Latest / Data Driven with Lou Peck / Tim Cheek and Matt DiSogra: heavy vehicle data and its proliferation, the modern HVEDR toolkit, and the future of the field.
Transcript
- Lou (00:00:19): This episode is brought to you by Lightpoint, of which I am the principal engineer. Lightpoint provides the collision reconstruction community with data and education to facilitate and elevate analyses. (00:00:29): Our most popular product is our exemplar vehicle point clouds. If you've ever needed to track down an exemplar, you know it takes hours of searching for the perfect model, awkward conversations with dealers, and usually some cash to grease the wheels. Then, back at the office, it takes a couple more hours to stitch and clean the data, and that eats up manpower and adds a lot to the bottom line of your invoice. Save yourself the headache so you can spend more time on what really matters, the analysis. (00:00:53): Lightpoint has already measured most vehicles with a top-of-the-line scanner, Leica's RTC360, so no one in the community has to do it again. The exemplar point cloud is delivered in PTS format, includes the interior, and is fully cleaned and ready to drop into your favorite program, such as CloudCompare, 3ds Max, Rhino, Virtual CRASH, PC-CRASH, among others. Head over to lightpointdata.com/datadriven to check out the database and receive 15% off your first order. That's lightpointdata.com/datadriven. (00:01:23): All right, good day, ladies and gentlemen. We have a twofer today, two gurus in heavy vehicle event data recorders, Tim Cheek and Matt DiSogra. And I'm going to read in their bios real quick, give you a little bit of a flavor about who they are and what they have accomplished, and then we'll get right into it. (00:01:40): So Tim Cheek is a vehicle products expert at DELTA |v| focused on vehicle design, manufacturing, maintenance and operator issues related to accident analysis and prevention, vehicle fires, and vehicle engineering. Prior to DELTA |v|, Tim spent seven years in product development at commercial vehicle manufacturer where he investigated accidents of various types, including fires of truck and bus products, and assisted legal counsel in the defense of product litigation. Tim is very active among his peers and professional societies, such as SAE and ASM, holding leadership roles at the local, national, and international level and receiving numerous awards for his outstanding contributions. He continues to be actively involved in various standards committees, publishes technical research, and instruct SAE's highly lauded seminar he co-developed called Accessing and Interpreting Heavy Vehicle Event Data Recorders. (00:02:33): And he is joined by Matt DiSogra, who, from what I understand, is the student that is becoming the master. And Matt has been an engineer with DELTA |v| since 2013, where he specializes in recovery and analysis of data from damaged EDRs, and has authored several related SAE papers. In addition, Matt co-instructs the aforementioned SAE seminar with Tim. Matt is also a member of the SA, SAE, sorry, J2728 Heavy Vehicle Event Data Recorder subcommittee of the Truck and Bus Council. He received his bachelor's and master's in mechanical engineering from the University of North Carolina and is also a licensed professional engineer. (00:03:16): So that should help you understand why we're interviewing these gentlemen today, because they are the elite, they are the teachers of the industry. So thank you, guys, for taking the time, making the time is probably the better way to say it, to join in today. Matt (00:03:33): Happy to be here. Tim (00:03:34): Yeah, thank you. Lou (00:03:35): So, I wanted to start a little bit and talk about the way that I understand DELTA |v| from across the country is you guys are a very high-end reconstruction firm specializing in heavy vehicles and heavy vehicle event data recorders, retrieval of that information, and interpretation of that information. Correct me if I'm wrong on that front, but if not, could you guys outline just real quickly each of your roles there and what you're doing kind of on the day-to-day level? Matt (00:04:05): Sure, I'll go first. So I oversee the group in DELTA |v| that handles basically all things EDR, specialty EDR. So that's onsite imaging of vehicles, that's benchtop downloads from modules from vehicles, and then that's all the way through the data recovery of damaged modules. So we kind of handle all facets of that aspect of things. And so we support our own engineers internally with that service, and then we also provide that externally to clients across the country who need that level of depth on their cases. Tim (00:04:42): Yeah, And I'd say I was Matt before there was a Matt. And then when Matt came along and he decided he wanted to fly on airplanes a lot, I said, "Matt, here you go." So actually what I do day-to-day now is I'm focused on case work, a lot of product liability work from my time with a truck manufacturer, work on various aspects from steering to braking to, it could be fires, things like that. Still involved in EDR, can't get away from that. Certainly in my day-to-day work, I have to use it, but that's what my focus is right now. It's not totally EDR. Lou (00:05:28): Yeah, and we were talking about it a little bit before recording and we'll go into it certainly in more depth, but it seems like something you have to throw your whole self at. And if you're not doing that, then you need somebody like Matt to kind of back you up. I think we all need somebody like Matt in the community that we can lean on a little bit. So yeah, we're going to be talking obviously a lot about HVEDRs. And the majority of the people who are listening will know what that is, but for those that might be outside the industry listening, the HV is heavy vehicle and the EDR is event data recorder. So I guess let's just first start out with, what is a heavy vehicle in this context? How do you categorize that? Tim (00:06:05): Yeah, I'd say a heavy vehicle, I guess if you were to classify it, they talk about light duty vehicles are 8,500 pounds or the less. So when you look at Part 563, that's what they apply to, under 8,500 pounds. Heavy vehicles, typically we're looking at it, the other way they classify it would be classes. So there's class one, two, three, four, five, six, seven, eight. So we're looking typically at class six through class eight. Remember, we're talking about heavy vehicles. So six would be medium duty trucks. Seven and eight would be getting into heavy duty trucks. Lou (00:06:46): Okay. And then the EDR part, event data recorder, the description, the name of it tells us most of that. But Matt, do you want to talk a little bit about what the E is, like the event? How are those modules trying to determine whether or not a, quote, unquote, "event" occurred? Matt (00:07:05): Yeah, it's a lot less well-defined than the situation is in the passenger vehicle world, where you have something like Part 563 that lays a foundation for what those things are, how they're supposed to work, and nothing like that exists in the heavy vehicle realm. The closest we get is an SAE recommended practice, J2728, which lays that out, but there's no requirement for manufacturers to abide by that, so it's all voluntary. (00:07:36): And furthermore, every manufacturer is free to implement the recorder as they please. So events can be things like a trigger based on a wheel speed change. They could be things like a trigger based on a fault condition. They could be a trigger based on, let's say, a forward collision warning if the vehicle has radar. The field of possibilities is pretty vast when it comes to what is the event in question since there is no set of regulatory definitions for it. Lou (00:08:06): And are these generally set up for accident investigators? In other words, why does Freightliner care about events? Matt (00:08:14): Generally, yes, right? There's not a whole lot of other reasons to put something like a hard brake recorder, for example, other than looking for these sorts of things. Although, to be fair, a hard brake doesn't necessarily imply some sort of accident has occurred, right? That can be generated with a truck just driving down the road. So there are probably some other uses, but I would say largely it's aimed at our community, despite maybe not saying that on the surface. Lou (00:08:39): Yeah, and making it very difficult for us at times to access the data and making the data somewhat volatile. So they're like, "Here's some data, but we're not going to make it that easy. You guys got to bring your game if you want to get into that." Matt (00:08:51): Yeah, or maybe not even stand by it. For example, Cummins, in their event data reports, at the bottom, puts a disclaimer that basically says something to the effect of, "This is not intended to be used for accident reconstruction." So they've done a blanket coverage of their own company there by saying that. Although, there's been plenty of research and papers published that validate that, so it is valid to be used. But right, they make it difficult in a lot of ways for us. Lou (00:09:17): Yeah, I remember, I think Toyota did that for a while and then you show up at depo and that's the first thing that gets brought to your attention. Tim (00:09:23): Yeah. I would add to what Matt said about the distinction between light duty vehicles and heavy duty vehicles. With light duty, you have the airbag that triggers typically the event. Everything's tied to that, to the airbag, and that's where the data's recorded. So it's pretty clean when it comes to passenger cars and event data with the CDR. But then we get to heavy trucks and most heavy trucks don't have an airbag, so where do you start, right? So if they don't have an airbag, even if they did have an airbag, how do you really define that collision event? So for a heavy truck, the crash pulse is going to be much different than it is with a light duty vehicle. So where do you set that threshold of sensing and then triggering for event data recording? That becomes a challenge, right? So that's why you bring in other types of triggers that Matt talked about, these strategies to try to capture various events. (00:10:28): And so with the SAE committee that Matt and I are both on, we looked at various triggers. There's now in Europe a UNECE, that they're looking at event data recorders for all sorts of vehicles. They wanted a one-rule-fits-all for every vehicle, from light duty to heavy vehicle, and it took a lot of education within that group to say, "Look, they're different and you have to approach this differently in terms of triggers and maybe even in terms of the event window that you're looking at." So if you're looking at five seconds of pre-trigger data for a car, is that sufficient for a truck? Or even you look at post-trigger data, do you have the same requirements for light duty versus heavy duty? Because with a truck, oftentimes when there's an impact, the truck driver has to control that vehicle and will be in control, can be in control, unlike a passenger car that, once there's an impact, they're probably not in control of the vehicle anymore. Lou (00:11:39): Yeah, that's interesting. I've never even considered that. Because I've seen enough reports and done enough work and worked with enough experts like you guys that I have seen that there is that post-impact documentation in the EDRs for the most part, but I never really thought about why, I just always appreciated that it existed. But that's a good point. I mean, you get a 60,000-pound truck, if that's a reasonable number, that it's rear-ending a Camry, and the delta-v for the truck might be two miles an hour and the driver's still fully in control afterwards. So it's nice to see. Tim (00:12:10): Yeah, and that gets to, when you do have an airbag on a heavy duty truck, where do you set that threshold for deployment? Do you want that airbag in the truck driver's face even for, say, a five mile per hour delta-v if they've got to control that vehicle, or even a 10 mile per hour delta-v? Lou (00:12:28): Right. Tim (00:12:28): Potentially, can they be controlling the vehicle in a 10 mile per hour delta-v? I don't know the answer to that, but therefore, it presents just a different set of challenges. Lou (00:12:42): Yeah, that's really interesting. And do all of these trucks have... Well, let's call them heavy vehicles, because they're not all trucks, some of them are buses. I imagine there's a big array of potential vehicles. Do they all have EDRs at this point? Matt (00:12:58): I would say for the most part, yes. When we take the broad definition of EDRs, you are going to find them in most of them. Some might have multiple, right? Because there's no standard, like Tim was talking about, everything in the passenger vehicle centers around the airbag control module. So on a heavy truck, you might have a data recorder built into, let's say, the engine control ECU, but you might also have one built into the ABS ECU, so you could have a truck with two, or maybe it doesn't have one in the engine, but it does have one in the ABS or whatever. There's all different combinations of it. So as of now, most everything on the road, at least one of those spots will have an event data recorder. But sometimes we have multiple, and that brings up a whole other interesting set of challenges. Lou (00:13:48): And that's where I'm like, "Okay, I don't even know what modules are on this truck. I don't know what modules I should be looking for. Therefore, I just need to know who the right expert is." Tim (00:13:57): So I think what we're seeing with heavy trucks is there's a lot of aftermarket devices that you can find on them. So you have ELDs, which stands for electronic logging device. You think of your truck drivers have to keep hours of service, and so these ELDs now are doing that automatically. It's a requirement for trucks to have that now. And so it's easy now, when you have an ELD that is based on GPS, to now add other functions to it, like event data recording, right? So that's where you have Omnitracs and PeopleNet and those various systems that you see out there. They're a pay-for-play type system, à la carte, so you're not going to have it for every system out there, it depends on what they are paying for. (00:14:55): And then dash cameras, of course that's unique to trucks for the most part. People have private dash cameras in their cars, but it's getting more and more common to see dash cameras in a truck. They may be standalone systems, but then more and more you see that they're tied into the databus on the truck and they're pulling in some data from the engine, the databus, the other systems on the truck to supplement that video that they're capturing. Lou (00:15:28): And are those generally installed by the fleet or are they sometimes OEM? Tim (00:15:34): Yeah, I'd say mostly by the fleet. Lou (00:15:39): Okay. Tim (00:15:39): You can get those as an OEM install, but that's a very personal decision for a fleet. And there's so many systems out there that it would be really hard for an OEM to be able to install all the various systems out there. And then that brings up another challenge for us is that, when it does get installed as an aftermarket system now and they're tapping into the databus, what does that do to all this data on this superhighway of data going across on this truck? Is it going to influence the interchange of data and even collecting the data? We see that where, when they have an ELD, we go out to the field and we try to tie into the OBD port on the truck and you have communication problems, and it may be because of this ELD system that they have on the truck. And you have to find it, unplug it, and then all of a sudden everything communicates well. Matt (00:16:45): It's kind of almost one of... The strength and the weakness of the whole CAN system is that it's open, any system can get on the network and start broadcasting and receiving messages, but it's built on the assumption that everything is going to play nice. And as we're seeing vehicles with more and more aftermarket and third party systems put on that want to have access to the CAN, we're also seeing systems that aren't necessarily playing nice. Not, I think, by intention, but maybe just by oversight, but it is causing issues on the back end when we're in the investigation phase. Lou (00:17:19): Yeah, there's a couple interesting things there. One, you have to have a really firm understanding of the electronic system of the truck. Even before you show up, you have to have this big foundation. And then, two, you have to understand how it might affect the data. Is there any chance that it is kind of overriding the vehicle's EDR? Not overriding it, but overtaking it so there's some more asynchronicity in the data that is being captured by the OEM modules? Does it kind of poke its head in there and change the timing ever? Matt (00:17:50): I think so far we haven't seen that, but it's plausible, and so it's something that I think we collectively in the industry have to stay vigilant about. Are we seeing instances where we suspect there's interference somewhere? And if so, can we really understand it and so that way we can maybe eliminate it? Tim (00:18:09): Well, and speaking about asynchronicity, another difference between heavy trucks and passenger cars is a passenger car is based on the airbag module. Everything's kind of in this one repository, right? We're seeing other systems on passenger cars now that have data that you can pull in various information, say, looking at Techstream or some of these other systems out there, the vehicle dynamics controller, but for the most part, with passenger cars, everything's coming in from a single repository let's say. (00:18:49): And somebody like Rick Ruth might kind of pick apart what I'm saying, and I might be overgeneralizing, but I think that for heavy trucks, what we have when you look at an event data recorder, there's not a single module that's primarily recording the event data. You have multiple systems on this truck, so it could be coming from the engine controller, it could be coming from the brake controller, it could be coming from a vehicle controller, it could be coming from transmission. All these various systems can record data. So now, when you pull that data out, how do you synchronize that data together in your analysis? It becomes a challenge. Matt (00:19:39): Yeah, that brings a good point, that there is... On heavy trucks, the event data recorder functionality is always built into some kind of ECU whose primary job is something else, right? There is no module that's, "This is an EDR." I guess you can kind of say the same thing about an airbag control module, but I would put it in the category of it is a purpose-built EDR. Its first job is controlling the airbags, but its job really is also to record. On heavy trucks, that's just not the case. They're building them into, like Tim said, ABS modules or transmission modules. And so it's never the case of everything is really easy to do or perfectly aligned, because that's not what its main job is. Lou (00:20:26): And then when you guys go to the inspection and you find that the wiring system is jacked and you're not able to just plug into the DLC, then you have to know every module that needs to be pulled so that you can go back and start to implement your specialty, Matt, which is the benchtop stuff. Matt (00:20:45): And that's like the moving target, right? If you're plugged in through the DLC, you might know, based on the truck that you are at, what things you need to get. But now, suddenly you're not in that position, and now you need to know, "Well, okay, of these 15 different pieces of data I want, which ones come from which ECUs? So which ECUs do I need to get?" Because the truck might have 15, 16, something different ECUs on it. You probably don't need all of them, but there's probably some you really do need. And so yeah, trying to maintain this encyclopedia in your mind of where every piece of information lives on all these individual components is almost impossible. (00:21:23): But that's the push, right? That's where I think the line for folks in the reconstruction industry is where it starts to say, "Is this feasible to know everything I need to know about accident reconstruction and also this information?" Right? That's asking a lot. You need a lot of brain power or memory for that, and I don't think that there's a lot of people that want to do that, can do that. You know what I mean? It's a complicated problem. Tim (00:21:50): Yeah, it's nice to have Matt DiSogra on your speed dial, 1-800-Matt-DiSogra. Matt (00:21:55): Right. Because I'm only doing that, right? I'm not having to remember how to do all the other facets of accident reconstruction, then, okay, I have a little bit of extra mental space, only a little, but some to try to remember all this. But even then, this is all I do and I still find it challenging, so I can appreciate why everyone finds that challenging. Lou (00:22:15): Yeah, I was speaking with Anthony Cornetto, Proprietor of HVE at this point, not HVEDR, but HVE the simulation package, and I told him that I was going to be speaking with you guys and he said, "Oh yeah, I got Matt on speed dial. Anytime I have an issue, he's the guy I call." And similar to what I was saying with Rick Ruth on our podcast, is that we're super appreciative to have guys like you in the community because it's impossible to keep that all straight. So we need people that are willing to share and willing to go down that rabbit hole and learn everything that's necessary, so that we can then ping the specialist and know that we're not missing data that's really important to figuring out how a crash happened. People are depending on us to put all the pieces together, so we appreciate that. And it's not as simple as the passenger vehicles. You have Part 563. You have the Bosch tool. There was a notice of proposed rulemaking at one point, right, for heavy trucks? Tim (00:23:13): There has been, yeah, and it comes up from time to time again. Matt (00:23:17): Right, yeah. Tim (00:23:18): And I think right now what we will probably see in the next month or so is a notice for proposed rulemaking for automatic emergency braking. So that, we kind of go back and forth between, say, with the heavy trucks what we're focusing on. And it seems right now, AEB is the focus; and so I think we'll see that coming down the pike. (00:23:42): Now, in Canada, there was a... I wish John Steiner was here to mention it because there was the hockey team collision that happened up in Canada. Lou (00:23:54): Oh, yeah. Tim (00:23:54): I forget where that was, but that got a lot of focus, and so, therefore, that brought the attention back to event data recording and the investigation of that crash. And so the Transport Canada, they invested into a research project and looking at the feasibility of mandating EDR for heavy trucks. And typically, Transport Canada will kind of align with what the US is doing with NHTSA, but this is a area in which I think they are kind of taking a little bit further step. I don't know where they are in the process right now, but we may see that become mandated in Canada. Lou (00:24:36): Yeah, that'd be nice. So we were talking about the Bosch tool a little bit and how something like that would be great. And that reminds me of Jeremy Daley, Synercon, and the FLA. From what I understood, that got bought out by... I can't remember who it was now. Matt (00:24:51): Dearborn. Tim (00:24:51): Dearborn. Lou (00:24:52): Dearborn, that's right. So I haven't seen any progress. Is that still in the mix or is it kind of sitting by the wayside? Tim (00:24:58): No, it's still in the mix. And I think I can remember. I don't remember how many years ago it was when Jeremy raised the idea to me, when he had started work on it, and I said, "Wow, that would be great. Sign me up. If there can be a Bosch-like CDR tool for heavy trucks, I'll be the first one to use it." But I did say that the challenges are immense here for heavy trucks versus, say, the passenger car industry because of some of the reasons we talked about. I mean, there's all these different repositories of information, there's not as much standardization going on with trucks. It's sort of the wild, wild West, if you will. Even just trying to get a communication device that will allow you to interface with a truck, that took some time. (00:25:57): Yeah, so it took the ATA, American Trucking Association, TMC, which is a division within that, the maintenance council, to really get together as an industry and say, "We need a common device that'll allow us to communicate with trucks. Because if I've got a Cummins engine and they're requiring their device versus a Detroit diesel, their device, Caterpillar, you name it, this gets expensive. We can't really keep up with all of it." So they came out with this common standard for a communication device. Well, that helped a lot, but that's just the first step now in communicating with the truck. And then the next step is where do you get the data? What's the Rosetta Stone for translating the data? (00:26:45): And so the challenges in trying to pull something like that together were immense. Now, hats off to Jeremy for doing that, and I think the device is good, but you can't stop, to your point, there. So with the Dearborn Group, it's going. And they're a good group. They actually are a maker of communication devices for heavy trucks. The RP1210 device, you can buy the DPA 5, I think, is their current generation. Matt (00:27:19): Yep, that's right. That's right. Lou (00:27:20): That's what I was going to say, it's the DPA 5, of course, yep. Tim (00:27:22): [inaudible 00:27:22]. Yeah. And one of the foundations for the FLA is really the DPA device. And so it's pulling a lot of just general information in first, and then it'll go out and it'll request some of the event data. And it can do that as a standalone, and you can also use it as a passthrough device. So if you want to use it as a passthrough device, it's no different than any other data link that you're using. It's when it becomes a standalone device and you say, "Okay, can it pull the data from the latest Detroit diesel engine?" Lou (00:28:04): Right. Tim (00:28:04): And I think the answer currently is, no, it cannot. Matt (00:28:07): It cannot. Tim (00:28:11): We probably need to have some of the guys from forensic training groups speak to that, but they have a chart out there that says what it supports, what it doesn't in terms of a standalone device. Lou (00:28:22): Right. Tim (00:28:22): And as technology has marched on, I think the abilities of the device has remained static. Lou (00:28:32): Yeah, they ultimately need to get a team on it that's excited about it and continually working with it, and probably partner up with guys like you to just help them understand when changes are coming or have come. And I imagine that you're learning a lot about a lot of those changes out in the field when you go to actually download a 2023 truck and you're like, "Wait a second, what is this?" Tim (00:28:56): Yeah. Now, having said that, even though it's remained static, it's a good tool to have in your toolkit. There's no doubt about that. To have it be your only tool in your toolkit could be a problem, depending on the work you're doing. So for example, if you have a Caterpillar engine that experienced a power interruption, it is the tool that we're going to pull out of our toolkit to try to get that data, because you won't be able to get that data with the manufacturer software. Manufacturer software sees a incomplete file and it says, "I can't retrieve it." Whereas, the FLA will retrieve that incomplete file from the power interruption. Matt (00:29:40): Right, right. So it has utility for sure, but very much I think in the context is, for the way we use it, is a supplemental tool to OEM software, specifically because the challenge is staying as close to the cutting edge as you feasibly can. And to date, really, the only way to do that is with the OEM software. Lou (00:30:01): Yeah, I remember a couple of times where I would download with the F... Matt (00:30:03): The OEM software. Lou (00:30:03): Yeah, I remember a couple of times where I would download with the FLA and use it as a pass through and then find out that I wasn't pinging a couple modules that I could get with the NEXIQ tool. So something to be aware. And if you have an eye up for it, then great, but if it's not something you're looking out for, then you might walk away without all the data, which is a nightmare. And one of the things that it seems you have to do to prep, if I'm going out to download the ACM of a Toyota Camry, it's like, okay, it's a Toyota Camry. I'll look up in the CDR tool and good to go. But if it's 2020 Freightliner, the most important thing from what I understand is well, what's under the hood? What is the engine? A, is that true? And then B, what are the engines that you have to be aware of and have the appropriate tools for at this point? Matt (00:30:51): Maybe I can answer the first part of the question. So I'll give a slightly different example, but in the same vein. Let's say you get a call and they say, hey, we have a 2020 Mac. So you say, okay, well it's a Mac, so it'll have a Mac event data recorder on it in the vehicle ECU, which is in the cab. So you say, all right, great. If you do nothing else and just show up thinking and then that'll be what I need. And then you look under the hood and it happens to be bright red and it's a Cummins engine, so congratulations. You have a second data recorder now in that Cummins engine. (00:31:22): So then you say, okay, well what kind of ABS do I have? And it has Bendix ABS. So we've got BDR, Bendix data recorder, there's an event data recorder, a third one now in the ABS. And then you look around the cab and there's a camera in the windshield and a radar on the front bumper. You say, ah, this has Wingman, so there's more data in that system and there could be videos in the camera. And there's also a controller for the camera called an SDP, which can also have videos. So suddenly what looked at the surface as, okay, I get what I have, I've got a 2020 max straightforward download, you show up and your scope of work has ballooned tremendously. (00:32:03): What is the most important thing is a difficult question to answer. And the reality is there's so many facets of what is equipped on a truck that become important. The answer is kind of all of them. And some of those things you can know ahead of time. For example, with the VIN, VINs can decode and tell you what kind of engine are in the vehicle. But you can't decode a VIN to tell you if it's got a Wingman camera on it. That's not data stored in a VIN from a generic VIN decoder. So you can look at possibly getting build sheets. But you know how the timescale we work on a lot of times is you don't have days to obtain data about a truck. You're going out there the same day. Now again, there's the pressure on you to triage this vehicle sitting in front of you for where are all of my sources of data just to even know a roadmap of how to get there? Tim (00:32:53): I kind of laugh when I go out and download a passenger car, which I probably download maybe five to 10 a year at the most. Heavy trucks, I might download five heavy trucks in a week. Lou (00:33:07): Yeah. Tim (00:33:08): And so when I go out and I pull out my CDR equipment, I always just pray, I hope it works. I hope it works, right? Because I have trouble. But then with heavy trucks, I do it so often, I usually don't have trouble. But that's because I'm doing it every day. So for most people, it's the other way around. CDRs, easy. You hit the button, boom, pulls in the data. With heavy trucks, what Matt was saying, you have to know what you're searching for. What are you asking the truck to give you back? Do I have the right software? Do I even know what software I need? And that's the big challenge is just getting the data on the heavy truck. And communication problems, I mean, if you want them, start getting into heavy truck downloads, because you'll get them. (00:34:04): Again, that's where it's really helpful to have Matt on speed dial, is to help you troubleshoot some of that. But really even that's a lot of what we talk about in the class with SAE is how do you triage the problems that you're having? How do you triage that question of what data am I looking for? How do I triage communication issues, things like that. (00:34:28): And we always tell people in the class, hey, grab a dozen donuts, go to your local truck dealer, walk into the sales department and say, hey, can I have some keys? And go out and play with trucks. And they're usually happy to do it. And it just takes practice. And the thing is, with passenger cars, I feel like the challenge in EDR is really analyzing the data. There's a lot there with more and more modern vehicles, the data can get pretty complex. And what's the acronym that Rick Ruth- Lou (00:35:06): R-T-F-D-L, I believe. Tim (00:35:08): ... R-T-F-D-L. I mean that is the key to going through that. And really going back to those classes, it becomes really important. Whereas heavy trucks, I don't want to downplay the analysis on heavy trucks, but I'd say there's a difference. Heavy trucks, it's hard to get the data, maybe a little bit more straightforward in analyzing the data, whereas with passenger vehicles, that's flip flop. Lou (00:35:39): And the DL there in the R-T-F-D-L would be data limitations. And we get a lot of guidance, fortunately, on the CDR side of things. You guys are making my heart rate elevate just listening to the idea of showing up to this truck where you're not going to know exactly what's on there, what you need to be prepared. Especially if you're traveling somewhere, you're flying, you have to keep a light kit. You show up and you find out there's a module on there that you didn't expect, so then you have to let the client know, hey, it's going to be a few more K, because I've got to go back to the office and come back out. So one, I'd love to hear you just talk about that. And then two, what are processes and procedures that you would recommend to recons who are faced with that? What can they do to best prepare for what might be there? Matt (00:36:24): Some days half that battle or a large part of that battle probably is, is it me or is it the truck? Right? Do I have some software problem? Am I out of date? Do I have a driver issue? Is the reason I can't get what I'm trying to get because I haven't fired up this piece of software in four months? Or do I have a perfectly good working copy of the software and is there some issue on the truck? Like you said, a wiring issue, a fuse is blown to that specific module. There's some firmware problem with that module that no one knows about except the manufacturer, which prevents it from talking to you. And so you're sitting there, your client's staring at you, staring at a laptop and nothing's working. Tim (00:37:04): That's the worst. Matt (00:37:06): And you're trying to figure out, can I blame the truck and say, listen, I got everything right. It's the truck's fault. Or are you the idiot who then has to say, well, yeah, I have to come back and write off all this time because it's my fault. And that's sort of a universal feeling. The only way to maybe get close to solving that problem is just by doing a lot of downloads so you're using the software regularly. And that's hard to do when you figure there's, whatever, 20 some pieces of different software you need to collectively handle all of heavy truck downloads. What's the likelihood that you're firing up all 20 every two weeks and using them. Low. So it's almost a guarantee you're going into downloads, being rusty. That's just the nature of heavy truck work. I don't think you can avoid that. What to do about it is a whole other question. I don't know, Tim, do you have ideas? Tim (00:38:03): Well, the thing that's coming to mind of me is that with a heavy truck, you've got all this power sitting there with the batteries. So after a big crash, what the tow truck driver do? Cut those battery cables. So one of the things that everybody needs to be prepared for is how do I reestablish power? As much as I've been doing this, I'm amazed at how forgiving it can be sometimes when you've got a pretty damaged truck and you're able to reestablish power and everything just communicates well. It's surprising sometimes how that'll work out. (00:38:50): But then you can go to a truck that has very little damage at all and you can't communicate with anything. And somebody's jacked the wiring back with the Deutsche connector or something like that. And you have to start digging things out of the dash and truly really try to understand it. And even then, sometimes you may never figure out the problem, and then you have to start going to direct module. And some of it comes down to how much time do I want to spend troubleshooting versus just going straight to the module. (00:39:29): So knowing some of that helps, having the experience helps. But really this goes back to again, as these vehicles have become more and more complex... It used to be when we first started teaching the class that the focus was on the engine module. I got the ECM, which meant the module off the engine. Lou (00:39:55): Done. Matt (00:39:55): Yeah, you're done. Tim (00:39:55): And as we've talked about, that's really just one component of many. And oftentimes maybe the component now that has less data than some of the other components on the truck. So I think it really is underscoring the challenge for us as investigators, back to our discussion earlier, just how do you keep up with it? Are you ready for it? Are you attending training? Do you have a plan in place? And for the companies that are larger, you can do that. You can have a Matt DiSogra in your group that specializes in it, shares the information throughout the company. But if you're a small group or a solo practitioner, the challenge is immense. Lou (00:40:45): Yeah. And that's fortunately Matt DiSogra for hire, right? Tim (00:40:50): Yes. Matt (00:40:51): That's it. I'll give out my cell phone number for $20 coupons to everybody. Tim (00:40:58): And it's the same way for me. I can reconstruct a general motorcycle crash, but if anything, if there are issues that become more of a focus on the motorcycle, I'm not a specialist in motorcycles. That's when I call up Lou Peck or I refer Lou Peck to the client, is a better way to do it. Matt (00:41:23): I think it comes back to that, it's a volume thing. Because say you're sitting in a truck and getting some particular error, and that's the first time you've ever seen this error, but that might be an error that we sort of have seen, right? It's infrequent, but we've at least seen it. And then there's some arcane workaround for it. Take this module and unplug it twice and plug it back in and it'll work. Which sounds stupid and actually is the solution, you'd never really intuit from that. (00:41:50): And then that's just it. If you're not doing a high volume, the challenges are even greater. It's not that you can't do it, but you're going to run into scenarios like that where there might be a solution, but it's difficult to know. And so yeah, if you're at a company like Tim said, you at least have that advantage. You maybe have a couple of colleagues you could call up, but I think in some ways we might consider ourselves competitors in some veins. But also we're all colleagues in this sort of quest to try to understand all of this heavy truck data. (00:42:22): And so fortunately, a lot of people are very willing to share experiences and feedback from the field and war stories. And that's helpful for everyone because that's sometimes how we learn about problems or realize that, oh, hey, that thing three other people have now seen that. And then maybe we figure out how to work around that. And if we all just keep the information to ourselves, then we're done, we're toast. Lou (00:42:48): I think that's a huge point that you made there. And I think that it's one of potentially the unique things about our industry. And there's people like you guys, and I try to teach as much as I know, I don't hold anything back. I teach everybody anything they want to know. And if they want to be a specialist in that, great. But the willingness to share and help everybody out, no matter. Somebody asked me the other day, who's your competition? I was like, I really don't see it that way. I don't see that we have any competition. There's first of all more work than anybody can handle. (00:43:21): And second of all, I'm willing to help anybody out at any point. And ultimately, I hope that that's something that the whole community takes away from this conversation is that there is that willingness to help and you need to have the appropriate contacts and be willing to help if you're on the receiving end of that inquiry. (00:43:44): So I want to talk a little bit about some of the practicalities. We mentioned already some of the things that trigger an event, hard brake. We have fault codes like you were saying, Matt, and then we have last stop. And last stop seems like potentially the best one at times, but also the most volatile given that it's the last stop. So when is that overwritten and is there any advice we can give to fleet manager or accident investigators to help retain that data? Tim (00:44:17): That's a challenge is the last stop data. It's volatile. And oftentimes it will get overwritten. So one of the things that's in the SAE, not a standard or recommended practice, J2728, is actually two stop events. So that hopefully you have that second to the last stop event. International adopted that approach. Matt (00:44:46): Yes. Tim (00:44:48): It may not always work the way they intended it to work, but that's the plan was they had two stop events in there. Lou (00:44:56): Yeah. So they can drive to a rest stop or back to their facility and then park the truck and hopefully that data still remains. Matt (00:45:04): Yeah. I mean, not to use the answer, but just for the sake of you. If it's only two last stops, it's still almost, you can't really drive it. The double last stop maybe gets you to that, hey, just get it off the travel lane and move it into the shoulder. It covers you on that one. But even driving it to a rest stop, he has to come to the off-ramp. He stops. He has to come turn in, he stops, then he has to back in. Another stop. So you're already three or four in your overrides. Tim (00:45:31): Think about the garbage packer. It's July. It's August. You have a full truck of garbage in the back of the body of that truck. You can't leave that truck sitting there for very long. Lou (00:45:46): Yeah. Tim (00:45:46): If you've ever gone to a truck that's been sitting down in Florida in July for more than a week or so, I challenge you to crawl under that truck. Matt (00:45:58): Yeah, the hopper juice. Well, now as inconvenient as hopper juice dripping on you is, now these guys are taking the approach of, it's a safety issue. Because if there are lithium batteries in the hopper, they want to get those hoppers emptied because that's how they're having truck fires now, is people have improperly disposed of full lithium batteries. And now the truck's out of service because it's been an incident and it sits there and they come out the next morning and the whole truck's burned down. So they get very motivated to try to want to get the hopper emptied. Tim (00:46:29): Yeah. So it's an OSHA issue, right? Matt (00:46:33): Yeah. Tim (00:46:33): So they need to do it. So there are challenges now if you're going to crank that engine up as to how to do it, make sure you do it so that you are not overriding data. Especially now when you tow a truck, typically they're going to disconnect the drive shaft. So you crank that engine up now and because the speed data is really coming off the tail shaft of the transmission, you have an automatic transmission, that tail shaft is going to start spinning because you have a viscous clutch. And so now as far as a truck thinks it's moving down the road. (00:47:16): Or let's say you have a manual transmission, somebody cranks the truck up, the drive shaft's disconnected, you could crank that truck up easily with it in gear if the drive shaft is disconnected. Now it starts spinning. It's seeing speed supposedly, and you've now overwritten data. I mean, if you've done heavy truck work, you've probably encountered situations where you've downloaded a truck and you see some event data from a date after the crash, but yet that truck's sitting there on blocks and nobody was driving that truck around the yard. That's probably the reason. And so it is a challenge. (00:48:02): I can remember, going back to the garbage packers, crawling under some trucks down in Florida, in the dog days of summer, even just a week after the crash. And fortunately I was in a shop and I take a creeper and literally just crunching over all the maggots that were flowing out of that thing. And you think I'm one inch above this whole floor's covered with maggots. It's fun. And then you go home and tell your kids about how glamorous your job is. Matt (00:48:39): Oh, yeah. Best job. Yeah. Lou (00:48:41): My kid went to school a couple days ago for dress up like what you want to be, and he chose forensic engineer. It's like, I'm not sure you get the... He just sees me crashing motorcycles and things like that. And I have a pinging pong table in my shop, so that's probably... He doesn't see the rolling over the maggots. Tim (00:48:58): How old old's your son? Lou (00:49:00): 10. Tim (00:49:01): 10? Okay. Yeah. I can remember when my son was, I think he was about three years old, so he's still a toddler. He can speak. The only thing he knew about my work was when we would have these fun parties like fall bonfire and we'd have marshmallows roasting and hay rides and everything. And one day I'm going to work, in the morning he comes down and he says, daddy, are you going to work? I said, yeah. He goes, are you going to roast marshmallows today? I'm like, yeah son. Lou (00:49:36): Something like that. Tim (00:49:38): I've got the coolest job in the world. Lou (00:49:39): No, I am going to be creeping over roasting maggots today. Tim (00:49:43): Yeah. Lou (00:49:45): The other thing that's interesting to me, I'm a layman, so I hear about fault code setting events and then the associated snapshot data. Is that ever something that can be practically implemented and integrated into the accident reconstruction? Is that data useful in that front? Matt (00:50:05): Definitely, yes. With some caveats though, especially I think the biggest one, would be that now you're really into a system that is wholly not designed for us. Tim (00:50:17): Right. Matt (00:50:17): That's a system designed and implemented for technicians to diagnose and fix trucks first and foremost. And then we are now taking it and trying to kind of shoehorn it into our mold of, well, can we use it for our purposes? And the answer a lot of times is, yes, there can be useful data in there, but it's a matter of understanding the limitations of that system and knowing how to properly apply it with the relevant, let's say, uncertainties that can come with that sort of data. But yeah, sometimes that can provide key evidence in cases also. Tim (00:50:49): Yeah, we talk about in the SAE class, some examples of where you can use snapshot data. And that's really been the only data that we had after a crash to be able to determine a speed and impact. But the other thing is think about other situations like let's say a truck that was having mechanical problems with the engine. They didn't get the truck fully out of the roadway because the engine shut down. (00:51:17): Now you're really going to have to go and do a deep dive into some of these fault codes, look and see what was going on, why that truck shut down, when the operator was getting the shutdown message, how many times. Did the operator hit the engine shut down override switch on the dash 10 times to try to get down to that next exit where there's a Pilot station and they can have a cup of coffee when they call the guy to come look at the truck, and maybe it finally shut down on them and they weren't able to get off the roadway. So there's definitely times when fault codes become important for the analysis. (00:52:01): The other thing is we talk about in the class tire marks. You look at heavy truck tire marks and you've got a solid set of dual tire marks. Did they come from the trailer? Did they come from the tractor? And one of the things you can look at is, for example, ABS fault codes. So it's often overlooked as downloading the trailer. And let's say the trailer, you have a wheel speed sensor fault, and because it was not getting good wheel speed sensor data, now the ABS disabled itself on the trailer and it reverts to non ABS, regular foundation brakes. And so therefore those brakes lock up on the trailer, for example, and you get these solid tire marks. Well, if you had done a calculation, let's say you didn't have event data that told you how fast the truck was going, a little bit older truck, and you did your calculations based on the assumption that those dual tire marks were from the tractor versus the trailer. That's going to make a difference in your calculations. Matt (00:53:12): Yeah. The other area that I think that fault codes are evolving into is not just locally stored fault codes, but a lot of OEMs now are implementing systems where vehicle faults can be uploaded to the cloud or to a remote server. So fleets can have access in real time to the health of their trucks. They're selling this as an add-on service, but the functionality and capability is there. So that can open the door to even wider fault code analysis, where now you might have a sort of time history with GPS coordinates, vehicle speeds and all sorts of stuff plotted on a map in a web portal showing where the truck was and these fault codes developing. Not stuff that is available through a DLC download, something that you would have access to through the fleet's portal if they have that. But that world of fault code analysis is not just a static one either. That's evolving into more and more cloud-based and remote data. Tim (00:54:12): Another example would be stability control. Stability control, collision mitigation systems. These depend upon a lot of things that are in good operating condition. And so looking to see why a system may not have responded the way you expected it to respond is going to come down to where all systems go at the time of the incident. That's one thing. So did you have a radar fault? Did you have a camera that maybe was out of alignment? Did you have a wheel speed sensor that was not giving correct information? Was there a pressure modulator valve that may have had an electrical issue, so therefore it's not cycling the way it's supposed to cycle? (00:55:09): All of these things become really important. Now in looking at that overall crash. And again, back to it's no longer good enough for us to fall back on our traditional reconstruction techniques. We have to be experts in understanding these systems if we're going to do a proper reconstruction, in some cases. I don't want to say this is important in every case, but certainly if you look at a truck rollover, you're going to want to know was the stability control system functioning the way it should have if it had stability control? Matt (00:55:48): That can be a tough analysis, right? Where you now, as the expert, are expected to be able to answer questions like, if this particular fault is active, then what mode and state would the system have been in? And what capability is lost because of that? Or what capability does it still have? You've got an active fault for a radar partially blocked on a truck with a collision mitigation system. Well, how does that affect that system? Is it still operable? Is it going to respond, is it not? And where do you get all the answers to those questions? It can become a rabbit hole. Tim (00:56:24): Yeah. And if the fault code is radar blind, what does that mean? Matt (00:56:29): What does that mean, yeah? Tim (00:56:30): Does that mean they were operating in the... What are the stretches out in I-10 going through the desert that you don't see any metal for a long time. That may throw a radar blind fault and have nothing to do with the radar not being aligned properly or some kind of problem with the radar. It's just not seeing anything. It's seeing empty space out there. Lou (00:56:56): And so do you find that you have to do testing at times to figure out how something like that might affect the overall operation of the truck? Matt (00:57:04): Sometimes. That's the only way to answer some questions, right? You're not going to find detailed performance explanations in the diagnostic trouble code manuals for these. Those are geared towards the technicians, again, who are fixing the trucks. But it might not give you a comprehensive answer of what's going to happen to the... How is the system going to perform with this code? And so does the manufacturer know? Yes. Are they going to tell you? Sometimes, maybe not. And so the only way would be to try to perform some representative testing and understand that yourselves. And yeah, we've definitely done that. Tim (00:57:43): Boy, this gets into the discussion about ADAS, driver assist systems, and really getting to level four autonomy. And in higher levels of autonomy is are we really ready for it? I mean, is our whole infrastructure ready for it? The maintainers inspecting the truck. And I don't know that we have time to really get into all that, but at this point in time, say even just A DOT roadside inspection, roadside inspectors for DOT, they're supposed to look and see if the ABS light is functional. But really that's the only thing they're looking at. (00:58:31): So for a truck with collision mitigation, they are not looking at all the other things on that truck to see if it's functional. And so now you pull into the mix a level four truck. Are we ready for that to be able to say all systems are a go? So there's a reason why we don't have self-driving trucks today. The technology is there, but maybe as a practical way of getting it on the road, there's still a lot of limitations. Lou (00:59:05): Yeah, from doing a little research in preparation for this conversation. I saw a couple interesting things, Waymo with their Via. And if you believe their website, they're basically there. But like you said, Tim, there's all sorts of other issues that come along with that. And then was it Detroit Assurance that indicated they have a level two autonomous truck where the driver can kind of just... I mean still has to pay attention, but can take their hands off the wheel for a bit. Tim (00:59:30): Yeah. I was out in Portland not too long ago and got to see, I think they called the truck Maryanne, if I remember right. And one of their self-driving trucks was sitting there right in front of Daimler's headquarters. And I looked at it, on the back of this truck, they had these double aught, these big battery cables coming out of these boxes on the back of the truck. And I thought, oh my gosh, is this an electric truck? And no, it wasn't an electric truck, it was just... (01:00:05): I think it was the Level 4 autonomous truck that they were testing. And it is so power hungry for all these systems that are working on the truck because now you have to have, say for steering, they're going to use electric power steering, and that requires a lot of power. And so as I looked at that, I thought, well, am I ready for that when I go out and investigate a crash to even know how to be safe around a truck like that when you've got that much amperage that could be coming off of that truck. And then you bring in these electric trucks, and fortunately, we haven't had to deal with one of those yet, but I'm going to start putting into my plan for training just safety around some of these vehicles. Lou (01:01:06): Yeah, and the Tesla truck, from what I understand, was delivered three months ago to Pepsi and Frito-Lay. So I suspect that you guys are going to be called out to some of those before long. Matt (01:01:16): Right. And then there's all the normal questions that arise that now we basically have no answers for at the moment. What data is there? How do you get it? What modules aren't on? And maybe you have a truck that's intact, maybe you have a truck that's in 17 pieces. And yeah, all of those questions will be there and we have to figure that out because the manufacturers don't have some published guide that says, "Hey, crash investigators, here's how to make your job easy and all the things you need to get, and where we keep all of our data," and all that. (01:01:49): When they sit down with the objectives for designing that truck, the electric truck or the autonomous truck, of course the first thing they do is say, we want to make s ure we have all the event data accounted for in this truck, and then they plan for everything else. And I say that kind of jokingly, but as you think about autonomous vehicles, whether it's heavy trucks or passenger cars, who's the driver of the truck? The driver of the truck has a story to tell and if the vehicle is now the driver, there really needs to be some data that's recorded. And I don't know that we're ready for that. A lot of influencers for that. Clearly, government is an influencer from a regulatory standpoint and the administration that we have, say here in the US, because that can change every four years. We can go back and forth from, say, being regulatory heavy to regulatory light, and that's going to be a powerful influencer. Industry groups will be a powerful influencer moving forward. So American Trucking Association, TMC, that's a group that I think will influence the processor. (01:03:12): Insurance, will influence the process. You think about autonomous vehicles and questions that come up now would be, who's liable for the crash? You have a 80,000 pound truck going 65 miles down the roadway, and you have something like 15 million jewels of energy to deal with versus a passenger car, the equivalent is probably 5% of that kinetic energy for a 4,000 pound passenger car. So it's a big challenge. So when you have the first crash with an autonomous truck, it's going to be newsworthy. Tesla crashes have been newsworthy. Just imagine that first crash with a self-driving truck. Now, you think about platooning, that may end up being something that we see more prevalent before we actually see self-driving trucks is platooning. There's a lot of drivers for that because for fuel efficiency, things like that. But now you have a platoon of trucks that crash that's going to be newsworthy. Right? Lou (01:04:31): Yeah. Oh yeah. Tim (01:04:32): For the other thing, for the insurance standpoint, as you think about not only who's liable for the crash, but how do you assess risk for underwriting? How do you put a dollar figure when a fleet now is invested heavily in autonomous vehicles? It does have a lot of promise for reducing crashes, there's no doubt about that, but how do you assess that risk? So these are things that it's coming, but a lot of questions that surround it. So I wish I had a crystal ball and I could peer into it and make a lot of money putting my money onto the right guess, but right now it's a guess. And when is it coming? Lou (01:05:20): Yeah, over the decades, well, probably a decade or so, seems like anybody who's tried to predict the advent of full autonomy has been very far off. We all predicted it earlier, or some of the futurists predicted it earlier, but it's a more difficult challenge than people were expecting. And one of the things, like you're talking about, Tim, just the involvement of the manufacturers, they are putting something on the line there. And I applaud them for that because their goal is to make driving safer, but as they do that and insert themselves more into the driving task, they are putting themselves out there for product liability cases. So product liability, in my estimation, is going to skyrocket over the next 10, 20 years and hopefully crashes go down as a result, but they're going to take a lot of flack. (01:06:12): People like Tesla have already publicly announced that they are gearing up for battle because they think it's a fight worth fighting and they're going to continue to try to make the cars safer. But they know there are going to be problems along the way and they're going to fight the product liability cases that come their way, but it's a tough situation for them. Tim (01:06:30): Yeah, I think you're right on that. If there is a crash, who's responsible? Is it the motor carrier, is it the manufacturer? Clearly, you can make arguments for one or the other or both. The motor carrier has to maintain the vehicle, back to what we were talking about before. So the court system's going to have their hands full with that. Lou (01:06:59): Yeah. So one thing I wanted to get back to, you're talking about the maintenance, and then you guys were talking about the radar systems and how sometimes they might act like they're checked out, but that's just because potentially they haven't seen any metal for a long time. And the active safety systems, there seems to be just a tremendous amount of progress there. And new machines, new contraptions, new logic, new things we have to look out for, new methods of accessing data, new methods of interpreting data that seems to have accelerated things quite substantially. And that is when I started to raise an eyebrow and just be like, this is probably not a good idea for me as somebody dabbling in heavy vehicle event data recorders to go out and perform a download anymore just because I might miss some of this data. And I'm not serving the case if I do that and pretend that I can be you guys. So wanted to the first start with Bendix, what do they have out there and what tools do we need to get some of that data. Tim (01:08:05): With the Bendix, there can be data in the ABS controller. So you have the EC 60, you have the EC 80, clearly, you're going to need the Bendix software, you'll need a data link to be able to read the data off of it. And then within that, there's what Bendix calls BDR data, Bendix Data Recorder. Matt mentioned earlier, Cummins and Bendix, they have a disclaimer for that data for crash reconstruction purposes, but that's something that you may want to have for your analysis in a crash. There's a paper that we have coming out in April through SAE on BDR data, really taking a look at the reliability of that data through some controlled testing that was done, comparing it to instrumented data that we capture, as well as other data on the databus. So that should help in the analysis. But to get that BDR data, you can read the data with the Bendix software, but it doesn't translate the data. So you have to send off to Bendix, ask them to translate the data, and then you get it back. (01:09:23): Some other information that you can get out of Bendix systems would be the Safety Direct, which Matt had mentioned, SDP. And Safety Direct is more of a dedicated event data recorder, that's specifically what they designed it for, so it's gathering a lot of really good data. SDP is a component of what Bendix calls their Wingman Fusion system. And so most trucks with Wingman Fusion will have SDP data, and it's something that the fleet can subscribe to, kind of like OnStar, for example, where you can get information in a portal, for example. Another source of data that you can have with a Bendix is the Safety Direct processor, which is more of a dedicated event data recorder, which is part of the Wingman Fusion system. It's a component within that system. And really what Bendix designed that for is to be a subscription based service where a fleet can subscribe now, get information in near real time for looking at the operation of that truck. (01:10:41): And then if there is a crash, for example, or there's some incident out there that they want to take a closer look at through the web-based portal, it will flag certain type of events. So let's say a driver has a hard brake event, or a collision mitigation event, or something like that, or a crash, they can look at that in the portal and try to understand it and perhaps use that for driver training purposes or use that in an investigation of a crash. If they are not subscribed to the service, it will actually record data in the processor and then that processor would have to be taken off the truck, sent to Bendix, the data extracted, and then sent back to you. But that can record video as well as event data. So It's a really good source of information for crash investigation purposes. The other thing is just the camera up on the windshield itself for the Wingman Fusion, that can store video in it as well. So there's a lot of different sources of information with the Bendix system. Lou (01:12:02): So it sounds like the process for you when you're going out to interrogate this vehicle might depend on what service the fleet has subscribed to. So how do you handle that? Matt (01:12:14): Yeah, definitely. If if it's a case where you're working for the fleet, then that's a conversation you need to have about what services are they subscribed to. They may already have some of that information. They may already have a video and you might not have to, or they might have no clue that their truck was able to record videos. So in cases where you're working for the fleet, that's a joint effort there to make sure that you're comprehensively covering all the sources of data. And in cases where it's not your truck, that can become more difficult, but you have to do your best to have those open conversations about, okay, well what is there or what might not be there, here's a proposal on how we should try to cover all the bases of what data could be there. Lou (01:13:05): It just sounds like your guys' process must be ever evolving. And every couple of weeks, it's like, all right, this just came out. How do we handle this? What kind of preparation can we do before we fly out to the middle of Texas to interrogate this truck? It's a lot. Tim (01:13:21): Yeah, I tell you, you probably see this too, Lou, the benefit of actually teaching is that it forces you to keep up with it. So every time we teach the class, twice a yearly through SAE, we really are working hard at trying to keep the materials updated. And the byproduct of that is really keeping us up to date and our daily work as well. So not only do we enjoy teaching the class, back to your point of sharing within our community, we like to get to know people out there that come to the classes. And when we show up to an inspection, now there's not all this posturing when you know somebody. But the other thing is it forces us to keep sharp. And a lot of people that come to the classes, they can teach the class themselves. (01:14:16): I remember the first time we taught the class, John Steiner and I we're sitting in front of a room full of 25 people and we felt like we were talking about physics to Einstein and Newton. Those were the people in our class, they were the people that were really doing heavy truck crashes. And John and I get up and we say, we're here to talk about just our collective experience, but you guys have a lot of experience and we're expecting to learn from you as well, and so that's been a great thing. And now 12, 13 years into teaching this class, I still enjoy going to that class, teaching, and I'm always learning something new through that. Matt (01:15:06): And we kind of joked earlier about the key to success is call my cell phone. And to be fair, that's one of the things I like the most is those calls because that's a lot of times where we hear about something for the first time. Because I might not be the first guy at a 2023 Freightliner, somebody else is probably going to be that very first one. And so when I get the call, "Hey, I'm out at the truck." I might not have the answers, but that's the first time we're hearing about something new. And so then maybe I get subsequent calls and then we start to build our understanding of a new system that's out that no one's ever seen before, and the manufacturer just quietly implemented and didn't really tell anybody. (01:15:46): Because it's not something anyone outside of our industry might particularly care about, but it makes a big deal for us that, "Hey, they changed ECMs and it's completely new now." And like, "Hey, what? Send me a picture of that." And then we have to figure out what to do. So yeah, like what Tim is saying, and to the point earlier, it's a team sport in a lot of ways. Lou (01:16:13): And it's perfect for you guys who are out there teaching, and I experienced something similar, you become the point man, the contact person when something weird happens. And it's perfect that you're that contact person because now you're going to go disseminate that information to everybody twice a year. So then it propagates from there. Matt (01:16:30): It does, yeah. Lou (01:16:32): It's really valuable. Matt (01:16:33): Some of the questions that get called into us get turned into slides in the SAE course. Absolutely. Yeah. Lou (01:16:41): It's same thing happens here. I was teaching a class in Chicago a couple months back and student in there said, "Hey, do you know if Zero motorcycles have any data?" And I was like, "I do not. I've never seen any." And then the next day he comes to class and said, "I called Zero and they told me, yep, bring the motorcycle in and we'll be able to download. It's got event data on there." I said, "Perfect." Matt (01:16:58): Oh wow. Yeah. Lou (01:16:59): Now I can spread that info to anybody. So it definitely takes, like Rick says, it takes a village, Rick Ruth. So WABCO, they have the Onguard system in ProView. Onguard being collision mitigation from what I know. And then ProView, being lane departure. Can you interrogate all of that with the WABCO tool? What is that called again, WABCO Toolbox or something generic? Matt (01:17:28): That's right, Toolbox. And in the case of WABCO, it is less complex than Bendix by contrast, just as far as accessing the data. Basically, all of it you can access with the publicly available software that they release, and there isn't a need to send certain certain ECUs to Wabco for download. Lou (01:17:51): That's nice. Tim (01:17:52): Yeah, one of the things that we talk about is with the WABCO, with the Onguard system, the data, where's it stored, it's up in the radar sensor right up on the front bumper. Lou (01:18:04): Oh geez. Tim (01:18:05): Which is a great place to store it when it's the first thing at the crash. Right? Matt (01:18:10): It makes sense because that's what's generating the data, so we should just leave it there. Tim (01:18:13): Exactly. Matt (01:18:13): Yeah, it's logical. But yeah, oftentimes that is the very thing that's destroyed. Which they're the only system that works that way, but that continues to be the case. Tim (01:18:24): So you've had a number of those where you've actually had to bring it into your lab and go to chip level data recovery with those? Matt (01:18:32): That's right. Fortunately, there's always, at least most of the time, there's a method available even when you have damaged ECUs. And so the radar sensors are common ones that we deal with because yep, they're destroyed, but there could still be data on the chips as long as the chips are still there. And so that's how we end up having to go about that. Lou (01:18:51): You got to be young to be doing that stuff because you need the eyes for it. Matt (01:18:56): That's it. Lou (01:18:57): I got the readers already, so I'm not sure I'm doing a chip level anything. Tim (01:19:02): Well, I look at Matt and he's looking through his microscope and soldering like this and just going down, and I'm like, man, I couldn't even take two wires, big wires and solder them together very well. And you've got this chip with 48 pins coming off of it and you're just going down it. Matt (01:19:22): I would love to take all the credit for that skill, but there's something magical about 40x magnification that makes any task really easy, even if it's tiny. But yeah. Tim (01:19:32): Steady hands too. Lou (01:19:33): It's a challenge. Right? Matt (01:19:34): Yeah, no coffee on those days for sure. Forget it. Lou (01:19:37): Yeah, you're a surgeon and your background's mechanical engineering, so I suspect that's not something you predicted in your future. Matt (01:19:43): It's not. No, I think we're trained to hit most problems with a hammer. So delicate soldering work under a microscope is definitely not, I think, a typical mechanical engineer activity, but somehow I've managed to figure it out with a whole lot of practice. Lou (01:19:57): And I suspect, in speaking with a lot of reconstructionists, it's probably something that you didn't plan on, but once you started seeing the need for it and you had a certain passion for it, it just developed from there. Matt (01:20:11): Yeah, it's kind of a fun problem. I think one of the things most people like about this industry is a lot of days are not the same. You're always into something different, and that's just an extension of that same philosophy. A damaged ECU is always going to be a little bit different, a little bit of a different challenge. So I find the problem interesting, doing that kind of work. And as time has progressed, vehicles are getting more data, data from more places, and we as investigators are coming to rely on that even more. When was the last time we had a crash with no data from any vehicle? It's pretty uncommon at this point. So the techniques have evolved to really rely on the data, which ultimately means that we want it more. And so in cases before where we've got to damage ACM, but it's fine, we'll still do it without it, no big deal. (01:21:08): We're seeing a shift in the philosophy and people really, they want that data, even if it's damaged we want to go through the effort because we really need what's on there or something from a heavy truck. So we see the demand growing for that, and I think that's just going to continue as we all become more interested, or reliant, or maybe sometimes the data is the only way to know. If we're investigating the performance of one of these collision mitigation systems, that's it, the data is what we need to understand that. So we need to have the techniques available to get that data no matter what condition the ECU is in. Lou (01:21:44): Yeah, I feel like that's where we're heading. We're heading towards more and more sources of electronic data, more reliance on it. And like you said earlier, Tim, I think you mentioned you can do a recon without it, but it will never be as good as a recon with it. And that's what I'd tell my clients is my best recon blind data-wise is not as good as your average CDR download. If I get that CDR download, it's a lot more information. I'll never be able to tell you what the driver is doing five seconds pre-impact via my science unless you have video of it or something like that. So that's huge. Tim (01:22:18): Yeah. You've seen this shift now in reconstruction where it used to be about how fast was that vehicle going at the time of the crash, and that's where we spent the bulk of our effort trying to do the analysis and come up with that answer. But really it's shifted now from that question, it's still important, where we get that information pretty quickly and check it, make sure it's right in the context of everything. But now the focus has shifted to everything leading up to that. How would things have been different if something else happened? And so it's just a different set of challenges with reconstruction as we've seen. And then technology as a whole, we have seen how it's impacted the industry. We think, oh, well, when event data recorders came into being, we're going to be out of a job. And really it just created these specialties that we've been talking about. You have to be a specialist now in event data recording. And then really just taking that data, how do you use it in the analysis to answer all of these other questions. Lou (01:23:38): Yeah, it's so important. And I think it is one of the themes that's come out during all these conversations that I'm having now, is the importance of the electronic data and the importance of knowing your limitations in where you want to focus your efforts, where you want to dive completely down the rabbit hole, and where maybe for instance, I just want to go to your class, get exposed to the technology enough so that I can intelligently direct my client and coach them on what's necessary in this case, and then just leave it at that. And I think with me personally, as you guys probably know, I've chosen to go fully down the motorcycle rabbit hole, the photogrammetry rabbit hole. And then the rest of the stuff, I need to know enough about it through conversations like this, and through courses like yours, to just be able to be the quarterback for my client to say, "Here are all of the things that we're going to need to consider for this case, and I can handle X and Y. And Z... (01:24:40): it's usually not just Z, it's Z, and then AA that other experts are going to have to come in to handle." That's a great Excel joke for the engineers here. And then like you mentioned, one of the other things that I just wanted to touch on before we go into some tools and then the future is the trailer data, which a lot of people don't even know exists. So those are modules that are fortunately not on the very front of the vehicle, so I imagine that that data is very often intact and can be interrogated with the Bendix and the WABCO tools. Is that how we're getting that data? Tim (01:25:19): Yeah. We talked a little bit in a earlier question about trailer data and really when we were talking about diagnostic trouble codes and where that can be handy, even just looking at a basic ABS and it's often overlooked downloading the trailer. So WABCO, in particular, they have a lot of data in one of their trailer ABS modules for stability control. For example, if you have a bulk liquid holler trailer, it may have one of these stability control modules on the trailer from WABCO, and there is an abundance of data being stored on that module. It's kind of a unicorn, I've only come across one recently. I was aware of it for many, many years and just had not come across a trailer with one of those systems on it. I think it might be a little bit more popular in Europe than it is over here. (01:26:21): But when I did see that, that actually turned out to be the way we got the data for the speed of the truck because it was a tractor trailer that was in front, and it got rear ended by another truck. And so we were able to download the data from the trailer and we captured an event. When that trailer got pushed sideways, it had a stability event and it captured the speed of the truck. There was no data in the tractor on that crash from that collision event. Lou (01:26:57): Yeah, that's something that's pretty new to me. I first learned about it from Steve Anderson not too long ago, and I was like, "There's trailer modules?" Okay, just further confirmation that I'm out of my wheelhouse and I should not be involved just other than telling my client to hire somebody who knows more about it. Matt (01:27:14): And usually it's even more difficult to interface with the trailer than it is to interface with the tractor because you're going through another separate set of adapters. And yeah, it presents its own set of challenges. Tim (01:27:24): Sure. Lou (01:27:25): Yeah. That brings me back to the point where you're traveling around trying to do this stuff, which just exacerbates the difficulty. Because if you have a Toyota Tundra and you can just put all of your tools in the back, you can bring everything and you're probably ready to handle everything. But when you're flying on Delta and you got a Pelican that's as big as they'll let you bring, what does a complete modern toolkit look like showing up to a truck? Matt (01:27:51): That's an interesting question. Yeah, because I think that's going to depend, I think on a lot of where your ceiling is in terms of capability. Right? Because depending on where your ceiling of capability is, might dictate you need a lot more tools to try to get you to the point where you're done saying, "Okay, I can diagnose to this point and then I'm going to finally throw in the towel." So do you stop at just a data link and maybe a multimeter, make sure I've got good battery voltage? Do you also travel with tools and direct to module cables or simulators in case you have to go direct? I don't know of anyone traveling with a soldering iron and a microscope doing chip transfers in the field? Lou (01:28:35): You'll be the first one. Yeah. [inaudible 01:28:37]. Matt (01:28:36): Conceivably, that could be something. To back up in a broader sense, it might be all of our jobs to set expectations of our clients about the way that these downloads can go. That if we have an attorney who's witnessed a couple of CDR downloads, explaining to them how this is not that, and that these can be a lot more complicated, take a lot more time, or sometimes might run into issues that we just can't get around and then have to use other methods to get there. Rather than being fully prepared for everything, is more so preparing your client for the different outcomes that could come from one of these. And then maybe there's a little less pressure on you to perform right, because we know it could get tough. Tim (01:29:25): Yeah. I think Lou, what we see with when you specialize in heavy truck crashes, you oftentimes get called out to the crash within 24, 48 hours of the crash, which may be a little bit different than other types of crashes, with passenger cars, and say things that are driven by the personal lines, insurance versus commercial lines. So because of that, it does present a different challenge to the investigators like us. You have to be ready with boots on the ground right away. You have to be... (01:30:03): equipped. So when you talk about where you're going, we look at a four-hour radius as kind of our rapid response radius to the point of being able to take your vehicle that's well-equipped, you have all the equipment. You don't really know what you're going to encounter when you get out there. So you need to have a full complement of tools. You may be doing a brake exam on a trailer, you may be doing a lighting exam, you may be handling various EDR issues out there. So to try to fly somewhere and have all that equipment, I don't think it's practical. Certainly not somebody that wants to be doing this for a long time. Good way to develop back problems if that's what you want to do. So some people have private aircraft and we've done that. (01:31:04): We've had private aircraft in the past and did rapid response work and it helps for that. But again, it's just over the long term it's really, really difficult and it can wear you down. So for that reason, we look at a four-hour radius. When we fly someplace, typically it's not the 24 to 48 hour response from the crash. It's something that happened a while back and we're able to triage it a little bit and figure out what exactly we're going to be seeing, what the objective is, and we can plan what we bring to the inspection. (01:31:46): So the other thing is, that's a great thing about teaching a class and getting to know people. Sometimes you know exactly who that investigator is going to be that's going to show up at the inspection. And it may be somebody in Philadelphia area, it may be somebody out in California, Sacramento, or someplace. And you can say, "Hey, are you going to have this equipment? We need to do a brake exam. We're going to..." A lot of brake exams on heavy trucks, it's a group effort. So there's tag teaming that goes along with that, and that's a great thing. So a lot of times that's how we handle those situations. Lou (01:32:30): It kind of goes back to the cooperation that we were talking about earlier with the exchange of information. I have found the same thing when you show up at an inspection and there's already some sort of relationship or mutual respect that you get that cooperation at the inspection and it benefits everybody. We're just hopefully looking for the truth and to do the best investigation possible. So that joint inspection cooperation is huge. So kind of going from the big toolkit, everything that would be necessary to do everything, to the minimalistic toolkit, like a sole prop who just wants to be able to do the basic stuff and then bring you guys in or somebody like you when necessary. It seems like with a NEXIQ and a couple pieces of software, you might be able to handle a decent amount of things. Tim (01:33:20): Yeah, I think so. You see some equipment's more common than others from market share out on the roadway. So if you look at the market share of various truck OEMs, that's a driver of that. So are you more likely to encounter one brand of equipment versus another brand of equipment? That can be a regional thing too. So you may be somebody that has a client that particularly likes to use you because they had a good experience in the past. And so knowing what equipment that client runs, whether it's a fleet is helpful so that you're ready for that. So I can't really give a specific answer to that. But you mentioned the FLA. The FLA, maybe that's something if nothing else, put the FLA in your tool bag. It may not be something that's going to be... If you want to do a lot of truck work that you're going to rely upon, but that's why it may be a good thing for law enforcement. (01:34:28): And Jeremy, I think targeted law enforcement with that device initially because something is better than nothing. And so that gave them a tool that in some cases they could hit a button, pull in data, and that was better than nothing. And then now you add, "Hey, if I can get the Cummins software, now I can add a little bit more capability to it." So that might, I think, be one tool to add to your bag of tricks if you're just getting into trucking. Matt (01:35:06): Yeah, when you talked about that philosophy of what can I get away with as a minimalist approach, like a NEXIQ and some software, I think we're seeing that as a philosophy for some folks in the industry where they say, "Okay, I obviously want to do heavy truck work, but I don't want to be a heavy truck EDR specialist either. I don't have the time and the desire to specialize like that." So that seems to be where a lot of folks are kind of drawing that line is they'll go out, assuming the truck can key on, they can plug into the DLC. They maintain the software and they'll do the download that way. (01:35:44): And then once it gets beyond that either direct to module or chip level work, then they're contacting other experts in the industry, other firms who can do that work for them. And there's some people that are real comfortable with that and like that arrangement because then hey, as soon as it gets difficult, it's kind of somebody else's problem. But on the flip side too, we are experts and we like to think of ourselves as experts and fully capable of doing everything. So there's that too. And so how far do you get? How much do you want to wrestle with this stuff? Lou (01:36:18): It's come up in some of my other conversations as well. And then I started thinking, "Well, what about these poor law enforcement officers who might not have that same luxury to sub out, who knows, a $10,000 download to somebody to have them fly out or they can't bring on Jeff Muttart to analyze the human factors. They can't bring in Rick Ruth to talk about some sort of anomaly." Maybe they can, maybe they can't. But it just seems like they're very often in a situation where they're being tasked with analyzing the whole case, which nowadays is a huge challenge Matt (01:36:50): In some senses I think yes, that's something, we're all going to feel it and they're going to feel it just as much as us in the private side. But I do think they have a bit of a luxury that we might not have on the civil side of things where on the criminal investigations, I think they have a little more discretion about what specific things are relevant to the charges that they're potentially investigating. And so from a standpoint of heavy vehicle data, there can be some cherry-picking there. We don't need to go after all of these different things if we can get the specific stuff we need from one easily available event data recorder. So yeah, the truck's got seven EDRs on it, but there's one that we can get for basically free, and we got it and it answered our question, easy. On the civil side, that's almost never the case. (01:37:40): We're being tasked with getting everything. This idea of preservation or trying to do this comprehensive sweep of all data that could be on a truck. So the challenge is very different for us where a lot of times we don't have discretion because our client is specifically asking us to get everything. And what does the word everything even mean? It's a whole other discussion. So to the point, for law enforcement, I think that's the secret tool in the toolbox is that with the discretion, sometimes there's a free way or almost free way or very cheap way to get maybe just enough to do what you need to do. Lou (01:38:19): Yeah, that's a good point. A lot of the times you talk to the law enforcement officers and they're just looking for gross evidence of some sort of negligence. If the guy's going 42 in a 35, that's not going to do it for them. They need to just pin down some methodology for establishing that they're at say, 70 in a 35 or something like that. So you're right, that's a great point. There's some more subtlety that we're after on the civil side that they fortunately don't have to pursue in every criminal matter. Tim (01:38:50): We see sometimes where the law enforcement actually works with a fleet. Let's say if the fleet may say, "Look, I'd like my expert to be able to come in and inspect the truck, and there's a discussion going on there where they say, "We will share the data with you." So that happens as well. We've had that in many cases. And sometimes that can be a benefit too, because now you're being able to see the truck before something moves and they have everything locked down. So there's a beneficial exchange that goes on there. Doesn't happen very frequently, but we see that from time to time. Matt (01:39:39): No, that's kind of like one of those great symbiotic relationships there where on the civil side you might get early access to a vehicle that's on hold otherwise. And for the LE guys, they get someone whose job it is to deal with all the problems of EDR to hand over a complete download so they don't have to worry about what tools and software. Here it all is, and you can look through it and see if it has the things that you need and sort of benefits everybody there. So yeah, very much is an arrangement that happens. Lou (01:40:08): You were mentioning earlier, Tim, that at the beginning, the advent of EDR, just light passenger vehicle EDR, that some of the recons were kind of thinking, "Well, there goes our job, let's start focusing on what's next." And then the video being omnipresent now, I think some people are thinking that that's going to solve a lot of their cases. And then the progression of EDR data becoming more and more advanced and including video like the Bendix Systems where you get what, 10 seconds pre 10 seconds post. Do you think there's a world where event data recorders become so good and so easily interpreted that us pesky recons are sidelined? Tim (01:40:48): Well, hopefully there's a world in which we are sidelined because there's no more crashes. That's what we hope for. I don't know if it'll happen in my lifetime. But no, I think if anything, it's just going to drive more specialization. Now you need to be able to understand how these systems are working, just like the collision mitigation. So it's no longer good enough to just understand how crashes happen, you have to understand how these systems are working. And so I think that's where the things are going to shift in our industry. Matt (01:41:33): On paper, it seems like more data equals more answers. And in practice it equals more questions, and you need someone who can provide those answers. And so I think the more data that we find on vehicles, the more modules that are recording information, the more systems that are flying around, just means there's more work for us as the investigators to understand and explain them. Because it's already at a point where we do this and feel behind, so someone who doesn't do it at all isn't going to be able to suddenly pick through it. So you mentioned does the data become, I guess interpretable enough or something? Maybe, but it doesn't seem to be going that way. It seems to almost be going the opposite direction where it's becoming more fragmented and more difficult to put together. So that's only going to drive the need for people with specific expertise in that. Tim (01:42:25): Well, and Matt, when you think about the SAE class that we teach, how much of the time would you say we spend on talking about what we call issues with the data? So it could be up to half the time we're talking about the issues. We introduce, what is there? How do you get it? But then the second half of the discussion is, "Can you rely upon the data?" And back to the designers of these systems, is this the first thing they start off with and do they spend a lot of time testing it out and making sure it's working correctly? Does it sell trucks? It doesn't sell trucks. So are they really going to spend a lot of time really doing the kind of work that we do to understand the data, how it gets created, how reliable it is? Probably not. So therefore you get a disclaimer on the data that should not be used for cross reconstruction purposes. Lou (01:43:33): Yeah, the EDR data's kind of done the opposite of what I think a lot of people thought might happen in that now we have this data, it's coming from all sorts of different sources. You have to be in tune with the sources of it. You have to be in tune with, like you were saying, Tim, the anomalies and what's required to interpret it. And now we can perform a more sophisticated analysis. We will know more at the end because of this data, but getting to the end is a monumental effort in a lot of these instances. Take video for instance, I'm sure you guys are working cases where you get video and it's like, "All right, we know the basics of what happened," but if you actually want to interpret that video evidence to determine speeds and use a scalpel to assess avoidability, well, you kind of doubled your bill at that point, so you'll have a better answer, but it's way more work. Matt (01:44:24): Yeah, the simple cases get much simpler, but the complex cases get far more complex. Tim (01:44:30): Yeah, I've heard somebody say that 90% of what there is to know about the crash, we know pretty quickly. The other 10% is really how the check's going to be written. And that's where all that time and of course cost goes into figuring out that other 10%. Lou (01:44:49): That reminds me of a quote that I love. It says, "You're 95% done. Congratulations, you're halfway there." Matt (01:44:56): Yeah, it does feel like that sometimes. Lou (01:44:58): So how can we best adapt and keep up with all this? That seems to be... I was going to ask this question. It's on my handy dandy journalist list here, but I think we've already kind of covered it, it's like where are we heading from here? And it seems pretty obvious. I'll let you guys chime in if you think differently, but it's like more data from different sources, it's going to result in more complex analyses that is going to require more specialty. Would you guys agree with that? Matt (01:45:26): Yes. Yeah, absolutely. That's a great summary of it. But also could there be regulation coming down that suddenly shifts that whole direction with the stroke of a pen? Yes. Also that. So it's a matter of I guess being nimble, being ready for whatever ends up showing up. Tim (01:45:48): Yeah, I would agree with that. And then with heavy trucks and what we've discussed with ELDs and just the various aftermarket systems and trying to understand what's out there, and you look at just the challenge for us as re-constructionists to keep up with that. But now look at it from a fleet standpoint. You say, "Man, all this data coming across." And some of the data can come across in real time. And you look in at our world and litigation and you say, "How can a fleet possibly even keep up with that?" It's talk about drinking from the fire hydrant. Yeah, and I think there's this expectation sometimes that they should know everything that's going on at all times. And I'm not going to debate whether that's the right perspective on it, but you can just appreciate the challenge that's there. (01:46:53): And even a fleet that's large to be able to do that is an immense challenge. Because the level of data that's coming across only increases. If you have 1,000 power units in your fleet, you just have more data to try to manage. And when you're looking at a fleet that's operating on pretty thin margins as it is, how do you allocate those dollars that you have? You're a business owner, you have to make decisions about how you allocate your dollars and tough decisions have to be made. And just as we look at the challenges in the transportation industry over the next years, right now trucking is one of the most dominant jobs in America today. (01:47:49): What's going to happen with autonomous vehicles coming in? That's probably going to impact the trucking industry and transportation more than anything is. Now we look at what are those people going to do? So seismic shifts coming across. So not only from what we talk about event data to talking about electric vehicles and the autonomy and the technology coming in, just lots of things that are going to be ushered in because of the pace of change that's going on. Lou (01:48:31): And you kind of both mentioned it, Matt, you mentioned being nimble and Tim, you mentioned drinking out of the fire hose and keeping up. How do us re-constructionists in general, how do we make sure we're getting some of that fire hose but not getting our face blown off. Whether that be a mixture of courses. Is there a group? I know there's a CDR group, there's the INCR group, is there an HVEDR group? Just what do you recommend for other re-constructionists to keep their finger on the pulse and make sure they're not falling behind in something as important as this? Tim (01:49:03): Yeah, I think it depends on the type of work that you're doing, but we've talked for this time about really being a specialist and choosing that. And I think, Lou, you would probably agree with this as you've specialized... It's counterintuitive, you actually have more work now as you've become a specialist than when you were a generalist. Would you agree with that? Lou (01:49:29): Yeah. When I had pipe dreams of just taking motorcycle cases, probably five or 10 years ago, I thought it was just that. I was like, "That's what I'm going to work toward, but it seems unlikely." And now I'm turning down most case calls and they're all motorcycle case calls and I can't handle all of them. So yeah, I would agree with that. Once you develop that expertise, that specialty, that knowledge, people find you. Tim (01:49:54): It's scary though, if you're... You were self-employed and how's the phone going to ring? When the phone rings, I need to take that job. So being part of a larger organization brings a little bit of comfort in that regard. So if you are somebody that's in a larger organization, I would really encourage you to look at developing a specialty. Because that's been my experience, and the experience really is DELTA |v| from its origins, we looked at, hey, we're going to focus on just crash reconstruction as opposed to going out and looking at some ladder failure or a trip and fall or whatever. And because of that, it really enabled DELTA |v| to grow pretty rapidly because we became known for when that truck crash happens, you can pick up the phone and call DELTA |v| and we know we're going to get a good consistent product from that. So that would be my message to people that are maybe starting out in this field is look at how I can specialize. Lou (01:51:08): I would totally agree, and when I hire younger people I think you need a certain marination period where you're getting exposed to all of the elements of collision reconstruction. But if something starts to blow your hair back, dive headfirst into it. And if you find out that after you do that it's not what you want, then feel free to change. But you will be rewarded for diving headfirst into something in this industry, as I guess all three of us are a manifestation of that concept. Tim (01:51:43): Of course, Lou, you know Brian Anders who started DELTA |v| and Brian's not practicing the craft anymore. But when I started working with him back in 2005, he really had developed a reputation, at least in the southeast region for heavy truck crashes. So it was a good marriage for us, for me to join him and take my experience with a truck manufacturer and compliment what he was already doing. And so at the time, he was doing a lot of work where he would have to fly places, so he got his pilots license. And I can remember real clearly one day he and I were flying somewhere for a crash and he goes, " You know, Tim, these event data these ECMs, I really have been relying upon dealers to get this data. And I think we need to develop that expertise ourselves." (01:52:44): And that was the start of where we are today. It grew from that conversation really just going to dealerships, asking a lot of questions. There were really no classes at the time, and learning a bunch of things. And then joining the SAE committee on J2728 and developing a relationship with people on that committee like John Steiner. And then John and I had a likemind where we said, "Hey, let's share our knowledge." And that led to us developing the class and going out and teaching. And so it's been a great journey and I just see that continuing on. Lou (01:53:35): And that led to so many good things, like you were talking about. The development of the SAE class, which is just a real benefit for the whole industry. I don't know where we'd be without that class, on the HVEDR front. Tim (01:53:46): Thanks. Lou (01:53:49): To Mack Volvo, you guys obviously have that relationship with Mack Volvo, where from what I understand, you're kind of handling the east side of the country and Mecanica and John Steiner are handling the west side of the country, and you're the experts on the acquisition of that data. And I imagine the subsequent interpretation of that data. So man, it's really cool to hear the origin of that. Just a quick conversation in a private airplane, it sounds like, heading out to an inspection. And look where we are now. Tim (01:54:19): And people have asked me, "Well, how did you become the Mack Volvo people?" And that was really just that J2728 committee. John and I were on the committee, and some people in the industry remember back in the time where you would send the modules off to Mack and maybe six months later you might get the data back. And maybe a few months after that, the modules might get shipped back to you. And it just really wasn't working. And so they decided they needed to get out of the business of downloading and made the decision that because the tool was not really ready for just providing to everybody, they went the direction they did. (01:55:07): And so we happened to be on that committee with Scott Hinkson that would do those downloads. And Scott said, "Hey, would you guys be interested in doing this?" And that was in 2008, and I thought, "Probably go for about a year or two, and then they'll release the software to everybody." And here we are, how many years later? 15 years later. And it still hasn't happened. Maybe it'll happen next month, I have no idea. But until then, we're going to do the best job we can for everybody out there and make sure we can answer the questions. And I'd say the upshot to all that has been, if there's any heavy vehicle event data that is the most well understood, it would be Mack Volvo data because of that reason. Matt (01:56:00): It's it. When you're able to see that volume of downloads from a single system, you can see all those edge cases and anomalies and strange things. And then you're like, "Man, if I could just see this for everything. If I could see that many downloads for Detroits and Cummins, we'd really understand everything." And that drives the whole philosophy behind trying to be open with the knowledge that we have, not just hoarding it all into DELTA |v| but teaching the class, sharing what we know, being accessible so people can share that with us. Because that's the closest we get to having hundreds and hundreds of the same kind of download come through, where then you can really feel like I've seen every permutation of this type of system. Lou (01:56:42): Yeah, I love it. And I mean, it relieves a lot of my anxiety knowing that you guys are out there and Mecanica's out there, because I know that when I get a crash involving a tractor trailer, I don't have to be the one to know everything there. If that was on my shoulders, I don't know what I'd do. I guess I wouldn't sleep, I wouldn't eat. I would just be constantly working and trying to learn new things. So yeah, this has been super fun and I really appreciate you guys making the time. A couple hours it turns out. We might've gone a little bit over what I had advertised going into it, but I really appreciate you taking the time. Is there anything that you wanted to say or cover that we haven't talked about yet? Matt (01:57:25): I don't think so. Tim (01:57:27): No. I think we've covered a lot. Matt (01:57:29): Comprehensive. Tim (01:57:29): Thanks for having us, really enjoyed it. Matt (01:57:31): A whole lot of fun. Lou (01:57:32): So where do people find you if they want to reach out, they have a case or they just want to talk shop? How do they find you guys? Tim (01:57:39): Yeah, so DELTA |v| Forensic Engineering, we have a website. And then also if you're specifically interested in heavy vehicle event data, hvedr.com. So that's all things Mack and Volvo you can find there if you get a Mack and Volvo truck. How to find one of the service providers, a little bit about the data, that kind of thing. And then also Detroit Assurance as well. If you have a Detroit Assurance Collision Mitigation System where you need data, we talk about that on that site, and we can provide assistance in downloading that data directly from the module if needed. Lou (01:58:30): Awesome. Well, that's great. It's good to know where to find guys like you, that's for sure. And thanks again for taking the time. Matt (01:58:38): Thanks, Lou. Tim (01:58:38): Thank you. Lou (01:58:40): Hey everyone, one more thing before you get back to business, and that is my weekly bite-sized email, To the Point. Would you like to get an email from me every Friday discussing a single tool, paper, method, or update in the community? Past topics have covered Toyota's vehicle control history, including a coverage chart, ADAS, that's Advanced Driver Assistance Systems, Tesla vehicle data reports, free video analysis tools and handheld scanners. If that sounds enjoyable and useful, head to lightpointdata.com/tothepoint to get the very next one.