The Blind Apex Podcast
The Blind Apex Podcast
Episode 169: Aero Myths with Andrew Brilliant
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Andrew Brilliant, of world time attack fame and beyond, joins me to talk about aero myths in club racing... and what we can do to better our downforce where rules allow. Yes, we even talked about rules... because I just had to bring it up.
You want me to tech term that up for you? So what you what you're describing is absolute accuracy versus relative accuracy. And and so you don't actually need absolute accuracy at all at that level. You know, you could say, how much down force does this car have? And that's really not important to you. The question is, do you have more downforce than you had yesterday?
SPEAKER_01Welcome to the play Apex again. Where amateur club racers tune in to get faster.
SPEAKER_02This episode is brought to you by Deechworks Fuel Systems. Use Apex 10 at checkout to save 10% on fueling your passion. Speaking of discounts, use Blind Apex 2025, all one word. Save 20% off a monthly or annual race setup pro plan. And if you want to support the podcast directly, go to buy me a copy, buy meacoffee.com backslash the blind apex podcast. Still getting used to that one. Business is over. And so on this episode, we're going to talk arrow. And it's a subject we've covered a couple times, but in the arrow industry, I think people have differing opinions and different levels of experience. And I really want to talk about the myths that we perpetuate in the paddock. It's just stuff that I see all the time, and it just doesn't make sense to me, but they people swear up and down. So joining me from the land of small bespoke garages, which I endear. They are truly endearing to me, and that because they produce highly competitive cars there. He's known for his time attack arrow. His cars have informed, been like covered by moto IQ, race engineering, race engine development, all the stuff that he's touched, right? He's got experience at the GT level with open-wheel cars. Andrew Brilliant, welcome to the podcast, sir.
SPEAKER_00Thank you for having me.
SPEAKER_02It's good to be here. Yeah. So I came across, oh man, some of the time attack stuff that you had done years ago now. It's 2026, so yeah, like over a decade ago, it feels like. Yeah. And then I started digging and I was like, he's still around. Let me talk to this person because he's built some crazy things in his life. And then I started looking at your website, and I was like, oh, he's been doing this for a long time and he's deep in it. So yeah, let's give our audience your background. Where did this journey start for you in Aero? And sort of how did you leave the United States to go to Japan? And where this journey, how did it come together?
SPEAKER_00Yeah, okay. So there's there's a bit of a story. I gotta try not to make the whole podcast the story, but long story short, it's probably started when I was probably let's see, how old would I have been? Maybe six years old. And my my dad had done his residency in Utah, and we were moving back to California and we crossed the Bonneville Salt Flats. And you know, as a kid, I was just, you know, I had my my toy cars, and you know, I was one of those kids, right? Like the future engineer kid. And and I just was the place is just visually stunning. Like it's just gobsmack, right? Kind of feeling. And I asked my dad, what is this place? And he explained it to me, and he said, you know, this is where the fastest cars in the world, you know, that run and and set the records. And, you know, he told me a little bit of the history about it. And I was just, you know, staring out the window for the for that hour that you cross over it, just imagining these these things. And my toys changed and my interest changed, and you know, that was from a young age. And then, you know, I followed whatever I could see on TV. That was like GTP, IMSA GTP, et cetera, et cetera, you know, growing up in the 80s, and then started doing my own land speed racing. That was my first sort of competing on my own in motorsport. And I was I wound up building my eclipse and I drove it to the dry lakes from Berkeley and changed out the seat and you know, try to set a record and drove home. So that that, of course, arrow, you grow into Aero there. And the a lot of the guys that compete out there are former, you know, Skunkworks or you know, Lockheed. These kind of Southern California community has a lot of that going on. So there was Aero expertise, and I just found it fascinating. And it just came at the right time while I was a student, and I learned so much and was just hungry to learn more. That's a that's about the size of it.
SPEAKER_02So uh okay, let's talk about your company uh and what you guys are doing for work, how it started, and what are you guys doing these days?
SPEAKER_00Yeah, so it so it actually started out where I was, you know, I'd always been into cars, as I mentioned, and then I was working and I paid my way through school working for a software company in San Francisco, and the the company president had his own team that ran in in e-production with a with a Mazda RX3, and they were frequently, you know, the local champions and would go to nationals. And I I got involved working with that team, and I was just sort of the guy who was you know good with the computer, and then eventually we started looking at you know, could we do any arrow things within the rules? And I started having to play with that, started getting into CFD. At the time when I was in school, that was really you had to be in graduate program to touch CFD. So I just kind of did it on my own, started finding huge gains, helped teams here and there, and uh it just grew. It just grew totally organically like that.
SPEAKER_02Yeah, it's funny that you mention CFD being you know graduate level when right now you could probably with enough money, I mean it's it's not necessarily cheap, buy your own license today, like random person, right? So it's something to be said, and maybe we'll talk about that later. I'm sure I have a question about uh CFD stuff, especially the modern things. But you mentioned before we hit record, you know, that you're less into the mass production world and you're more into the bespoke one-off, very custom setups.
SPEAKER_00So yeah, yeah, we don't we we will we will do development work for for a manufacturer if they want, but we only do it private label. So and we don't so there's a reason you have a conflict of branding, you know, you sort of build this brand for motorsport things, and then they'll want to use that brand to sell a product, right? So they're and you kind of are you're you're changing your brand to a different market to do that. So I've built our brand around pure motorsport stuff, and we do some one-off, like a manufacturer will ask us to design, help them design a part, but we usually have them do most of the design and we sort of run CFD and give them results back. So we're not really a production parts kind of workshop. We're a we're a high-end specialized motorsport shop.
SPEAKER_02Okay. I just I wanted to get that out there and set the stage, right? So this is you you've had the hands-on, you know, from your time with your own cars, working with the e-production cars and and the evolution of your business, right? But I just want to set the stage on where you are today. So yeah, sure.
SPEAKER_00That's more, I mean, I still have those roots and I still keep them. I mean, I do take on sort of garage build projects, but those will be where they're producing their own parts, not where they're buying, buying pre-made parts. For me, that's sort of a that's a bit of a uh, you know, I'm kind of like a I don't really understand the world of I can only bolt things onto the car. It's sort of like, well, if you're not building your own car, you know, I'm not sure what kind of race car you're building, but you know, if you you gotta learn those skills, that's sort of the for me, that was the point of building a race car. I shouldn't talk down to anyone who has a different purpose, but for me that was it was like you know, learning how to build it all yourself. And so, and once you start doing custom things, they always outperform. So you so if you're really focused on performance, you you would do that. That said, for most people, that's probably not where they want to go with a car. But for but for people who want to go there, you know, we can help them design that for sure at any sort of manufacturing budget, really.
SPEAKER_02Okay. I I actually think that's part of these bolt-on aerodynamic compromises we make. Um let's say I buy a splitter kit from Acme Aero. Uh hopefully that's not a real company, but it's fictitious. And I I have certain criteria as a driver, right? I have an open trailer, I would like it to be quickly removable. Quick is it means something to me, it may mean something different to you, right? Some people want to pull some tabs and the thing drops to the ground. Me, quick is five minutes, right? Or ten minutes, something more easily accessible. So we're all making compromises on our requirements, and then we buy the product and they've made compromises for max mass production, and so the optimization of that part may or may not be influenced by the compromises you made and they made, right?
SPEAKER_00So, yeah, yeah, and well to your point, right? Like, you know, when I lived in the States, I had two cars and one of them was a streetcar. And the streetcar had completely shelf parts from from manufacturers on it, right? So so I'm not I'm not trying to talk down to those things. It's sort of like a co horses for courses sort of question. But you know, we were we were in the context of you know club racing or something else like that, where I would assume you know, if you're running a club racing car, you probably you probably could, you know, buy a sheet of carbon fiber and make your own splitter, you know, presumably. So it's just different, a different situation that we're talking about.
SPEAKER_02Oh, I mean, the world is our oyster depending on our skill level and how much budget we have, right? And how much time we want to put into it. So I I do want to put disclaimer out there because we're talking in generalities. Yeah. And for club racers, we're generally in production-based cars that we at some level are retrofitting with race car type era. So not everything applies to all cars, but we're hoping I'm hoping to coax out some of that universal knowledge that I sure yeah, that I heard you hint on in an interview probably 11 years ago.
SPEAKER_00So okay, yeah, yeah, yeah. No, there's plenty of stuff we can talk about there. There's some there's a lot of things we've learned along the way for uh about that stuff that I'd be happy to share.
SPEAKER_02So the arrow world and the paddock, it's it we seem to see a lot of people who copy each other. Like if the leader's running some part, somebody's gonna run that same part. You know, we get copied in industry, we get copied from the paddock area when we perform well. And you were doing a uh the interview that I mentioned, and you talked about the limitations of rules and what they put on you, especially when you were doing some GT car development. And you started going into the process of CFD to wind tunnel, and then you later talked about like the question was do arrow parts and designs work with this work the same on different cars? And you said rarely, but there might be a couple of things that translate across all cars. So yeah, that's true. I'd agree with that. Yeah. Okay. Could you talk more about what translates? So let's set the basis of what applies to all of us.
SPEAKER_00Sure. Yeah, so I I mean, there's a the the basic rule is that the more optimized something is, the less universal it is, right? And that's a pretty straightforward assumption for anything, really. Like, you know, if you design a really nice piston for a Chevy V8, and then you put the same shaped piston in, if assuming you could even put it into, you know, a Honda four-cylinder, it'd be terrible, right? So the more any any sort of optimization problem is is like that as you get deeper and deeper into it. You you're tweaking finer and finer things, and then you have a system of synergies, and they rely on each other to happen. And it just becomes so specific that it completely malfunctions on another car. And then there's some really general things, you know. So take a step back, and then I've never heard of a splitter not making down force on a production-based car. That's just an absurd thing to even say, right? So, you know, obviously that's one end of the spectrum, but then you go and design some, you know, super complicated, you know, multi-profile, multi-angle F1 type of airfoil and try to put that onto every production car. You'll you'll find that without optimization work, you know, very few things will translate across. So, you know, you might be optimized for the the flow field coming off of the roof line. Well, that's different on every car, or how wide the bubble is, or you know, what the wheel offset is. So you you get down to like a wheel offset. You change a wheel offset, you have to redesign for in our world, you'd completely redesign the front of the car if you changed a wheel offset by 15 millimeters. So you you know that's that's where you get deeper into it. So those questions are always open, and you really never know for sure whether it'll perform or not until you test it. And testing it is so much work compared to actually just designing something for that car that people kind of never go there, right? Because you you've got a job that's 80% set up, getting a car into CFD, doing your baselines, doing your sensitivity testing, getting rolling is 80% of the job of designing just a splitter, let's say, for a club racing car. So, you know, why would you go and test a bunch of generic things? You wouldn't, right? It once you've gotten in that deep, you just design your own.
SPEAKER_02Okay. So I feel like we're poking around your philosophy on arrow a little bit.
SPEAKER_03Sure. Yeah.
SPEAKER_02So can we start talking about that? And maybe specifically, because we've talked about CFD a lot. So CFD time versus wind tunnel time. Do you think all CFD is the same? Because we no, no, absolutely not.
SPEAKER_00Oh, actually, you know what? Sorry, can I skip back and answer your question? So there um I'll give you a loophole and your your listeners might enjoy this. So there was a there was a front diffuser design that was, I believe, and I'm sorry if I've quoted this wrong and I offend anyone at Ferrari, but there was a Ferrari GT2 era design that was just a series of channels that were about 100 millimeters wide. And that design is like almost universal. I found that worked reasonably well on every car I put it on. And that was one of the earlier designs that I that I used, because at the time that was actually cutting edge for GT, and it was really easily translatable to different cars because you would, in a generic way, you could sort of just modify the geometry of this hand fabricated simple devices to the specifics of the car in a way that you could explain in two sentences. And it was made out of just channels, so anyone could construct, purchase, or or you know, what have you, fabricate simple, you know, four-inch wide channels to the American audience, and and and that that was it. And that that design probably outperforms uh more than half of you know car-specific parts out there, but it does require you constructing things and being comfortable with that. So so there are those sort of like loophole designs. Once in a while, I've come across these things that just translate well across cars, and that won't perform anywhere near what like apex of time attack is, but you're still talking about things that'll make you know GT car-esque spec down for us on a broad range of platforms. So you wouldn't really say no to that kind of thing, right? So so there's a question, there's a bigger question, which is what makes sense for this use case? And that's the more important question than what is the shape.
SPEAKER_03Okay.
SPEAKER_02So were you talking in the diffusers, the splitter diffusers, is that like the tunnels?
SPEAKER_00Well, I've I've heard that word tunnels used. I'm not exactly sure what they mean when they use that word like that. But you know, in in my world, that's called a front, it's called a front diffuser. And that's the you know, a diffuser is something which is growing in volume in the you know, as you move in the direction of flow of the air around the car, it's expanding volume. And so typically, you know, you have that in the underbody where the ground is part of your your your volume expansion. If you draw a rectangular on the exhaust of a diffuser, it's there's more ground clearance as you move aft than there is forward, right? So in that case, you're expanding volume, and that's a diffuser. So what you have underneath the splitter is is is a series or a single diffuser that's expanding in volume. So whatever you want to call that device that's expanding volume under the splitter, that I call that a front diffuser.
SPEAKER_02Okay. Well, I mean, I know them as tunnels, but okay.
SPEAKER_00Uh, tunnels tunnels traditionally that that's a name that came from venturi tunnels, which was a very old, you know, a very old word relating back to Bernoulli. And I'm not and it they called underbody tunnels venturi tunnels in the 80s because they look like half of a venturi tunnel. So I wouldn't I wouldn't think that's the correct word to use in this case because uh it they don't really have that shape.
SPEAKER_02Okay. I I the way you define diffuser, it's what you said. So yeah, yeah. We are we I guess we have to do some settle some common vernacular between us all.
SPEAKER_00So yeah, look, I mean I mean you can edit out this this this this part of it, I suppose. But I you know I'm just sort of so that we can talk in the same way. So for me, that's called a diffuser. If I say the word tunnel, I'm talking about something that goes from the middle of the underbody out the back of the car. That's called a tunnel. And I'm I'm relatively confident that that's the nomenclature because it's I mean, it's there's a lot of manuscripts in out there, you know, where they describe that that part that way. So I'm pretty sure that's the academic vernacular. I'm not sure about any specific industry vernaculars that I might not know about.
SPEAKER_02So I'm just the dumb guy with the microphone, so I will take your word for it and we'll use diffuser if we get into that later.
SPEAKER_00I can just really quickly define that and say that the in the tunnels are the underbody portion of a diffuser and why they differ from the word diffuser in an F1 context anyway. So, for example, we had a diffuser era in F1 and then a tunnel era in F1 that just recently ended again, and and that's defined basically by how far forward it starts. So you get something that starts in the you know, a rear diffuser that starts somewhere in the underbody, and typically it has a different design than diffuser. Diffuser is a simple expansion. A tunnel is where you start wide, then you get narrow. This is when you look from the top down, and then it expands again. And so that comes from venturi tunnel, is the original term. So when I say tunnels, what I'll mean specifically that that mid to rear underbody section and diffuser will be either from roughly the rear axle aft in a simplistic growth in volume, or the same device underneath the front of the car is called a front diffuser.
SPEAKER_02Okay. I follow you. Hopefully, everybody else does, if not, rewind and listen again. So, but okay, let's get back to the CFD wind tunnel, treating them all the same. Do you think they're they should be treated more like an engine dyno, or if you look at one, you should sort of consistently look at the results from that one? And then like sort of sort of like what what's your overall philosophy? So, and I can set the stage more if you want, because we do bolt on car apart, and they all have these claims, you know, and they show us pretty pictures with the colors and the pressure areas. And so I how do we sort that out?
unknownWell
SPEAKER_00I mean, okay, so the there was a there there's a past era of of CFD and then there's the modern era of CFD and the modern era of C F D is really, really available. Now this is a good thing, I think, for consumers, and it's actually what even made the work that we do or that we started doing to be feasible in the first place was that compute was growing at a at a high rate, and what was, you know, like I said, a graduate program 20 years ago is now my MacBook, right? So so you have a totally different scale of that. So there are there are different scales of CFD and there are different qualities of CFD. The idea that you know CFD equals CFD is kind of is a bit absurd to anyone in the field. You know, there's a huge, I mean, aside from the base tool you're using, you know, there's anything from$100,000 of licenses per year down to like a phone app that's free, you know. So that that's a really huge spectrum. Now, where you fall on that spectrum is is a matter of how you want to spend money. But my base philosophy to your to your question is that you know all simulations are wrong, some are useful. And that's that's the more important question is is how you scale the accuracy of what you're doing to what you need. So if you are developing something very simplistic, if you're dipping your toes in computational fluids, there's no reason not to use a software as a service, cloud CFD, or or a desktop level. Like maybe, maybe running on your gaming desktop, it might take you 24 hours or even a week. You know, when I started, I think doing an F1 grade simulation would take me so long I wouldn't even have done it, right? So so we started there too. And it's a question of how accurate you need to be. And I can I could delve into the math of how we determine that, which is a whole different topic probably than what you guys want to talk about, but there's certainly a stratification of the levels of CFD. So I I think now it's become very accessible because you have services like Air Shaper. You know, they have like a cloud, a cloud CFD, and it's based on, I think it's based on open phone. And it's a it's it's simplified compared to what we do, but that doesn't make it useless. Like it's got a higher, uh, a much higher cost per test, but if you're only doing five tests, would you want to do what we're doing? Would you want to spend, you know, a quarter of a million dollars on hardware? No, you'd you want to do 10 CFD tests. So, you know, you have to sort of think about your situation to answer that question. But in general, there's a massive stratification across across what CFD is out there. There's everything from like absolutely worthless, useless information to up to you know, super high-end services like Ancysis Cloud or you know, et cetera, et cetera. So so yeah, that does that answer your question.
SPEAKER_02It does. It does. I just I wanted to set more of the stage so people know your thought process and where you're coming from when I ask you these next questions. So I think that will help people understand where you're coming from and and what your mindset is, because you may or may not answer the same as other people or things may be different. They may think differently than you, but it would give them a perspective to understand maybe your answers coming up.
SPEAKER_00So yeah, no, no, it's it's good, it's good. There's a little bit of like trying to sync our lingos here, but but that's all good.
SPEAKER_02Yeah, so okay. So let's get into the heat a little bit. What are some of the most common arrow myths you see club racers repeat?
SPEAKER_00Let me think back. It's been a it's been a while since I've looked at this, but I remember I had there were two things that I always thought general that so there's general things and there's specific things, right? And it's I guess I guess let me ask answer your question with a question of like, are you talking about specific questions about shapes or about general ideas about how it works?
SPEAKER_02Ideas about how it works.
SPEAKER_00Okay, so so I would say the first misunderstanding I see is the idea that that downforce is made through a response of inertia, right? Like if I turn air up and I push air up, I make down force. It's like a it's a it's a Newtonian reaction of the mass of this air being pushed up, pushing down on the car. And that's not what's happening at all. And I'll give you one very basic, simplistic example of that that just destroys it instantly. If I put a flat plate parallel to ground, we'll presume that makes zero down force. And if I angle it up, it'll make down force, right? If I if I angle it so it's inclined relative to ground plane. Now, if I leave it flat and I put two diverging turning vanes, vertical turning vanes underneath, so I'm expanding that volume without any vertical displacement, that will make down force. In fact, it makes more down force. So that happens because you're expanding the volume and you're accelerating the air underneath. And so in that case, you produce down force without any vertical force reaction happening of the air. So that's so that's how I can prove that that's not true. So the idea that like a canard being on a car makes down force that way on the canard itself is typically not the case at all. So you we might put a canard on a car that actually only supports a load of 10 or 15 kilogram force on it, on the part itself, but its effect on the car might be 150. So that was an example that we did on what was formerly the MCA suspension car and is now like Tanuki. So that car picked up around 120 or 150 kilogram force from a canard, which is huge. That's at 150 miles an hour. So, you know, you're talking about hundreds of kilogram force. That part would never support that load on its own. It would have snapped right off the car. So obviously there's something more complicated going on, right? So the the vertical displacement theory, I'd say that's that's completely wrong. Another thing would be getting air out of somewhere equals down force. So that's there's a bit of that in an old school kind of way of thinking of like an air dam on the front of the car. That makes down force by getting air out. And that's not untrue, but it's only part of the picture. You know, if no air enters something, obviously it has a low pressure. But you you actually will make more down force by controlling the air and accelerating it than you will from having none of it there. Well, if you you could never really achieve zero, that's why. So so those are totally different to an aerodynamicist, those are completely different situations, what's going on there, the the the causes and the reasoning of how it works. So those are two uh huge misunderstandings that I noticed when I was involved at the club level. So this would be, God, 20 years ago, but but I think they still persist. I think another another misunderstanding is that in aero-controlled, quote unquote aero-controlled classes, that there isn't anything to be gained from downforce. I've never found that to be true, not even once in my whole life. Even classes like I was running in Bonneville in a in a production car class, which is supposed to be the stock, you know, OEM or OEM fitment parts all over the car. And I completely changed the aerodynamic performance of that car within the rules without breaking a single rule. So there's, you know, that's that's not true. I mean, NASCAR, IndyCar, these are one or lost on arrow, and they have spec arrow, you know, templates that are being checked, and and and they're still winning and losing on arrow. So so I'd say those are the three biggest ideological myths.
SPEAKER_02So I actually I saw an interview with you many moons ago, also, and I had to bring this up, and it was one thing, and maybe you can explain how it works, or or if you remember saying it. But you said that a lot of people misunderstand how the cooling system is a part of the complete front arrow package.
SPEAKER_00Yeah, absolutely. Absolutely. I'll tell you a story about how I quote unquote cheated. I don't know if you want to call it cheating, but but I had a problem in the land speed days. I had front lift. My car was completely unstable, like absolutely unstable. And I spun, I spun out at 180 something miles per hour. It was terrifying, and I really had to resolve this problem. So we were in a stock body class, and you're only allowed to have a factory fitment cooler in the front. And after much research and deliberation, I determined through a combination of physical testing, instrumentation, and CFD that the lift in the front was completely coming from the cooling system. Which, in retrospective, now you realize that that's the case for a lot of cars, especially 90s cars and earlier, 90s and earlier. But it like if you drive a 1960s car, there was a thing they actually had vacuum-actuated headlights that would point down as your speed increased because the front of the car was lifting so much that they'd high beam everybody. So this is a real effect. And I learned that after I'd figured out my own car. And what I did to solve that on my car was I just had a radiator made with a really fine fin pitch, right? Because we were going very, very fast. So air was coming through that cooler no matter what. And by increasing the the loss factor of the air coming through the cooling without breaking any rules, we were able to significantly improve, in this case, a reduction in front lift, aka front down force. And that totally changed the car, right? So so you know that on a road course car, that probably would have been worth a second a lap because you went from front lift to front down force. And we did that with returning the factory ducting and sealing it up and changing the the loss factor of the of the you know the rate of the flow through the front radiator, and that was it. So cooling is absolutely integral to everything you do. Not just that, but you have to think about where that air goes because anytime you try to make down force, especially on the front, which is usually the challenge for for cars like this where they're allowed to wing in the rear, you you have to fight that cooling air. And if you create low pressure to make down force, that air just wants to flow right into that space. It's it's to air downhill is from high pressure to low pressure. It always rolls downhill. So it'll just go right into the space you do. So you you you struggle to simulate it because you have a really complex thing where air is flowing around, you know, you know, you got wiring harnesses or turbochargers and hoses, everything going around in an engine room that you do, and then flowing out into the underbody. That's a really complex thing to simulate, and almost almost without exception will result in front lift.
SPEAKER_02So we talk we're talking about down force. Okay. Right. What do club racers think creates down force, but actually doesn't?
SPEAKER_00You're asking me to tell you how someone thinks that's hard. I don't I'm not quite sure how they think. Honestly, I haven't I haven't really sat down with that kind of audience for for some years. So maybe you can tell me what what do they think makes down for us, and I'll tell you.
SPEAKER_02Well, the they cut up fenders immediately. They're quick exhaust from the floor, for example. Yeah. Right. They're they're venting their fenders.
SPEAKER_00That's generally true. I'd say that's generally true.
SPEAKER_02Okay. They're quick to throw away their fender liners and any of the plastics in the front end making space, those types of things.
SPEAKER_00Yeah, I'd be careful about that one because that there is a lot of work that goes in, goes into those designs in the modern cars. You know, anything it depends on the car. You know, like if you go back to the 80s, fender liners were about like rust protection, you know. And then you go to like a car produced in the last you know, 15 years or so, those fenders have been carefully sculpted for arrow, drag, down force, whatever it was that they're after. In general, back it just to the last point we were just talking about, a lot of cars will stylistically go for a really large intake for the cooling system. So a really good example is like the FT-86 that has a really large cooling inlet area, much larger than the car needs, but the exhaust side of the cooler is essentially a completely closed-off engine room, right? There's there's almost nowhere for the air to get out. And that's how they prevented the lift. And there's actually a law here in Japan that the cars have to be NZL near zero lift. So they, in order to comply with a rule like that, you know, either formally or informally, you you would have to carefully control that cooling exhaust. And the bigger you make that intake, the worse that problem gets. So they're shoving a lot of air in and then they're not letting it out. And that's how they're keeping that car from having front lift as a result. So if you simply removed those on a on an 8.6, like if that was the only change you made to the car, I would expect that you would see significant uh increase in front lift.
SPEAKER_02Okay. I I always thought getting the air out of the engine bay would reduce the lift.
SPEAKER_00Well, it is if you get it, if you put it out to the top, right? So going back going back to the earlier example about the the turning veins underneath a plate, that was my example about the force reaction theory of downforce. So if you go back to that that story, you can imagine what's actually going on is you're creating a low pressure underneath this this plate. And the the bot that plate is facing down on the bottom and facing up on the top. So if the top and the bottom have the same pressure, you have you have zero force, right? You have no down force and no lift. And if you in if you increase the speed of the air underneath, then you change its pressure to be lower. And so now you've got the top pushing towards the bottom. It wants to push that plate into the ground now, right? Because there's a difference in the pressure. So the way you actually make down force is just by having a downward-facing working surface that has lower pressure than the upward-facing top surface. That's it. So if you if you take that air and you dump it out into the bottom of the car, you'll generally speaking increase pressure underneath. And if you do that the other way, where you you let's say vent your your bonnet or your hood with louvers or or well-placed ducts, et cetera, et cetera, in general, in cars like that, that'll be that'll be a really significant way to to improve the front down force or reduction in lift because you're you're taking what was once making lift and turning that into, you know, you get rid of it. You just exhaust it somewhere else where it won't do that. And generally what happens on top of the hood is that that air is turbulent and dirty, but it runs into the windscreen. And so that's sort of the end of that turbulence. Okay.
SPEAKER_02So we live in the world of do-it-yourself stuff for club racers, yeah. Uh obviously, we don't all have access to like scanning tools and CFD, although it's becoming more affordable, you know, sub-1,000 scanners, proliferated CFD. But for the do-it-yourself or the club racer in general, how should we be prioritizing our arrow upgrades and then testing those upgrades?
SPEAKER_00Sure. Yeah. So I'm gonna flip that order on its head and I'm gonna say that without data, you don't know anything. And so you've you've gotta do, you've gotta collect data. So when I when I was doing my land speed car before I had a CFD model, I was measuring with zip ties on the shocks. You know, that was that was how far I took it. So I had to drive very carefully to not bias that result. You know, but but with the front, when you're accelerating on the front, it you only get it'll squat, right? So you you can see, for example, your peak compression in the front would be really consistent on a on a dry leg to know what your front lift slash down force would be. So you have to build, you have to build a test method first because you just you struggle to have the resolution to know by just raw feel once you get past really basic sort of things. Like you can't you'll struggle. If let's say you made three rear diffusers, one at six, one at eight, and ten degrees, and you put them on the car and you try to go by driving feel, whether that's worked or not, that's really going to be difficult for you to determine, right? You'll need some sort of a method to measure that. And that does not have to be something extreme or expensive. So, you know, in the early days, I had test methods that were everything from 18th scale toy cars in the bathtub to you know, magna helic gauges for pressure, you know, airspeed testers that were$20 from eBay, you know, it just goes on and on. You just but you have to have tools, whether that's a GoPro and some tufts or you know, a few pressure taps, whatever it is that you use, you you have to build that. I think in this modern day, you know, data loggers are really inexpensive. You can even if you step outside of motorsport data loggers, you just get like a laptop connected USB logger. That stuff is so inexpensive. There's really no reason not to do that. I mean, you hop on DigiKey and buy you know pressure taps for five dollars each, you know, it's hard. You sort of run out of excuses if you have any DIY sort of motivation, you know.
SPEAKER_02Yeah, it as long as you stay consistent or evolve your testing and transition it from one to the other, so you'll have a period of overlap, like if you're doing the yarn tufts and tape and and your GoPro, and then you're gonna add pressure units or whatever. As long as you're transitioning and overlapping, so you understand the transition of the prior data to the new data, then you should be okay. Uh my data nerd by trade. So as long as there's some overlap, you know, you should be okay in transitioning from one method to another. So right.
SPEAKER_00So the so uh you you want me to you want me to tech term that up for you? So what you what you're describing is absolute accuracy versus relative accuracy. And and so you don't actually need absolute accuracy at all at that level. You know, you could say, how much down force does this car have? And that's really not important to you. The question is, do you have more down force than you had yesterday, right? That's the really important question. So your baseline may not be accurate, right? You or you may not even have a baseline value. So let's say you use a pressure tap situation. So you might know that you've got some pressure value near the kink line of your rear diffuser. Okay, if that's if you have lower pressure than you started with, you've improved, right? That's all you need to know. You don't need to know how much down force you had. So once you rule out a need for absolute accuracy, that's where huge amounts of money gets spent. Correlation, accuracy, you know, even even a really high-end F1 grade wind tunnel, I mean, it was only not that long ago where I was looking at F1 tunnel test data where the correlation was too was double-digit. So, you know, that gives you an idea that absolute numbers are are different than than people, you know, sort of imagine them. So, so knowing that you only need relative data, yes, you you do a baseline and then you measure everything relative to that, to your point. That's really all you need to look at at that level, I think.
SPEAKER_02So I I agree. I absolutely agree because it's it's those baby steps on are you, it's the simple question is are you getting better? You define better as more downforce or whatever, less drag, right?
SPEAKER_03Yeah.
SPEAKER_02You you pick you pick what that question is, but as long as it's improving and you have that methodical way to measure, as it could be simple or whatever, but as long as it's consistently used, then you're doing okay. So right.
SPEAKER_00So the more the more consistent it is, so you so you you do tests on on repeatability, right? So you'll so in a wind model, for example, a repeatability test will be you you test the same car, the exact same configuration. You don't touch the model, you test it, you know, across multiple times across a day, multiple days, multiple weeks, summer, winter, whatever you want to do, right? And you you look at that and say, okay, our our plus-minus is, I don't know, we'll make something. I will say five points. Five points of uh of down force is our is our tolerance. And so you have to throw away any gains that are inside of your repeatability window, right? So if you've gained three points and your repeatability was five, you don't actually know if you've gained or not. So when you start out doing your earlier development work, I believe really strongly philosophically in a sliding scale of accuracy. You know, when you're doing your first arrow things. You can get away with a large tolerance and not repeatability. Because if you're talking about a one-inch splitter versus a four-inch splitter and you want to quantify that gain, well, you're going to be able to pick that up with a with a really coarse tool, right? I mean, you could you could see that on shock pods, a driver will feel it. You know, there's a million ways you can measure that that are coarse, and that's all within your tolerance. So as you get to the higher end of motorsport, you start saying, okay, well, our relative accuracy really matters. And that goes back to your question about how good a CFD do you need, how good a testing do you need? Those all tie together and say that comes down to that repeatability. And repeatability matters for that relative gain performance because you might have an absolute number or you might not, but you still have to be able to trust that gain. And so that's what you have to be constantly looking at is your tolerance in your measurements, how does that compare to the gain that you've that you've measured, gain or loss? And you have to know what results really mean something and what don't so you touched on something and I don't know.
SPEAKER_02Uh it's sort of a touchy subject, right? Because it's the you you talked about the splitter lip size essentially. Does it really matter? Absolutely. Okay. The end. Yeah. Okay, so here's my follow-on because I it wouldn't be an arrow podcast without me uh beating the drum that I think these are drum rules, but I don't want to influence your answer. But okay. In the United States, most of the popular classes, for whatever reason, have all started moving towards flat horizontal splitter blades with limited lengths, usually so you can't see them outside of the bumper when viewed from above or with a certain measure, right? Three inches from above, right? Some generic, but they are taking the shape of your splitter away from you.
SPEAKER_00Right. Yep. Yeah, they're the well, they're trying to make it an arrow controlled class, I suppose, is the goal.
SPEAKER_02Yes. But okay. So if you could rewrite club racing rules, what part of arrow rules do you think would need to be re would need to be changed? What about influence you?
SPEAKER_00Yeah, I mean to what to what end am I am I trying to change these rules? Like, what's the goal?
SPEAKER_02Like a budget-friendly but effective because uh because it's really plain Jane, right? Like our rules in almost every class in in the club racing world in the United States is your wing can't exceed the width of your body, can't hang over the bumper, and then you know, you have a this flat horizontal splitter up front, which is really boring and not always that effective.
SPEAKER_03Sure.
SPEAKER_00So yeah, okay, well, so there's two there are two I okay. I'm gonna have to do this this this annoying thing where I give you two opposing answers to the same question because there's there there are two competing things in my mind, and this is a philosophical question, really, right? So on the face of it, on the face of it, if you ask me what I thought a club racing class should be, I completely disagree with the idea that arrow should be controlled and engines should be unlimited, for example. I find that to be a that that extreme end, that edge case, is dangerous, I think. And the reason why I'll say that is because if you crash, you've you've got a situation where your inertia is going up at the square of your sp your velocity, right? So crashing at high speed is how you die. No like how many people have died crashes at five miles per hour. How many people? It's a very small number, right? And how many of yeah, and how many die at 500 miles per hour, right? So yeah, yeah. If you crash at 500 miles per hour, that's a hell of a thing to survive, right? So, and that's just the physics, and there's nothing we can do about that. So if you want to think in terms of inertia and safety, you know, because humans just have a max capacity for for inertias we can endure, right? Because we weren't evolved for 500 mile per hour crashes. We were evolved for bumping into a rock or something, right? So so you if you want a fast lap time, quick lap time, and a very safe car, it'd be low power, high drag, high downforce, right? That's that's the ultimate lap time to safety ratio. And but there's another competing issue you've got, which is you know, good racing, where you have this dirty wake behind cars. And it and the more downforce you make, the harder it gets to maintain that behind another car. And so that's where you see F1 sort of hemming and hawing with various methodologies to try and improve racing. But that's that's at a really high end, a really optimized end of the spectrum. And I don't really see that happening for amateur racing. So I think in general, I disagree with the philosophy of controlling arrow, and I think they don't actually achieve it. So going back to your earlier question about making that splitter flat, well, I've done enough, enough developmental work on that to know that I mean you play with that rule. You play with that rule every way that the rule allows you to, every way that's specified, you redesign the entire car around what that rule contains. And you still they're never gonna stop someone like me from making down force on a car with a rule like that. It's just not gonna happen. We'll find it somewhere else. You spend money wherever your biggest gain to, you know, your biggest cost performance is gonna be. That's where you spend your money. And it's not like you go, uh, I've got this much money and let's give it all back because it's the right thing to do. If you've got the money, you're gonna find a place to spend it to go faster. And that's always how it's worked. So trying to control cost via aero technical regulations, I feel has been really well proven not to work. Every time they do that, the costs just explode because the margins become smaller and smaller and smaller. And the and that accuracy, we were just going back to that absolute and relative accuracy thing. You start drilling down to needing, you know, 10 times more accuracy, 100 times, 1,000 times more accuracy. That cost goes up at the cube, not at the square. So you just blow out cost and put it on little tiny fine details. You go and say a template on an indie car has got a tolerance of one millimeter. Well, everyone's making their own body parts at one millimeter, like 0.999, right? That's expensive. So I don't think it works. I don't think it works. I don't think it's ever worked. I'm sort of shocked that no one's figured that out. But but that said, I do understand that they want to make the cars in a certain image, right? They want to have a class that's this type of car. And they don't want it to just be dominated by error. And I understand that. So there is a balance to walk. That particular rule, I don't find it very effective, not to someone like me. But but I do think that what used to be inaccessible, you know, like I said, that you that was a graduate studies thing to even touch CFD when I was in school. And now it's a software as a service cloud thing on your iPhone, right? So it's not, it's not completely outside of the reach of everyday people anymore. It's truly not. And it's it's an area of massive innovation. It's the most once you have the basic things on the table, you have a good suspension, you have a decent engine, you have a good driver, you're not going to find more bang for your buck than arrow because it's something that's basically untouched on cars like that. So I think it's a a huge area of interest to young people. I think it's something people quite enjoy doing when they start learning how to do even lightweight fabrication or fiberglass work themselves. It's quite, it's quite enjoyable work. It has a huge return on investment in terms of performance. It never ends, like you can just keep playing with it forever and just keep milking more and more performance out of it. So I find I I I've found it to be a really enjoyable experience for the teams and for us to develop club level arrow. And I don't think those regulations achieve what they're looking for, to be honest.
SPEAKER_02Yeah, I I 100% agree with you because of what you said. You you nailed it. I couldn't have said it better because okay, you limit me in this area, where else can I find it? And then the floodgate opens, right?
SPEAKER_00Yeah, yeah. Well, you you know how they get there though, right? You know, they get there because someone does it. Someone calls up, calls me up, goes and does it, and the car's two seconds faster. And then everyone complains, and uh, what are we gonna do about this one car that's running away with it? And not everyone has access to those resources or whatever it is, and so they react to that, right?
SPEAKER_02Yeah, but the re you didn't change the resource level, you just changed the allocation of the resources.
SPEAKER_00Yeah, yeah. Well, that's always been my argument, too, because we were always beating teams that were had 10 times our funding. So my argument was always like, here we are, you know, those teams spent$100,000 on engines and we spent$10,000 on Aero, right? And we're winning. So, you know, is this really a cost thing? It doesn't really make sense to someone like me. And so, but then but then what they said, you know, for example, one sanctioning body came back to me and just said, well, not everyone is you. And I just said, but I'm not the only one of me on planet Earth, right? And and I wasn't born, you know, like I wasn't designing F1 cars, you know, from from the womb or something like that, right? It was a learned a learned skill, right? So so I don't really agree with that. You know, once again, we can circle back to that. I don't agree with that, but I I understand the motivation behind what they've done. I just don't think it works.
SPEAKER_02Yeah. I don't think so either, but hey, you know. Which is probably why you you wanted to interview me, because you were struck. I I've uh I I've gotten in trouble. Well, I've been accused of some things in the Aero department in the past, but I was I did go to the wind tunnel and got outed by somebody present. They went they needed some Instagram content, and they took a picture of my car in the tunnel, and it immediately got to everybody I raced with, and that was a quite an interesting weekend of answering questions.
SPEAKER_00So yeah, there's a there's a bit of a stigma about it that I don't understand because it's kind of like like I said, you know, oh, I can spend twenty thousand dollars on an extra three horsepower, but if we've touched a wind tunnel, you're just it's like you're cheating or something, you know. Yeah, so I I don't I don't know. But like, and the thing is, I mean, what are you gonna do? You go and make a rule. I'm I'm digressing a little bit, but you go and make a rule that says no wind tunnel testing. It's like, okay, well, what if I light some dirty rags on fire and a big fan on the front of my car? You know, is that a wind tunnel? I learned something, or you're gonna say, well, you can't 3D scan it. I mean, come on now. Like, what how how are you gonna enforce something like that? And nowadays you can 3D scan it from photos to some level of accuracy, right? Right. And that technology is gonna get better and better and better. I mean, the how long until a 3D scanner is built into your phone and you're taking 3D photos? Years? Not even a decade, I doubt. So you you know, what's a thousand dollar scanner today? That's an iPhone 25 or whatever, right? I don't know how you'd stop it.
SPEAKER_02And then one of the the AIs automatically knows how to assess it for you. Yeah, like translates it, measures it from your photos. Like it may not even be on your phone. You could just upload it, maybe. Who knows? Stuff's accelerating at a huge pace.
SPEAKER_00So yeah, yeah, those things are going to change, you know. And that's you know, we're I mean, to that point, I think there's some there's some cool new startups in that space, like doing AI assisted things. And I think those are actually a really good thing for a lot of people, you know, where they're there there's an expertise gap, right? Where they they're you know, it's just uh if you go and ask Chat GPT about like quantum mechanics or whatever, it's not gonna tell you what what you know the people at the top of that field would, but it'll certainly give you a wonderful introduction interactively. And I think depending on what level you're at, that's that's that's super useful information. And we couldn't possibly serve that market, right? So, so I think those are those are high value to a lot of people as they get better and better over time. And you know, to that end, you know, I've uh about three or four years ago, I rebuilt our whole system to be building up every test we do for training data, right? So we could train our own AI on that. So I certainly am a believer in that technology.
SPEAKER_02Yeah, and only the future will be able to tell us what's coming, right? Like and it can be tomorrow, it could be a week, it could be a decade. We don't know how far out it is, but yeah. I I if you talk to Chat GPT or something like that, Grok or whoever enough, maybe maybe it'll convince its developers that it needs to learn more about Aero or something.
SPEAKER_00You just have a limitation of what you can train train an LLM on. You know, if you had an LM that had specifically trained on, you know, the the corpus of all aerodynamic, you know, stuff in the world, and specifically that one topic, it will be a completely different level, right? A model of similar size, say chat GPT. It might even know things that I don't know, but I still think it won't replace a craftsman or an artisan because you get into this whole like, you know, I'm gonna test infinite things, and you just don't have the time or the resources to do that to sort of get it up to domain-specific knowledge level of what experts in the field have. And that's not happening soon, but that to that but to that point, people like me, we have to just keep learning and keep accelerating the pace that we're learning at and using the AI tools to teach us as well. And that's how we stay, you know, five steps ahead of what uh you know AI version of us is going to do. And that's just a reality, not just for us, but for every industry, I think, moving forward.
SPEAKER_02For sure. I completely agree on that account too. So I have a I have a segment for 2026 that I want to get your take on. It has nothing to do with Aero, but you have worked on race cars, you had or your own race car, you played in the e-production realm. So the segment is on your dream race car or on your race car, abs or no abs.
SPEAKER_00Who's the driver? That's my first question. It wasn't and it's ABS. I'm a terrible driver.
SPEAKER_03Okay.
SPEAKER_00Terrible is wrong. Terrorism is I in the context of the drivers I've worked with, I'm a terrible driver. Okay, that's fair.
SPEAKER_02That's fair. Yeah, so okay. That here's a here's another question. I I'm sure you're of the same era as I am, or similar, and you played a little Gran Turismo in your life. So in that game, washing your car improve the arrow rating of your car. Is a smoother car, a slicker car better, in your opinion?
SPEAKER_00Yes. Yes. But you're talking about something really difficult to measure. Like the the the tool that you need to measure that difference is pretty expensive and ridiculous, and it's a very, very fine difference that you're talking about. Now, that's context specific. Like if you go and said, Okay, you're I I'm trying to remember what it was. It was like John Force or someone was, you know, they they did something with the paint to help the top speed. Well, that thing's going 300 and something miles per hour, right? That's a totally different speed regime than than a 120 mile per hour top speed at Button Willow, you know. So so yes, it does, it does. It does. But it's it's really small. It's really small.
SPEAKER_02Yeah, so you guys putting on 30 layers of ceramic coating, it's not probably helping your laptime, but it will help you remove the rubber marks easier.
SPEAKER_00Yeah, I think there's a you know, I I'm not gonna take away from the sort of pride of ownership aspect of it because I think there's a certain thing of like when you keep a space clean, you notice things. Like have you ever noticed that uh when you come into the living room and it's and it's a mess, you leave things because you don't realize that they're there, they don't stand out to you, right? You you leave that cup instead of putting it putting a dishwasher or whatever, right? So you there's a kind of thing like that where when you keep that car pristine, you notice a crack forming in something, or and so there's a there's a sort of indirect way that I think that affects more than than the direct way.
SPEAKER_02Oh I'm I'm 100% in the clean race cars are fast race cars because you can notice things, because things become out of place quickly. You know, well I still have body damage I haven't fixed, but in the grand scheme of things, I should fix it, right? Because if my fender is out of alignment on one side versus the other side, if I walk from one to the other and they were actually matched and in the factory location, maybe I'll notice that my camber is out of spec or toe is out or something. So taking pride in your car is important. So I I agree with that.
SPEAKER_00Yeah, and I think that should be true. That should be true regardless of the the budget or the you know, that's just a matter of of your time and your passion. You know, there's no reason, I think, to have a car in disarray, you know. For sure. And I think there's often rules about that too, I think, in some of the amateur classes.
SPEAKER_02Yeah, it's a 50-50 rule. It's it's kind of it's 50 miles an hour at 50 feet. Does it look okay? It's just a little crazy.
SPEAKER_00But well, I mean, that there's a certain point where you're like, okay, well, now you've got a maintenance cost that blows out to keep this thing like absolutely mint when you're just you know hammering on it at the 25 hours of Thunder Hill or something like that. You know, that's just you're not gonna keep up with that. But but the basic idea that it shouldn't be a huge heap of rust, you know, that there's there's safety reasons for that, right? For sure. And promotional reasons, too. But I I I I to that end, I will say, like, to me, like an LMP car beat up, like just in a in a state of of you know, layers of dust and dirt and whatever on it at the end of a uh at the end of Le Mon 24, that's a beautiful thing to me, too, you know. So yeah, that's that's racing too.
SPEAKER_02Oh yeah. I mean, it it put in the work for 24 hours, you know, when it comes off. So it it did its job. I'm actually surprised they don't ceramic coat it well enough that it's not that dirty, but they come back pretty pretty gnarly looking. They're definitely a matte finish sometimes.
SPEAKER_00So yeah. You know, I think I can't remember how much weight came out of my car, but I actually went and started scraping off the layers of El Mirage dust when I was changing my car from from land speed to to like road course config, and there was so much built-up SoCal dust in that car. I think it was probably a few kilograms worth, yeah. Because it just it piles on and then humidity and then it sticks, and it's basically like caked on mud kind of thing, and then layers and layers and layers just build up. It's pretty wild.
SPEAKER_02Yeah, that's crazy. Well, Andrew, where can anybody find you if they want to do some work? Because I know you mentioned some club racers that have worked with you in the past. So if people feel like reaching out to you and getting some consultation time and some projects looked at, working at sure.
SPEAKER_00Yeah, I mean you can go to AMB. That's that's my initials, andrewmichael brilliant-arrow.com, and that's the the AMB arrow side. Although we're transitioning now to a new corporate structure, which is applied dynamics research, and that's ad-research.org. Uh you can go to either of those websites, and those, those will all, it's the same team. We just AMB arrow is now a brand under Applied Dynamics. So you can find us there. There's feedback forms. My personal contact information is on there as well. If anyone wants to reach out, there's also a feedback form and you can you can inquire.
SPEAKER_02And he will answer your email. I swear. That's how I reached out to him.
SPEAKER_00So uh yeah, I I try uh to I have to disclaim that we are a small shop and there are times when I'm traveling or I'm at an event and I can't check for some days, and uh I tend to work in clumps, you know. I'll I'll respond to emails in a clump and then like that. So so give us a little bit of time sometimes, but but we but I do I do read that email inbox. So
SPEAKER_02Well, Andrew, I really appreciate you coming on the show, sir. Yeah, it's been a pleasure.
SPEAKER_01Yeah, my pleasure. Thank you. Until next time, keep working on yourself. Keep working on the car. And let's get faster.