah, we took the boat. And then when there's like little ice cream shops on the water and like, you know, us adults had a margarita, well the kids had ice cream.

3:02Everyone had a good time. But yeah, it was, you know, again, summertime. And it's just wild that you could live here for 10 years. And I didn't know about this place until maybe three months ago. But just, yeah, Texas is so big. There's so many spots that you just, you never know. There's things that are places that people mention. It's like, I'd never even heard of that. So, but anyway, it's done like we both had a great weekend. It sounds like it. And now we're ready to get into drilling fluids topics. Yeah. I mean, are we ready to pound into one? Yeah. It's like limping out of the weekend. Yeah. No, for sure. So, man, it was something I kind of had brought up and was well-worse stability. We've talked about it before. But you know, it's one thing when we talked about what, when we talk about well-worse stability, just operationally, you know, on a rig. But you can get really deep in it and actually model well-worse stability in different mechanics and really try to get a better understanding from a scientific perspective of what's going on down hole. And so just kind of going through that, walking through it and talking about, you know, things

4:04that we would need to properly model what's going down on down hole and then just talk about some of the, you know, key words in different, I don't know, terminology, if you will. Because it can ultimately come in totally different language unless you know what you're talking about. So, again, I think it was, it's a worthwhile one. What do you think? I think so. You know, it's also interesting. You were having a conversation about a particular term that we worked into this and it was one of those like, yeah, we throw these things around and we don't think much about them until somebody asks us what it means. Yeah. So maybe a review is long overdue. Perfect. No, I like it. So let's get into it. Just sort of at a high level. Let's discuss well-worse stability models and just the overall mechanisms of how we come about these. Well, you know, a lot of it is, are you just doing something based on physics, you know, do it? How heavy is my rock? What are the stresses?

5:05And you know, that's a mouthful in and of itself, but like a purely geomechanical model which is basically saying I'm going to pick my mud weight and you know, angle of attack and that sort of thing. There are folks who have made efforts actually to include chemical effects. So when we've talked about, you know, reactive clays or perhaps even chale embrittlement with oil-based mud, recognizing that auto-sumatic effect and how it can not only affect the shell that you're drilling through, but then how that affects the mechanics in light of the changes that take place. And the important thing is if you're able to get these models down, which is quite the feat, you can identify maybe a window or a set of conditions for the most stable well-wore regime. So sometimes, you know, we've had customers in South Texas where they have a bunch of instability and it's clear that they are drilling at an angle that promotes well-worn stability. The question is, you know, a lot of times if you say, well, don't do that, can they actually

6:08get to the reservoir target if I say change in a different direction? Sometimes it could be just like nudge a little earlier or you know, try and get through that and then, you know, maybe even do a harsher kickoff. There may be some risk to that too, but you're trying to balance out all of these things in light of the circumstances that you see. Yeah. No, that's a great point. What would you say is a lot of time the biggest challenge to come up with these models? It is just getting enough data and this can come from all over the place. So in essence, almost every well-wore stability model does not have a perfect data set to perform the analysis. If you think about a big offshore well or something like that, it's much easier to get core samples and, you know, in situ pressures and, you know, very specific measurements because the wells are expensive and if you're drilling a bunch of wells in the field that all costs a lot of money, the savings speak for themselves. Yeah. A lot of times what we run into is somebody wants a model, but you know, it's a confidence

7:12interval thing, right? So you give your best guess or you give a range and then you sort of try and see how that falls together relative to what you're experiencing. You know, it was quite interesting a while ago because even on the high spec wells, there were several well-bore stability experts from big service companies that spoke and it was sort of ironic because they were basically like, yeah, you can't actually get the in-situ salinity of shale without a formation tester like down there. Like you're spending a lot of money on a log or a tool. The other thing, you know, we'd ask how easy is it to get core samples and a lot of well-bore instability incidents happen usually in the intermediate. Not to say they can't happen in the reservoir, but they said, you know, a lot of times we can't get core samples and the only time that we do is when they didn't get their depth correlation right and accidentally pull the core from the intermediate when they're trying

8:15to core the reservoir. And so you just talk about, you know, these worldwide experts who are being brought in to every major place across the world and they can't get the data. And so one thing we illustrate is, you know, there is a table. There's actually an SPE short course on well-bore instability that was pretty great. And it just talked about the sheer volume of data you have to ask for and then the question is, is any of it currently reliable? Has anything changed? That sort of thing. In essence to the point where for most of our types of applications, the conversations we have, it's going to be, well, let's do our best guess, but it may help us rule something in or out. Outside of that, it's not a very precise mechanism or model. And so when we kind of do a little intro before each section of our mud school and we do

9:18well-bore instability, we've sort of compared to chaos theory. And I don't know if you've heard. I've heard the term. So how this was sort of discovered was way back when they thought, the thing was like the 60s, they thought they could perfectly model the weather. I don't know if you heard of the butterfly effect. Right. Where it's all these like knock on circumstances lead to an outcome. Well, what they were doing is they were doing these computer models and they were somewhat confident that they could very precisely model the weather within a matter of years. They really thought they were close to it. And so they're doing an extensive computation. It's down to like six decimal points or whatever. And something gets fat-fingered. And so one of the computer scientists is, you know what, I'm just going to run it again, but I'm only going to run it to the thousands so I don't have to wait as long for this big computer, room size computer to give me the answer back. Yeah. Anyhow, totally different unexpected data.

10:19And what it ultimately led to was the realization that there are so many dependent variables involved that this is going to be far more difficult to predict than we ever thought. And so now you get to today where, you know, weather is pretty, still pretty tough to predict because there are so many variables to measure. And, you know, relationships and dependencies. And so anyways, the short of that is there's this whole slew of data and it can be so overwhelming in some cases that a precise model isn't as readily available. There are things you can do. There's a lot of people, the more rudimentary ones. A lot of people like to criticize them as opposed to consider them. I'm sort of, hey, if it, if it, not just as in a direction with some sort of confidence, I would say like let's at least try it.

11:20But just thinking about all the things you would need for a well-wrestled ability model, the first one is just, you know, the vertical stress. So like, can I basically get, you know, how much rock is on top of rock? Or what is that on, you know, what is that effective density of the rock? You know, acoustic, seismic can usually give you a reasonable idea of this. Okay. Then, you know, poor pressure, a lot of times that's like an actual formation pressure test where you actually have to like press something up against the form of seal in the formation, draw in, determine the, you know, pressure equilibrium. This once again, maybe you can do something indirectly with acoustic resistivity, something like that. But, you know, with us, our poor pressures, we have such weird pressure regimes because we have subsurface injection. We have depletion, but, you know, it's in a shell which most of the depletion is tied

12:23to where you fractured. Other areas, shell doesn't have enough communication, you know, permeability, which is why you frack it. Yeah. To communicate that pressure loss elsewhere. So you know, that's a pretty challenging principle. The other, you know, another one, least principle stress, this is SH-min, horizontal minimum stress. The most common is would be like a leak off test or specifically an extended leak off test where what you have to do, and we've talked about, you know, leak off test and that sort of thing, is in essence, break down the formation and see the pressure and then see how it sort of recovers. So it's sort of like a two-step leak off. Sometimes called a mini-frag or, you know, if you're doing fracturing, a mini-frag to see what your rock stresses are. Cross circulation may tell you like how you break down the formation and what pressure does that take or like ballooning because you know the formation is stressing.

13:24Those are a bit more of a reach. But the other thing is, you know, if I know my least principle stress or my maximum horizontal stress, it should be like 90 degrees from that. So that one's a little bit easier to figure out at least, you know, looking at well-bore failures, sometimes, you know, an FMI log will be able to tell you based upon the formation movement. And then you got to figure out the strength of all the rock and how it behaves. And I've got a few buzzwords for our notes later, but this is where I need core material to see how the rock actually breaks down or stretches, how elastic is it, those kinds of things. Once again, you know, logs may be able to help you there. Cuttings sometimes or there's always an effort, can we do more interpretation with things we can acquire at no extra cost. And then well-bore failure analysis.

14:26So sometimes you have an idea of what's going on or you can back it out because of the cavings you see or, you know, what happens. Yeah. So in this obviously is a lot and it's not something that you just find on the mud report. Obviously, would you say in generally speaking it's challenging to get this data and if so, or regardless, where would one access this typically? This is always, so one, it's, can you get the data? Is the data current? You know, sometimes when we are trying to do this for our customer, we'll go, I mean, I'll be reading academic papers from the 40s and what I believe is a representative analog offset formation. Yeah. Some things don't change about the rock. Some things do as you deplete a field and, you know, some of the, you know, some of that stuff gets, gets modified. So some of it we do our best. We look at logs with a lot of our customers, even if they can give us like the log, it

15:28may be that, you know, they don't have one young's modulus. But they have the log that shows what the interpreted young's modulus should be with a range. And what we can do is we can take a statistical distribution of that, you know, minimum maximum most likely, right, and then use that range and we'll do what is called Monte Carlo simulation. So basically force it onto a distribution curve with all the other variables and say, okay, then min max most likely of what we have with the range of data we were given. This is probably where we are. So, you know, when you want to have greater confidence, this gets really expensive. Right. And this is where we're sort of pushed out of the mix of having a strong confidence interval for the types of wells we drill. But I did, you know, you mentioned Poisson's ratio as kind of a key term. I put a couple of these other terms that we acquire here.

16:29But you know, Poisson's ratio, it's basically, you know, does a rock expander contract when it's, you know, stretch or compressed along its length. The best visualization I've ever really seen is like, it was a bunch of short little straws all glued together. And when you bend it, so it was like a flat piece with straws all glued side by side, you bend it and, you know, some of it compresses and some of it flexes. And sort of the idea is, you know, how is this thing going to behave under stress? And then Young's modulus would be, you know, if I have a high Young's modulus, it's stiffer and it doesn't want to deform. A lower one means, hey, it's, you know, going to stretch, it's going to exhibit some mobility. And then the horizontal minimum stress is the lowest per per nuclear compressive force acting on the rock, where we're basically trying to figure out, in essence, like, if I

17:35break down a formation, it's going to happen on the maximum horizontal stress. And so I know I'm 90 degrees from there. If I look at a borehole. But those three calculations, if you're doing well, we're strengthening, those are like foundational to figure out, how is the rock, how is a fracture going to open up that I want to seal? How big is it going to get? How much am I going to like try and pack in there where, let's say it's one of those scenarios where I need, I need more mud weight to keep the whole stable. Yeah. Well, I need to strengthen the formation. So I need those variables as confidently as I can put them together. And then I figure out what material is going to pack inside that induced fracture. Yep. But to efficiently do that, I want to have some idea of how big it is. And I'm going to pressure up to what I need to to hold that mud weight, which means I need to think about those things. So all of those are some pretty key terms, both in well-wrestability and well-wrestling

18:36thing. And then there's other mechanisms that come into play. So temperature, there's actually an interesting paper where they actively heated the mud and did a different leak off test. And they got higher leak off as the mud was warmer. This is, you know, from my experience in geothermal, well-wrestling stability is kind of a huge problem because you have really, really hot rock. Yeah. And you're trying to keep the tools cold. And so you keep pumping cool water down a very hot formation. And if you just think about, you know, a circle of the borehole, if I shrink it, what happens? You know, all that stuff is more compressed. And so the likelihood of it breaking off. Yeah. So it's going to happen. Now, I mean, we don't want to heat up our, most of the time when we have mud chillers and that sort of thing, we're trying to keep things from heating up. Right. You know, but it could be, I don't think there's much you can do with that information per se, but it is, you know, a phenomenon.

19:36The other thing is, you know, sometimes we have well-wrestling stability issues with, like, you know, not just, in a lot of areas it would be you're changing formations, like drilling into a salt or what have you, which is, which is going to be true, but, you know, in the Delaware basin, for example, you have this salt you drill through that, you may wash out, you made a zal part of it, and some of it's soluble and some of it's not. And so you have, like, rubble and other things that are falling in. And that's much more of a, you know, chemical effect of the nature of what you have than it is, like, I need more mud weight to keep this rubble from coming in. Yeah. And so you have to think a little more holistically there. Okay. Talk about, so we've got obviously temperature. What about, talk about the dissolution when it comes to rock material. So that was a salt, I think, and hydrate can also, you know, sometimes we have a player in this. I've seen some criticisms, you know, oh, dispersive shell and you, you know, destabilize,

20:38it's got a rubble zone above it, so you disperse the shell and now I have stuff coming in. These are all, you know, very possible and things that aren't going to show up. They should in your formation evaluation, but they might not show up, you know, with your geomechanical model because you were so focused on the competent rock that you could measure. Right. Okay. Talk about some practical efforts we can make to help enhance our model or to improve it. I think the biggest thing, you know, outside of them, you know, a model, the big thing is caveings analysis. And we have talked about this some, but I think looking at your caveings, actually looking at the shape, taking really good pictures, get it, you know, correlating them with a depth, having a better understanding, you can actually see, you know, if there are sharp edges on it, okay, that's a relatively new section of breakout. Any of those kinds of things can tell you a little bit more, you know, splintery

21:42caves. Hey, I've got over pressured. This is a poor pressure issue. So those sort of things help you, one, on the fly, you can look at that and you have a pretty good idea of what's going on. And two, if you do have a model, you want to update, this will tell you whatever it was, got it wrong. You need to tweak your numbers. Yep. So there's a big part there, you know, in this whole thing, I would say one of the big challenges we have is, you know, we have access to offsets, but we don't necessarily have access to logs. And I understand why customers may hold those closer to the vest or frankly, I think there's also sometimes just a massive disconnect between our, you know, subsurface and drilling. And so the big issue there is, you know, the drilling folks will say, yeah, like I'll set up a meeting and like, nothing happens. Or you just don't get, you know, the subsurface people are like, look, I'm doing other things.

22:44This is the least of my worries. Or we just sort of struggle to, you know, get together on the data sets. But if we can look at logs, you know, if we can look at offsets, or even just having a subsurface person come back and say, I see the information you've gathered and here's what I think is reliable and here's what I don't. Can improve our confidence interval where we can say, all right, well, I got a better idea than where I started. But I will say that, you know, I've been privy to see outside party and mechanical earth model data sets. And they're pretty limited, I think, just because, you know, even a specialty consultant who's offering these services can't get the information. Yeah. However, it's also shocking how much is taken for granted where you catch the right person and they're like, oh, yeah, we got some of that. Like, we use it for a completely different purpose.

23:45We would have never thought somebody in the drilling realm would care about this. Yeah. And so when we can make those connections, things, I think, get a lot better. But generally, it can be kind of tough. So that's where I think going back to your Caving Analysis, looking at just offset data, what worked, what didn't, can be a big step. But if we want to make the jump, maybe not even to a model, but just a, I don't know what you call it, like a vague circle of confidence that we can maybe tighten up a little bit. It involves some information. We don't typically access, but when we do have it, we can at least know for wrong. Right. So I think there's a lot of that. And in general, we sometimes risk over complicating these things. And other times we under complicate them. Yeah.

24:45I mean, we've talked to the, in other episodes about how often somebody wants it to be water-faced salinity so they can just change their salt concentration and not have to wait up. And it's like, I'm sorry. It's kind of where we're at. Yeah. But yeah, so I think, I don't know if that, that rundown is helpful to the listeners, but, you know, this stuff is complicated. And I tell you what, there are some really smart people who do well-wrestling stuff, but normally they're packed away somewhere within an organization. And you know, it's, I've been very lucky sometimes to get to have a conversation with them and learn a little bit. You know, they're using, they carry the really big heavy laptop around because they're doing reservoir modeling, they're doing all this other stuff. Yeah. So it's, it's a lot. So let me ask you, so assuming an operator is willing to share some of this data map, what are you and the team able to do? And then what is the output to where, like, how does it actually affect the rig site?

25:49Like, how, you know what I mean? Can we connect the sort of the, to the mud engineers sort of day to day, if that's something that would be, again, achievable with the right data? With the right data, so one thing you can do is you can look at your stress regime and say, look, this is the, this is the mud way it's going to take to drill in this direction. Or guess what? You probably shouldn't try and drill in that direction. The reason you're having all these problems is because the rock does not like going in this direction and we do not have the mud weight window to do it at this density. Or perhaps it's our, our window is narrowing now and we need to look at managed pressure drilling and run a much tighter site, you know, cyclical pressure to manage this. I think there's other stuff that you would just, I mean, it can confirm that you should tighten your fluid loss. But guess what? Anytime we're having an instability incident, it just seems prudent to tighten it up, right? Like, you don't need to ask me for that.

26:51Right. And so we might be able to explain why it's helping. Sure. But some of that stuff, I think, I think, comes pretty obvious. Where I think it takes away the misery in the field is the times we can sort of make that recommendation in the office and then you never have the problem in the field. Right. Preplanning exercise, I think this would be something that would be extremely useful. And maybe perhaps more than an exploratory or exploratory type environment. Yeah. If you've got a little bit of data and it's a new area, maybe you're having issues, then obviously this would be good. But I would imagine that they don't do a lot of like rock mechanics or hostability studies in some ways like the heart of the eagle ferd where they drilled tens of thousands of wells and they know kind of what's expected, right? Like when have you used this type of modeling and you've seen sort of the benefit come from all the work? There are some places in, okay, the benefit is a bit more of a discussion, but it framed

27:52a better conversation. I mean, there has been some stuff in South Texas, Oklahoma, which, I mean, the hard part with Oklahoma is the geology is a mess. But at the very least, you can say, look, I know what you think you're seeing, but here is a reason that perhaps that's not entirely the story. Right. You know, I think in Oklahoma, what's harder is getting the field to not keep doing what they're doing. But there are also some people that are very open to, hey, we'll try anything. We're having problems. So there is a place for it. Once again, the hardest thing is we have someone who says, I have a problem when you take a look at this. And we get excited because these are fun. This is what we do, right? We live to solve problems for the customer. Yeah. So we say, okay, like, saddle me up. And we get kind of a mishmash of data sets that don't necessarily provide the foundational

28:54information we need with any confidence interval. It's just, here's some stuff. And then it kind of dies on the vine. But this is where we've obviously done a lot of improvements in how we can align data sets to drilling activity and that sort of thing. And so that should automate and streamline some of these things where maybe we can correlate things a bit more quickly. And so I am hoping both between the longer laterals we drill, which require more, your window is getting tighter. That means more precision is required. The risk of having problems is higher. But then the other aspect of that is the, sorry, I'm summing through my words. But the other aspect of that is I think there's going to be more interest in this because the need is going to be greater.

29:54And so when we can do better engineering to drill five miles, six mile out of a, what out you name it? Because we couldn't before due to some of these geomechanics issues, then I think it'll be like, okay, yeah, like let's keep the pipeline open. So we've been spending a lot of time looking at how we can leverage those data sets when it gets a little bit harder to get away with things we were getting away with with three mile out of a. Yeah. No, I think that's really where it's going to come down is people keep pushing the limits. The stuff's going to become just as more important because the risk tolerance is lower. Right? Obviously, it's a ton of risk going beyond four miles, but you want to do all your homework to make sure that you're not running into trouble at four and a half miles and you're trying to get to five in that last half mile cost, you five million bucks. So yeah, I could see that happening, but it's cool. It's an interesting conversation. It's one that we talk about well-washed ability, more from just sort of like an operations

31:00level, but a lot of science can go into making sure you're making the right decisions and planning the well correctly. And so anyone out there that's interested in this stuff, hopefully you found it useful and informative. If you have any questions or want to dive deeper into any of this, let Matt know and I'll be sure to relay the question. And again, with that being said, any other topics or ideas for episodes, please let us know, ask the questions. We'll do our best to answer them. Matt, anything else before we close up? No, I mean, I'll be honest, I think I'm a little exhausted from running through all these terms. So I hope we didn't bore any of our listeners, but it's a lot of stuff, right? This is one of those areas I continue to learn and I'm challenged by, but hopefully we inspired some thought and those who made it to the end of the episode. That's it, you got to keep the act sharp, buddy. All right. Well, that being said, folks, check us out on LinkedIn or you can reach us at the Flowline podcast at aesfluids.com.

32:00Take care for now, stay cool and we'll catch you next week. Cheers. Take care. Thanks for listening. Please tune in next week for another exciting episode of the Flowline. And remember, may your returns always be full and your trips always smooth. Choose express in this program belong to participants and not their employees. The program is for informational purposes only and cannot take the place of seeking professional advice. Copyright AES Drilling Fluids