Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
John Buckley: Real-World Tire Testing and the Breakpoint Debate
563 Plays2 days ago
**CORRECTION** John asked us to note that he gave incorrect figures for the wattage standard deviations during the episode. The best outdoor tests currently achieve standard deviations of approximately 0.8 W, while indoor testing has achieved a standard deviation of 0.3 W.
John Buckley of Streamlines returns to discuss the outdoor tire rolling resistance testing he conducted with cyclist and Escape Collective contributor Ronan Mc Laughlin — covering the methodology behind measuring CRR with an aero testing rig, the temperature sensitivity finding that nearly sank the dataset before it began, and what the data actually shows about tire pressure, width, and the widely-cited breakpoint theory.
Key Points
- Streamlines' method establishes CDA on baseline runs, then holds it fixed and solves for CRR as the energy remainder. Repeating baselines are the integrity gate for any outdoor tire test
- Tire CRR sensitivity to temperature is approximately four times larger than published literature suggests, and the relationship is nonlinear at low temperatures.
- Once vertical stiffness is equalized across widths, rolling resistance on smooth tarmac is nearly identical regardless of tire width.
- John has never observed the pressure breakpoint in any of his field data. His critique is both theoretical — the mechanism requires a sudden onset of vibrational losses not supported by the data — and empirical: pressure sweeps consistently show gradual, small differences rather than a hockey-stick curve.
Links & Resources
- Streamlines: https://streamlines.aero
- Streamlines in the field — aero testing with Mattie Dodd: https://streamlines.aero/blogs/news-insights/position-data-and-a-blustery-november-aero-testing-with-mattie-dodd
- Contact / inquiries: info@streamlines.aero
- Ronan Mc Laughlin's Escape Collective tire testing article: https://escapecollective.com/introducing-escape-tyre-testing-get-ready-to-rethink-accepted-wisdom/
- BicycleRollingResistance.com — GP5000 multi-width comparison: https://www.bicyclerollingresistance.com/specials/grand-prix-5000-comparison
- John Buckley on Endurance Innovation (August 2022): https://zencastr.com/z/h0CqHZy5
Recommended
Transcript
Introduction & Guest Welcome
00:00:00
Speaker
We're publishing these tests. they They have repeats in them. You can see the repeats. All it would challenge is somebody says, OK, we saw the breakpoint. Show us your testing methodology and show us how your repeats were for that day because we we're trying to be pretty transparent here. And as far as I'm concerned, you know the data we published on this escape collective, I sleep completely easy with that one.
00:00:23
Speaker
To me, it's like the burden of proof is on the breakpoint.
Streamlines Updates & Testing Protocols
00:00:35
Speaker
Hey folks, and welcome back after a very long hiatus to the Endurance Innovation Podcast. Joining me today is a returning guest, ah John Buckley from Streamlines.
00:00:49
Speaker
and John and I have a lot of things to talk about with you today, but before we launch into that, I just want to say, John, thank you very much for taking the time. I know you've got a lot on your plate right now, and welcome to the show.
00:00:59
Speaker
and Thank you. It's good to be good to be here again. Excellent. ah So look, it's been ah it's been quite some time since we spoke, I think easily a year. And ah before we launch into the the meat of the conversation, which is going to be about the testing that you did with Ronan McLaughlin of Escape Collective on tires, um catch us up on what Streamlines has been up to in that time.
00:01:24
Speaker
Yeah, and I haven't actually, i'm trying to remember exactly where we were with the the podcast. The last time I think that we were we were going ahead with, we were testing with teams and businesses and we're really kind of still there. we i think we were just, we were developing wind models at the time. And I think at the time, i can't remember whether we were using a ah constant airspeed at the time or not.
00:01:46
Speaker
But we've definitely moved away from that now. We definitely now tested variable airspeeds. And that's been kind of, to To keep the story, to keep an overview, I'd say that a lot of it has been just develop both developing the the models and protocols to test, figuring out what matters, how to make it simpler.
00:02:07
Speaker
And so now we work with about 10 different um companies and teams, and work Yeah, we provide them with what we call Cirrus system.
00:02:22
Speaker
And so that consists of three so three on bike sensors plus a power meter.
Testing Challenges & Strategies
00:02:29
Speaker
So you have... um going call it an aerometer, aerosensor. we tend to call it a wind meter just because, yeah, aerometer and aerosensor are taken. And also, really what it does is measure the wind. And I think that kind of causes some complication because if you call it wind sensor, least people understand just what its function is, that it's measuring the wind on the day. if you've got no wind, you don't need one.
00:02:49
Speaker
ah And then, so there's that. There's a body position sensor and then a speed sensor. The speed sensor is just a magnet to wheel, and they're just very consistent.
00:03:02
Speaker
you know You can get a whole debate. I think there is some very good GPS technology, and you can talk about the pros and cons of that, but that's separate. And then the body position set position sensor to where we pick up on the...
00:03:16
Speaker
on the chest and on the head position. And that just measures that so that we can track whether the rider is moving. And so those three sensors plus the power meter, they all get recorded. We run a um an Android device on the bike. We've got a ah nice little bottle rack mount. and we've tried a few different things. We've tried logging on bike computers, but The nice thing about the Androids is that they're, it's just super quick to use and it's super quick to upload. It just it just ends up being simple. And the fact that you can get these, we use this Samsung S10e and the fact that they're, but you you can't get anything like that for the price. So we know we've got a we've got a really robust and like bottle rack mount and it's just quick to use. And so so that allows us to like that allows us to upload to the portal.
00:04:07
Speaker
and then the web portal will show you results right away. So if you're creating a session, you just put in the combined mass of you and the bike, you put in the CRR, and then you need to have two gates marked out on the road.
00:04:20
Speaker
And if you've already done the course, you can just select them on the portal. If not, you just physically mark them the phone like you would drop in a pin. And then you're ready to go and you're Most of the time we would say to ride back and forth between the two gates. And we would generally recommend a usually run four kilometers per per run.
00:04:42
Speaker
And then- Four kilometers one way. no No, sorry, ah one kilometer each way. oh I see. then yeah And so just four traverses, it's no benefit to go any longer. you can go shorter, definitely more powerful riders can go shorter.
00:04:57
Speaker
i guess you're waiting you're you're looking for for things to stabilize, right? For the rider to come up to speed and like be stable in their position and stable in in speed. Actually, we we try to get them um we try to get them stable before they cross the first gate. So we tend to give 200 meters each side.
00:05:13
Speaker
ah So on the whole, it ended up being just under a mile. Mm-hmm. Yeah, let alone a mile. And then that just gives you an acceleration zone. And so then even when they stop, just continue down the road a little bit so you get up to speed on the next way. And we would generally say to ride at um something like FTP because most test people will do will be like, say, 30 kilometers. And you can usually get 30 kilometers.
00:05:39
Speaker
You can usually maybe try like six different, six or seven different things. Yeah, and because you get you get your brakes between the road. It's a workout. I've done them. I've done plenty of them. But it's not it's not like excessively heavy.
00:05:54
Speaker
Yeah. And just what I'd say really like where we've come is that we say now that, yeah, you can test, you know, you can test you need to, you test eight configurations in like two hours. And we've got people that are doing that. We've got teams who will test, they'll do the training camps and they will test, I've had teams test ah like 20 riders in like six days.
00:06:14
Speaker
So they just crank them through. And my my favorite question, because i yeah I've done a few of these tests, is what are your limitations on test conditions? Like specifically, ah the ones that are the most difficult for me are flat ah route and traffic.
00:06:33
Speaker
The big thing with traffic, I would say, is You can't get stuck behind a car or you can't have a car directly behind you. we're We're pretty good with sort of two or three cars. I would say two or three cars maximum per traverse. So take say say the traverse is like 90 seconds. Then you're talking about, so two cars a second.
00:06:52
Speaker
that that That's definitely doable. You want the car to make a quick pass if if they pass you. Yeah. So that would be two passes in in that time, which is actually pretty busy.
00:07:03
Speaker
Yeah, two cars per minute, yeah. Yeah, two cars per minute. So that'd be two passes. That's more just unenjoyable for the rider, just going to pass like that. I live in a big city, yes. I give back to that for sure.
00:07:15
Speaker
and So that's that's the traffic one, to be honest. Our biggest... Our biggest difficulty is that we need a relative straightness on the road. We don't need it to be flat. we We've got um the elevation.
00:07:29
Speaker
we We can measure elevation now to usually within, it's ah under 10 centimeters for over the kilometer. Is that through IMUs or or or GPS or a combination? or What do you do It's a combination of GPS and barometric pressure. And it's that's like so solid. It's just like it's it's something we don't even think about anymore. So the elevation is just not not an issue. The only thing with elevation is that you need to be able to get a decent airspeed both ways.
00:07:55
Speaker
so yeah So you don't want to do a big climb. If you had a climb, you'd want the climb in a headwind. Because it's actually, that you want the airspeed both ways. I got you. It's kind of counterintuitive. No, makes it makes it makes good sense. It's no fun to ride in. No fun whatsoever. But it makes sense because, yeah, you're getting more airspeed. So it's it's it's as if you're riding faster.
00:08:19
Speaker
Yeah. I mean, what what we see really is that everybody who's testing out on wind tunnels, who's doing skin suits in wind tunnels knows that that the CDA is dropping with speed. I mean, it's just really apparent you're testing a wind tunnel.
00:08:34
Speaker
And then people, not everyone is sort of like taking that knowledge and saying, well, how do we adjust our outdoor, you know, how do we take that into account when test outdoors? You know, and that was, that ended up being really, really quite important. What I would say here, the reason why we need the minimum airspeed is that above a certain speed, it's pretty easy to model that, to model that difference. And then once you get, once you start getting below sort of probably eight meters a second, it starts to really increase a lot, um which will tie in a bit the tire testing, which is, and just has to do with why we don't do, we'll get into that later on. Okay. it's Essentially the rollback. Pin put in that. Okay. Yeah. Yeah. Put a pin in it. So,
00:09:27
Speaker
can't be too steep uh we need in regards to bends what we would say is that we don't like to have a bend you can't tap the brakes on a bend we prefer that there isn't a large corner cornering force on on the bend because the cornering forces uh slow you down and um so So the main thing is we need a straight stretch of road with not too much traffic.
00:09:55
Speaker
And that that's that's the main thing. And then obviously
Finding Suitable Testing Locations
00:09:58
Speaker
it can't be, unless you specifically want to test like gravel stuff, which we've done, but it needs to paved. And so what we often do is we often just look, we we can usually find routes for people just from looking on Google Maps, look for look for areas near farms and farm roads that don't go anywhere are just perfect. It's funny you say that because so I live in Toronto, a pretty big city and it sprawls um and it's impossible to find, oh hu very difficult to find a straight enough road that's, let's say 1500 meters that doesn't have traffic on it.
00:10:31
Speaker
and doesn't have lights to interrupt it. And I finally found the the perfect place and it's farmland. It is in a, in like a little pocket of, of what used to be marshland, now flat farmland. And it's delightful because there's almost no traffic except for the occasional tractor. And it's actually an awesome place to train if you're doing like flat TT stuff or like long course triathlon training. That's not very hilly. It's amazing because you can spend a lot of time in the bars, but it's also phenomenal for aero testing.
00:10:59
Speaker
Nice. Yeah. Yeah. Farmland. And the one other thing which can be worth looking at are um sometimes the bike. ah Sometimes you can find bike trails that are suitable. The only in there that is an issue is if there are people on it, you might need a spotter there or somebody because...
00:11:15
Speaker
Especially if you've got somebody going, you know, you've got somebody going 30 miles an hour yeah and there's kids on bikes, it's just not safe. so No, so especially in like a TT position where you might not have all that visibility of the road. Yeah. Yeah. It's it's a trick. And also a lot of them tend to meander, like not not a ton of the ones that we have here. They're usually like paralleling rivers or or lake shores and they are not straight.
00:11:39
Speaker
Yeah. Yeah. But we don't need it that straight anymore now. Like as long as you actually the visibility can be almost as much of an issue just that. Yeah. Because as you said, you couldn but you just want to be safe on it. so Yeah. And it was funny because and I'm kind of jumping ahead. But one of the things I remember reading in Ronan's article is that he did his at night to minimize the traffic on his little path.
00:12:02
Speaker
Yeah. um Yeah, we can get into that. Yes. Okay.
Testing Baselines & Consistency
00:12:08
Speaker
All right. Put a pin in it. For sure. where I'm running out of pins, John. um Okay. So this is a ah great preamble. Thank you very much for for catching us up. Obviously, we're not, this isn't an episode, listeners, about where we tell you how to arrow test and all the You know, the whole bucket analogy that I've probably beaten to death in past episodes. um We're kind of rolling into this with the assumption that you you you have a grasp on the basics and now we're up to speed on what Streamlines has done in the last little while. So let's get into the the meat here and the testing that you helped Rona McLaughlin do, where interestingly, he wasn't really after the aerodynamic data. He was after ah rolling resistance. So he was he was doing some tire testing.
00:12:52
Speaker
So i think one thing that is really important for us to understand is how the heck do you get CRR data, which is the coefficient of rolling resistance for your non-nerds, from a setup that is designed to measure aerodynamic drag?
00:13:08
Speaker
Yeah, and that that kind of goes back to i mean the fact that what we call aero testing is really you're testing everything on the bike, but the, can you measure everything, but the aerodynamics and technically we say everything, everything, but actually we don't measure the rolling resistance as well. And, uh,
00:13:31
Speaker
If you have any spare pins, we can talk about what we do is we we fix CRR when we do aero testing. And then when we aero test, we fix, or we're sorry, when we tire test, then we fix the we fix the CDA. So the process here is that we will we start on a smooth road and we will estimate the role and resistance.
00:13:57
Speaker
And then we will get a a CDA for the bike. That CDA, that CRR just needs to be ballpark. So let's say you've got to set a set of GP 5000s. It's a very smooth road. You know, it's like 0025. It's bit rougher. Oh, wow, it's that low. Holy smokes.
00:14:14
Speaker
For GP5000 on a smooth road, it'd be that low. And then you get a bit rougher, you get up like 0.3, 0.4. But the important bit here, and this is where I think people get a bit hung up on, is that if you make an error there, that shifts your CDA a bit. It doesn't make a big difference on the final rolling results in terms of where you are,
00:14:39
Speaker
from the rough road to the smooth road or from the different tires. It's essentially an offset. So if you start out at, at say, you know, 003 and you measure someone's 004, it's essentially carrying that it's carrying that offset all the way through. sure And we've got a paper which we can which I think is published on our site or is published even in that Escape Collective article showing just going through the the mathematics behind that. And it's it's really solid. And I mean, I think you know you if you don't believe it, have a read read through it. you have any questions, you you can send us on. It it it really stands up. it's but Once you start doing the numbers, you can realize that that that it works out so that all you really need is a good estimate for the CRR. We then establish what the CDA is. We then carry that CDA through the through the arrow testing and
00:15:29
Speaker
In the example of how Ronan was testing, he tested on multiple days. And so in that case, we don't actually, we sort of offset the CDA. So he would always start with the same baseline.
00:15:43
Speaker
And so everything he did on that baseline was relative to that starting baseline. And he established a baseline on every single day, right, which was really important, I think. Yeah, yeah. And, you know, it's kind of a question people say, like, a do you have to the baseline? Of course you don't have to do the baseline.
00:16:03
Speaker
But it it casts an awful lot of doubt into any of the measurements. So one of the things that with that test, the reason we did so many baselines was because... So we thought that ah basically, yeah i you know I think that if somebody else wants to say, all right, is this are these results believable? I think the baselines there are the really the key thing that is like, I think the test really stands out because you look And just day after day, we're able to repeat you know the first to last.
00:16:36
Speaker
And that the we're not saying, oh, there's a bit of a drift. and we're And when we account for this drift, but if you say you're accounting for the drift, your baselines need to repeat. And that's essential. And so like technically, you don't have to you don't have to do baselines. But if you want so you know if you want to be rigorous, you have to.
00:16:57
Speaker
Yeah. So let me ah let me kind of ah paraphrase what you said just to make sure that I follow along. So for for this test, ah what you did first was you needed to understand what Ronan's CDA was. So you assumed a CRR.
00:17:13
Speaker
Because as ah as you were saying, you can measure everything except for CDA and CRR. So those are your two unknowns. So you have to start with one as ah as an estimate. And I'm 100% with you that with ah you know good tires, good conditions, you have a pretty good... I was a little surprised by the value of the number, but to your point, it's just an offset. So you assume a CRR.
00:17:33
Speaker
You plug it in, you do a of however many passes you need to do to have that confidence. I know Ronan likes to, he's a very diligent human. He does a lot of these. ah So you get a baseline CDA, then you now take that baseline CDA and you plug it into your, now it's now it's your known rather than your unknown. You plug it into the equation and now on the subsequent testing, you're now measuring CRR.
00:17:56
Speaker
Yes, yeah. And so if you can look at it another way is that We calculate everything so we get a certain number of watts and then we' got a remainder, which is the amount of rolling resistance in watts. And then we take that rolling resistance and then we divide it by the speed, we divide it by the combined weight of the bike and rider, and then divide by gravity, and then it out pops that CRR number.
00:18:22
Speaker
Or what, like for instance, the Escape Collective article, we published everything in and watts. So then we said, we take that CRR number and convert that back into watts at 32 kilometers an hour for Ronan's weight. Got it. Okay, that makes real good sense. um And I wanted to, when I first was thinking about the flow of this episode, I wanted to talk about some of your really interesting findings. But i'm going to park that ah because I think now is a really good time to talk about what i was was saving for the end of the episode. And that's the ah the signal to noise question that I have. um yeah
00:18:59
Speaker
And that is that is this. ah In my experience, and again, I'm not nearly as experienced as Ronan or you and in in this field testing stuff. But um i I believe that aero testing is super noisy. there's just There's a lot of inputs. There's a lot of things you have to get exactly right. um And there are a lot of influences on on the the final outcome. um And when you're doing testing for ah CRR,
00:19:29
Speaker
the the signal that you're looking for, that or in your case, the watt value that is attributed to rolling resistance is relatively small compared to the watt value of the aerodynamic drag.
00:19:40
Speaker
um So how did you ah how did you handle that problem of having such a high potential noise ratio to the signal that you were trying to tease out, which was your your your wattage cost of your rolling resistance?
00:19:54
Speaker
That's really valid concern. or It basically is the... It is kind of the core of the issue about this tire testing. And so I think first of all, when you're talking about the baselines, your noise is what your baseline repeats are.
00:20:11
Speaker
And so that's the whole purpose of the baseline repeats. They establish what your noise is. So in this case, our baseline repeats were generally within one to two watts of each other. That's a huge credit to Ronan because being able to repeat at that kind of ah that kind of fidelity with aero testing is is um is very, very difficult to do. So what I'd say here is that you kind of asked... In terms of holding your position, right? Like that's the that's the biggest... one What i want to say here is, and you were talking about um what have we've been doing over the past few years.
00:20:41
Speaker
We're generally testing now. We've got teams whose median standard deviation, like their median standard deviation is under one watt now. Are you serious? Yeah, and that's that's in all conditions. That's not on selected days. So we've, I mean, i think that was something of which I think there was, think a lot of I should say things are moving.
00:21:08
Speaker
Things are moving in this space. They're they're not standing still. Okay, I've got i've got i've got ah i've got a question. So clearly there's there's a gap in my understanding here, John, because I always thought that you know position obviously makes such a huge difference to aerodynamic drag. That goes without saying. I think anyone listening to this already appreciates that fact. So then my understanding was like sometimes even fairly minor movements of the rider's body, especially from one run to the next. Like we've we've had guests on talking about how even in the wind tunnel where things are more or less static, it's very hard to repeat the the exact same position from, let's say from the first run of the day to the last run of the day as folks fatigue. So hearing that you you folks are testing to under one watt repeatability is is kind of blowing my mind right now. So, I mean, just as a, or we had somebody who they wanted to do this, like this Tom Compton challenge, you know, this one where they put the the discs on the bike. Oh, yes. Yeah, yeah, yeah.
00:22:09
Speaker
And so not sure whether we'll get a chance to publish this data set, but I swear it exists. And this is an outdoor test. And the standard deviation on those discs for the baseline run was 0.3 of a watt.
00:22:25
Speaker
That's wild. I don't remember the test protocol. I think we were talking to... it's just Yeah, it's not even a test I agree with very much here. But we as I said, the median is under the median is under a standard deviation. We test this under a watt. So we're just talking about standard deviation. Standard deviation is like about...
00:22:43
Speaker
Two-thirds, yeah? Yeah, exactly. I'm not a statistician, but it's like... sixty yeah And I'm very worried because people will say here, you know, like, anyway, no matter what you say in statistics, you got murdered. yeah For sure. No, they they can yell at me for that. That's just from memory, from like engineering math from like a million years ago. Two-thirds of your stuff will be within one watt if you say your standard. And so here, two-thirds are within 0.3 of a watt. um And that's just been from...
00:23:12
Speaker
just just just Just refinement on essentially what we do is energy tracking. We attribute where all the energy in the bike is going to. But if the rider moves and that movement creates a different sort of set of aerodynamic conditions, I understand you're capturing the wind and you're accounting for that. I'm i'm okay with that. But if the rider moves because they're tired or they shift on the saddle or something and that, you know, creates some air, changing the way that the air flows over them and that creates a ah you know a difference. Because we're talking about, you know, look, like these these guys are riding FTP. Yeah.
00:23:45
Speaker
The total watts that they're putting into their bikes, you know, between 300 and 400 watts, sometimes even more. So we're talking about like incredibly, and and you're within one watt.
Real-world vs Wind Tunnel Testing
00:23:56
Speaker
That is, you know, what, like 0.2%? It's super low.
00:24:00
Speaker
Two big things here. One is um you're right to point out the rider. It's highly dependent on the rider. And so we've got a few... You know, there's there's a few repeatability of superstars out there. um So, like, we do a lot of testing with Q365. And I don't have his name right here. there There's a guy there who just, yeah, he just knocks him out. yeah we we We can take it out to four decimal points, but he' he'll do the three decimals on the CDA. That's wild. All day long. And this is with wind.
00:24:36
Speaker
And what...
00:24:41
Speaker
So yeah, it's it's very rider dependent. And I think the rider is definitely one. The rider is the biggest variable in this. There's obviously a test to test repeatability. But one other thing that I think a lot of people aren't realizing is just that there are wind tunnels which know,
00:25:06
Speaker
you know $4,000, you know, probably there's probably some aviation ones that probably cost $10,000 an hour, but there's like, there's motorsport ones that cost $4,000 an hour. Okay. And then, you know, there's wind tunnels that an hour. and there's a reason why the an hour tunnels cost four thousand dollars an hour and so
00:25:34
Speaker
that There's a thing about the the the wind tunnel and balance quality that go in that are a question. And then there's also the wind tunnels tend to oh tend to be a bit oversensitive to Reynolds number.
00:25:50
Speaker
So velocity, mean? so velocity healing yeah In this sense, though, the the the air in some places in the wind tunnel will be very laminar, which is why what you're not seeing. So therefore, you can have a certain drag reduction from a certain thing, which won't be which you won't be able to replicate outside. And you can you can end hyper-optimizing to that, but it's not even repeatable within the wind tunnel.
00:26:14
Speaker
Because then at the end of the day, you're not repeating to that. And that that's... in For instance, in Formula One, that is the thing of which buries a lot of teams. you know that was you know why i worked from to around and that was...
00:26:39
Speaker
yeah that's your background before before the cycling arrow stuff right yeah yeah and But when he was young there, we were always doing stuff that were like these tiny, you know producing this with these tiny gains.
00:26:51
Speaker
a lot of them um were just these wind tunnel optimizations and they actually make the repeatability of that wind tunnel model worse because you're you end up with these hyper-optimized things which are then, you're just hyper-optimizing the wind tunnel. it That tends to happen less in the actual, um in the dynamic conditions of real-world testing. Even, I'll give an example, I think the new benchmark should probably be, I mean, and really it's only really been the past month,
00:27:21
Speaker
which I've really come around to saying this, is they this Catesby Tunnel in near Silverstone. And it's it's it's like a three kilometer long indoor tunnel, but it's a straight smooth road.
00:27:32
Speaker
And you can ride up and down that. And I don't even know. It was like an old railway tunnel or something? Yeah, exactly. And so they paved over it. But... You get the advantage here, at least, that the rider is moving dynamically so that it tends to be less sensitive to some of the, to these sort of like micro, it's just, it's just it's it's it's more repeatable than a wind tunnel. Sure. Okay. So if I'm if i'm understanding you correctly, um there is there is a signal to noise question here, but you are able to, at least in this case with Ronan, you were able to get his repeatability so
00:28:09
Speaker
solid that you were still able to tease out the ah the signal that you were looking for in the CRR? So if you're if Ronan was one to two watts, if I remember you saying correctly, what is the what's the magnitude of the CRR contribution to the whole picture?
00:28:24
Speaker
It would generally have been around on the smooth road, I think in the high 20s and then on the rougher roads offhand, I want to say the or in the forties or something.
Temperature Effects on Testing
00:28:37
Speaker
I, let me look. ah I know I put you on the spot now, but no, it's all right. I've got the, I've got some of the, ah it goes from the high twenties up and up to, um, and this is obviously for both tires, right? Like that's, that goes. Yes. Yeah. Yeah. So, so,
00:28:53
Speaker
The highest stuff would have been in the 60s. 70 watts of rolling resistance would have been would have been the highest ones. on On the really rough road, yeah. Yes. Cool. I think for me, at least, this is such a rich topic of conversation. There's just so much. it's it's It is really bleeding edge stuff. And it's also, um there's there's a lot of...
00:29:14
Speaker
Maybe I've been out of it for too long, but there's some dissent in what you hear from one camp to the next. So there's always a lot of interesting stuff dig into. So signal the noise. You were saying that on smooth roads, you're around 20 watts for the pair of tires. You got up to maybe 60, 70. But what's the delta? Let's say you're on smooth tarmac because that's the hardest maybe.
00:29:39
Speaker
the hardest maybe um ah case. What's the delta between like you know the the the one step in tire pressure to the next step in tire pressure on smooth tarmac? And I want to compare that to your, let's say, one to two watt noise.
00:29:53
Speaker
Yeah. So, and just to qualify here, just because I am selling stuff, it's The 1 to 2 watt noise here, like we we can we can do better. And and some of the people say is like it's from knowledge from other tests to where to where it's under a watt. So what we really see in this one, even with the 1 to 2 watt noise, is that for the smooth roads, we see a decrease in rolling resistance with increase in pressure. And we're talking here, some for some of the tires, it could be as high as...
00:30:25
Speaker
ah I'm just looking at now, probably 6 watts, 7 watts. For Ronan's case, this is going from 40 up to 75 psi.
00:30:36
Speaker
Okay. so That's definitely a lot of signal to to to not as you know to ah not as much noise. Yeah. but And also here, just as we can speak about the noise, but we also did it in a number of cases. We repeated it. Also, there's multiple tires and the trend is consistent across everything.
00:30:58
Speaker
So like, I think Ronan contacted me of like February of last year saying he wanted to this test. Oh yeah. Okay. It'd be a good chance to try this out. And so then he started testing, I think around like April and I wasn't ready for this, but the initial results were just, yeah, they look terrible. I mean, the, the, the, the baselines were just, just terrible. I mean, so much so that it like your, your power meters, but something's just wrong. Yeah.
00:31:24
Speaker
And, Also, like I was busy. He, I don't know, he was just sending me, oh, I've got some results. I mean, he was testing like the middle of the night. What the hell's going on? I'll have a look at it next week or something. And then i look at him and was like, God, you know, going to break his heart because he'd done like, if you look, the bulk of the testing was done like in two weeks in April at night and he'd done all this stuff. And I looked at him like, this is junk. But then um didn't really get this whole architecture done until like,
00:31:51
Speaker
August and then this you know and then i was like I had already said to him like you're gonna need to redo a lot of this stuff and so I was looking through the results and I was really trying to figure out why What was wrong? why Why was it so bad? And people have published stuff. I think Tom Anhalt published some stuff. to stuff on bicycle rolling resistance showing temperature, but everybody sticks in the sort of reasonable range of like 20, 30, 40 degrees c or you know which would be
00:32:22
Speaker
relic but like sixty five you know like sixty eight sixty eight to a hundred degrees farenhe And so you kind of think, all right, it's just a linear extrapolation. I mean, the factors that we saw accounting for temperature, we were like four times bigger than that. and And this is one of the one of those like really interesting findings that that I think we teased out at the beginning of the show, yeah?
00:32:45
Speaker
Yeah. Just to close the loop on this, so that was... You can't just say, all right, here's published data and tire temperature. We're going to turn that up. We're going to multiply that by four and say we can compensate for temperature with this. It just doesn't make any sense. And then I ended up go reading through all of these like German and like there's these...
00:33:06
Speaker
There's this Finnish guy that does loads of stuff on truck tires and like, you know, Arctic temperatures. And so he's developed the whole model. And so with that, it's this um kind of exponential curve as you get colder. And so once we threw that in there, the temperature compensation. Sorry interrupt. As you get colder, just to ground our listeners, as you get colder, CRR increases, yes? it doesn't just increase in some cases it kind of, yeah, it really goes up, it goes up almost exponentially. So that was the thing that was like, so this guy did all this stuff on the truck tires and there was even a big difference in compounds in different truck tires. Cause apparently if you're driving truck tires in the Arctic, it really, dr I guess you live in Finland, you know, all winter long there, the, or even Canada as well, you know, they're probably, you know, lot of sub zero driving. Yeah. And,
00:33:58
Speaker
Those tire, the really hard tires, they consume a lot more fuel because the rolling resistance is so much higher. So I bike commute. Usually the last couple of winters have been super snowy, but historically there hasn't been a ton of snow in Toronto in the last maybe seven or eight years. And so I bike commute all all winter long and I've got, cover, this is for my nine to five job. And one of the sites that I cover is about 28 kilometers away. So it's a good little morning, afternoon sort of commute.
00:34:23
Speaker
And in the winter, I'm always slower. And I'm like, what is up with me? and I'm like, I was like, you know, I'm wearing a jacket. So I'm probably a little bit, I mean, I'm not but optimizing this for performance. This is just me wondering like, why am I, why is it taking me like five extra minutes in the winter to get to work? And sure, yeah, I'm wearing like a, maybe like a slightly bigger jacket. Maybe there's an aerodynamic cost, but I'm not going very fast. I'm maybe like 25, 28 kilometers an hour anyway, my commuter. But the cold could be a factor there as well.
00:34:50
Speaker
Yeah, so it's it's twofold. You've got the air density, which is you know probably it could be 10 or 20%, and then the rolling resistance, which is a smaller number, but that that can easily double um for for such cold temperatures. that That's what we were saying is we it was huge.
00:35:08
Speaker
That's wild. I know that that Ronan mentioned this in his in his results, but do you have any way of... uh calculating that or uh or of um of applying a correction factor like let's say if you have your you know your gp 5000s and you know that they roll and i should know this what are they like 11 watts or something based bicycle rolling resistance test yeah 10 something per tire uh in in ambient let's say like you know 20-ish degree conditions i'm not sure what he tests them under but they're indoors so i assume it's regular indoor temperatures and then if you go down to zero centigrade like what what's the what's the factor I mean, you don't need to tell me what the factor is. I'm just curious if you have something like that. we did. We did that. So it's published in that Escape Collective. I'm just looking for it here.
00:35:55
Speaker
So there is a correction factor. That's really cool. yeah Yeah, no, that that was the whole point was we didn't – I didn't want want to publish until we did this correction factor because without the correction factor, the baselines didn't repeat.
00:36:07
Speaker
And that's why your numbers were not – it would look like nonsense, right? That's what was telling them. This is like that's like the worst results I've ever seen. Ronan, you stick. ah Yeah, so that was the whole that that was the whole issue was why yeah, why if you want to give a bit of a timeline, like most of the testing was done in like a one week period, one or two weeks in April, then it was me.
00:36:32
Speaker
I wanted to get the whole We needed an automated system. Like you're not trying to do all of this stuff manually. It's just, a I mean, I guess if you if you want to take five months, it's fine. But our situation was that we knew that we were going to get loads more from teams getting bombarded with them and we needed a way to process them quickly. So this was kind of a test case. This was a case to to refine it. And then also um I'm not... um In this case, Ronan's... Yeah. Ronan's on a deadline. Well, no, actually, Ronan was on a deadline, so wasn't an issue. Ronan's on less of a deadline than the teams, and Ronan wasn't going to yank my money. He didn't have any money. There wasn't money. Anyway, that's a different story. But Ronan put a lot into this, and so it was...
00:37:19
Speaker
yeah Plus, that article has gotten in so much traction, I i imagine, then just based on what I'm seeing. yeah you you You folks got a lot of attention from that, maybe even even if it wasn't direct like money, a lot of attention through that through that work with him.
00:37:34
Speaker
Yes, no, I was that was was definitely glad glad i did it. But that was kind of the idea with it, though, was that the teams do it. This is how it goes. They test on Thursday. They want the result for Friday for doing the recon for the race on Saturday or Sunday. i mean, that's literally what is happening. I'm not like not exaggerating, in these days. The heat is on.
00:37:53
Speaker
Yeah, yeah. And so the whole thing here was with Ronan was that it would allow, we would have a data set here of which we could spend a long time on. Got it. There's two really interesting findings or interesting meaning surprising to me. That's definitely one of them. The other one is the pressure one. But again, before I go there, I just want to circle back to our signal to noise ratio because I had one final question for you there. No, but just I want to close the loop here on the thing about it. The reason I'm banging on about this temperature is that on this test, it was one to two watts. Ronan would have done that during the day. If you do that at 20 degrees during the day, it could have been sub one watt. Oh, I see. Ronan's a very meticulous tester.
00:38:36
Speaker
There's two things that made it very difficult. If you look at the temperature through the course of the day, it will it rises a lot in the morning and then it kind of levels off in the afternoon. and then once the sun goes down, it kind of just drops pretty steadily all night long up until the morning, in which it rises again a lot in the morning.
00:38:55
Speaker
So the worst times to test are early in the morning. For tire tests, yeah really for any test, and it's kind of early in the... Anytime the big temperature changes, it makes this stuff difficult.
00:39:09
Speaker
So you need to have that... tire temperature compensation model really solid. If the temperature isn't changing much, it doesn't make much of a difference That's really interesting because I remember, I'll start interrupting again, I remember like doing tests fairly early in the morning and just like not not being able to make sense of my numbers because my baseline's in the afternoon or not in the afternoon, let's say like late morning when the sun is high already in the summer versus like first thing when the sun's at the horizon, they did not agree.
Rolling Resistance & Tire Dynamics
00:39:38
Speaker
Yeah. And so one of the beauties of the set that Ronan did was that he's got these ones at night with the temperatures dropping and then he's got these ones in the morning with the temperatures rising with the very same set of tires and the very same set of road. And so it's like it's... it's a great It was a great data set for for like validating this tire because you know if you just come out and say, yeah, we think that you know it's four times more sensitive than people think, know you you you want something to back that up. and I do need to close the loop on on writing up that report, but it's a really good set of data for that. For the temperature finding, yeah?
00:40:16
Speaker
Yeah, for the temperature finding. Cool. Very cool. okay so and Signal to noise. Yeah, we're we're a little bit all over the place, listeners, but there's there's just it's just because this is such a, I think for me at least, this is such a rich topic of conversation. There's just so much, it's it's it is really bleeding edge stuff. And it's also, um there's there's a lot of,
00:40:38
Speaker
and Maybe I've been out of it for too long, but there there's there's there's some dissent in in in you know what you hear from from one ah camp to the next. So there's always a lot of interesting stuff to to dig into. So signal the noise. um You were saying that on smooth roads, you're around 20 watts for the pair of tires. You got up to maybe 60, 70. But what's the delta? Let's say you're on smooth tarmac because that's the the hardest maybe.
00:41:03
Speaker
the hardest maybe um ah case. What's the delta between like, you know, the the the one step in tire pressure to the next step in tire pressure on smooth tarmac? And I want to compare that to your, let's say, one to two watt noise.
00:41:17
Speaker
Yeah. So, and just to qualify here, just because I am selling stuff. It's The 1 to 2 watt noise here, like we we can we can do better. And and some of the people say is like it's from knowledge from other tests where to where it's under a watt. So what we really see in this one, even with the 1 to 2 watt noise, is that for the smooth roads, the um we see just a decrease we see a decrease in rolling resistance with increasing pressure. And we're talking here, some for some of the tires, it could be as high as...
00:41:53
Speaker
I'm just looking at now, probably 6 watts, 7 watts for this, is you know for Rona's case, this is going from 40 up to 75 psi. Okay,
00:42:05
Speaker
okay okay Yeah, so that's that is ah that's definitely a lot of signal to to to not as you know to ah not as much noise. Yeah, but and also here, just to so we can speak about the noise, but we also did it in a number of cases. We repeated it. Also, there's multiple tires, and the trend is consistent across everything. So you can see that it's just it's consistent. Like for this test, everything was consistently down and to the right. Yeah. Okay. We, we covered temperature, I think a really interesting finding and obviously relevant to anybody, not just doing tire testing, but doing, doing aero testing, like make sure that you are, you've got either John's ah correction factor plugged into whatever model you're using, or you are um testing when temperatures is quite stable. So I think I'm good with that. ah Let's talk about ah the tire pressure stuff. So, My journey with with rolling resistance started with bicycle rolling resistance and and that gentleman whose name I don't have left off the top of my head. Excellent, massive database attires on on indoor testing on a drum. Really great data set, often cited.
00:43:22
Speaker
And then, and his findings were generally, you know, higher pressure equals lower rolling resistance, which, you know, all of us running around on 23 millimeter tires, 10, 15 years ago, that's what we did. Whatever the rim could take, we pumped it up to the max and and rode that with the understanding that that was fastest. And then um I first heard about,
00:43:41
Speaker
this break point that we're going to talk about at which um higher pressure results in more rolling resistance. So the end result being there is a sweet spot for your system mass and your tire width. and based on all the, you know, the online calculators that are that are around the break point that they report.
00:44:01
Speaker
has led folks to ride progressively lower air pressures, especially as tires have gotten wider. But your findings in your testing with Ronan John seem to contradict that, at least to some extent. So I definitely want to have spend some time talking about that.
00:44:16
Speaker
Yeah, so what I'll say in regards to this is the model of which at least how I think of these tires now, and I've seen quite a lot of data, and i to me at least this this model holds up, is that The drums are very representative of very smooth roads or velodromes. And I think that if you test on very smooth roads or velodromes, that the higher pressures are just, the fate they do roll quicker.
00:44:48
Speaker
So this is bicycle rolling resistance testing you you feel is representative of those conditions? Yes, I think so. I think there is there is always a question of roller diameter that I think he uses I think it's Yarno's name. He uses, yeah, it's a great site. death right If you're interested at all, subscribe to the site. it's It's awesome. And it's like, it's also peanuts. It's like a big investment. 100%. Absolutely. If you're interested interested in this, give them, it's great reading. And yeah,
00:45:16
Speaker
And actually, I believe and recommend offhand because I'll tie into that in a moment. But first of all, he uses a one millimeter, a one meter drum. And you the ideal thing is a flat surface because the amount of deformation on a one meter drum, it's a slightly different deformation than you would get on a flat road. Yeah, the contact patch is a bit different, yeah.
00:45:38
Speaker
Yeah, I think you can say here that the results were indicative. i think the the actual roll in resistance might be a tiny bit lower on a flat road because the the deformation for given pressure will be a little bit less from a flat road. Intuitively, that makes sense. Yeah.
00:45:55
Speaker
Yeah. there's um There's some papers from, there's a guy in Milan who publishes some tire papers and he he basically shows all of those differences. They actually have a rig there that that will roll on a flat road.
00:46:09
Speaker
so But it's it's really solid bicycle roll resistance it's really easy to look at and understand. And what I'll ah say here to go into this, which is when you refer to sort of the hysteresis, at a certain point, when you get into rougher roads, this doesn't hold true.
00:46:30
Speaker
And you can see that lower pressures are quicker on rougher roads. That, I think, not controversial at all. And so the question is, the only things that are left in the balance then is how rough does the road need to be before you start, before you start, and it starts biasing towards lower pressures.
00:46:55
Speaker
And then the second more nuanced question is what's the sort of transition between that where you're getting sort of medium roughness and, you know, what does that transition look like between a decreasing slope with higher pressure versus suddenly you have some sort of optima here. I think what that breakpoint theory describes as is a very sort of sharp optima.
00:47:21
Speaker
Two separate things here from which I think this result is, or if I can say, There's, think, kind of two points of dispute, I think, in terms of interpreting tire results. And one would be about the effect of width just on rolling resistance on smooth surfaces. okay And then the second one then is this about the effect of hysteresis.
00:47:53
Speaker
And how basically how that transition occurs. Is it a sharp transition or is it a very shallow transition? And that's interesting because, because yeah, the kind of the way that I look at this stuff is is I try to look at it through ah a lens of utility. And so a sharp transition versus ah a gradual transition means different things for people.
00:48:17
Speaker
um for ideal tire pressures because if you have a sharp transition, you might err a little bit on the side of lower just to make sure that you don't exceed that point.
00:48:28
Speaker
Whereas if it's a shallow transition, then you can push it closer to that to the transition point and be more comfortable. That's the way that I would interpret it. Yeah, we've got the sharp and shallow transition, which we can call sort of one um discussion point. And then the other discussion point is about...
00:48:46
Speaker
how wider tires roll versus narrow tires roll just on smooth surfaces. At like equal pressures. Now, we're just going to, if we're going to look at widths or because, but you can't really fully divorce them, right?
00:49:00
Speaker
So let's let's let's let's unfold that, I think. Okay. Love it. Let's go. I'll even recommend here, if anyone's interested, there's a very good article on bicycle roll and resistance to where he runs a the GP5000s at, I think he runs four widths of GP5000s.
00:49:22
Speaker
across different pressures. And he also gives sort of spring rates for the tires. So he gives deflecction amounts of deflection for the tires.
00:49:35
Speaker
Just pure geometry. You imagine a circle and that circle touches the road. If you imagine at zero load, the road is exactly like tangent to the tire. There's no contact patch or the contact patch is infinitely small. The tire is just floating above the road. Now, once you start...
00:49:52
Speaker
pressing the road into the tire, you start getting a longer contact patch. And so as you move up, say 10 millimeters, your contact patch now might be 100 millimeters long.
00:50:06
Speaker
There's a tiny bit of that contact patch length is related to contact stiffness, but it's primarily just the geometric intersection between a line and a circle, yup if that makes sense. It'd be great to a visual here, but we don't.
00:50:21
Speaker
and So there's actually very... Maybe that'll the the the episode art will be a diagram of what a contact patch looks like. Yeah, yeah. Well, it's even quite a question of of exactly what they look like because around the They're not strictly oval, right? Yeah. Well, they're oval, but you've got this... um The tire has a little bit of stiffness, so right around the boundaries, they unload a tiny bit. So it's... You get it... And people love to... People like to argue about the case and stiffness stuff. But to be simple here, we can say that the length of that contact patch is very closely geometrically related to the amount that the tire is pressed into the road.
00:51:04
Speaker
The harder we press into the road, the longer that contact patch. And it's very much there's just a geometric formula, which is very close to that. Okay. And what I'll say is that the role and resistance of a given tire is very much proportional or basically is is a function for for a given tire is a function of both of either the length of that contact patch or the vertical deflection. And I'll say they're the same thing.
00:51:37
Speaker
Yeah, because they're correlated, yeah. Yeah, there's like a function between them. And so then you can say, let's say I deflect a given tire three millimeters. I get a certain roll and resistance.
00:51:50
Speaker
then if I increase the pressure, I've now reduced that deflection. For the same kind of load or weight or mass whatever you Yes, exactly. And so once you get into wider tires now, wider tires will have less deflection or they'll have a shorter contact patch for a given pressure.
00:52:10
Speaker
But the surface area of the contact patch is the same, right? Because it's still supporting the same load. So it's it's less long, but it's wider, no? It is, exactly. But there's a fundamental thing here. Now, this is car tire. If you pick up a car tire, an automotive tire, and you start bending around, it's got steel bands through it. It's got an extremely thick carcass. They have a very rectangular contact patch. yeah A bicycle tire, especially racing tires, they're very supple. mean, a lot of them are foldable.
00:52:44
Speaker
There is a casing stiffness to them. That casing stiffness is very small. and that the bulk of the force that holds up the tire comes from the tire pressure.
00:52:59
Speaker
Yes, okay.
00:53:03
Speaker
We've all ridden on a flat tire. There ain't much holding us up there. Exactly, exactly. And a huge huge contact patch, very slow. Not recommended. No. With wider tires, in this sense, if your contact patch doubles in width,
00:53:17
Speaker
what I'll call the the deflection will actually almost half. So if you were to go, let's say theoretically that you have a 25 mil tire and a 50 mil tire, and the 50 mil tire has got twice the contact patch width as the 25 mil tire, you've essentially doubled your width.
00:53:38
Speaker
That tire will then deflect half as much as the, um, as the narrower tire. And deflect you mean in the in the vertical direction? In vertical direction. So the contact patch is shorter, but that also means it's deflected. in The contact patch length is the vertical deflection.
00:53:57
Speaker
but Okay, except that they are um analogous in this conversation. Yes, that they're tied together. that you So if you have less vertical deflection for a given load, that's a harder, stiffer tire.
00:54:11
Speaker
So you've got the same pressure, but now, and just go out and try it. I mean, take a 23 mil tire at like 50 PSI. Take the big, the fattest mountain bike tire at 50 PSI. With one field. You will feel, you ride down the road, there is there's a huge difference in stiffness.
00:54:34
Speaker
Yes. And it's, and so for that reason, wider tires, for a given stiffness, need to run at a much lower pressure, and it's really almost proportional to the contact patch width.
00:54:49
Speaker
So wider tires allow you to run lower pressures. If you try to run a lower pressure on a narrower tire, you'll start bottoming out the w rim. Yes, yeah, there are those physical realities. And i would say here that you want to have a reasonable stiffness when you ride down the road.
Significant Data in Tire Testing
00:55:05
Speaker
a higher vertical stiffness especially on smooth roads will run quicker.
00:55:11
Speaker
You'll have less tire deflection, it runs quicker. And I think there isn't that much difference between like a 25 millimeter, you know, pumped up at like a higher pressure and then a 30 at a slightly lower pressure.
00:55:25
Speaker
The results are so similar. I mean, I would say, for instance, have have a look at this bicycle rolling resistance test at um on the GP5000. It's a really nice case study.
00:55:37
Speaker
Let me make sure that I understand. So this is how I make sure that I get it. I'll i'll parrot it back to you what you just said to make sure I got it. ah So if you control for, let's say, contact patch length or or the deflection, which we agree is you know ah similar enough, um regardless of what the width is and the the width of the tire and the pressure in the tire, your rolling resistance on smooth tarmac is going to be very similar.
00:56:02
Speaker
It's be very similar. And then what people will say, well, okay, well, what about the, look, this is for a given construction of the tire. so Okay, yes, also valid. Yeah, for sure. You know, if you change the construction of the tire, I'm saying for, like, I think the GP5000 test is nice because um it's it's i think it's, I think it's the same,
00:56:24
Speaker
um thickness going all the way through here. And one thing that's very important with this is to talk about like big numbers, small numbers. I really enjoy the big number, small number conversation. here It's so important because you get so many red herrings to where people say, oh, that's nonlinear. There's this. there You've got to think about this. And like those are small numbers. And here we're trying to figure out what's critical.
00:56:47
Speaker
And There are optimal tires. There are cases to where, yes, this tire, like ah definitely within a given series, this size tires a tiny bit quicker. But as you mentioned, signal to noise is that, or what's the noise of the experiment? And I think even, don't want to mispronounce now I think it's Yarno at Bicycle. He's really insistent, if you look at his documentation, I have like i have like a half watt noise and everything. So stop trying to quote things to like 0.2 of a watt. And... we don't we yeah We can't claim that resolution.
00:57:19
Speaker
But I think that trend is definitely there. they're on There are... non-linearities with the stiffness of the tire as you go wider that has to do with that case and stiffness it's primarily though they call this you can model the tire as a membrane to where you imagine that it's imagine almost like a balloon where there's almost no stiffness that's actually a pretty good model it is slightly non-linear there are Especially once you get into like, you know, six to seven bar and things of the tire start the case and stiffness starts to really pick up. It's complex because the the case is actually stiffer when it's inflated than it is when you just feel it when it's empty. Oh, because you're inducing some kind of internal stress in it and that's what's causing the... Okay, what I don't want speculate. I can't say I understand it either. It just gets a bit stiffer. I don't know.
00:58:09
Speaker
Exactly magic. um And interestingly, I think um you you folks, you and Ronan had a tire outlier, the 35 millimeter that had that had some results that were not not in line with what ah what you were saying and if what you were seeing with the other Wits. And if I remember correctly, Pirelli mentioned that this there was some potentially some construction differences in that tire.
00:58:33
Speaker
So with this series, yeah, we tested, I think, we tested 26, 28, 30, 32. So the when you go from 32 to 35, the tread thickness increases.
00:58:46
Speaker
And so we saw that both the 35 and the 40 are slower and that stuff shows up as well. And people who will say, well, why not keep the same thickness all the way through? And ah you know you can also say, right, well, there's a situation. if you blow If you blow up a balloon twice as big, suddenly you know it becomes a lot more fragile.
00:59:09
Speaker
And so pressure vessels, as they get bigger, they do need to be thicker skinned. So you can't have an extremely big tire with a really thin skin. There is an argument, the most if you want to get a really thin skinned tire, make a 19 again. Yeah.
00:59:24
Speaker
ah But then that that needs to run a very high pressure in order to not bottom out on the road. Okay. So I think, John, I think I'm with you ah on smooth tarmac. I think it's a pretty so pretty easy to understand relationship between contact patch and rolling resistance and how contact patch is affected by width and um and pressure.
00:59:51
Speaker
I think we can I can wrap my brains around that. But what happens when you start to get into less than perfect tarmac or then you know maybe even extending it eventually to loose surface?
01:00:04
Speaker
Okay. So I think uncontroversially, we can say things start shaking. Yes. And so I think that won't be disputed. Things start shaking and energy gets dissipated when things start shaking.
01:00:20
Speaker
and there's some split between is energy lost from from the tire movement when things start shaking or is energy lost from shaking the rider. Because we're a big mass spring mass system that's super complicated with all of our bits and squishy things inside our bodies.
01:00:39
Speaker
Yeah, so in this case it's it's damping losses and whether it's in the tire, whether it's in the body, it's some split in between and I don't know what that split is. i kind of have my ideas here. i'll keeping it myself for the moment or, you know, I don't, I think it's primarily the body, but I don't, I don't know that and that's, that's more anecdotal.
01:00:58
Speaker
um For whatever reason, once you start shaking, you start having this big energy losses come up. And so then the obvious thing or the thing of which people have been doing for,
01:01:10
Speaker
over a hundred years is you soften, you know, you soften the suspension, you introduce the spring. And so then you shake a bit less and there's less energy dissipated from that shaking.
01:01:21
Speaker
So pretty straightforward. And the easiest way to add a spring in there is is to drop the tire pressure. Exactly. that's that's the That's the most obvious one. Now we're getting into like gravel suspension forks, but that's that's totally outside of the scope of this conversation.
01:01:36
Speaker
yeah and so Yeah, so it's the whole idea of whether a fork or drop in a tire, then because the whole advantage, if the fork was just a spring and there's no damper in the nice because there'd be less energy loss than deforming the rubber, even though with a tire, it's primarily air deferent. Anyway, there are losses in vibrating a tire.
01:01:54
Speaker
That's pretty uncontroversial there. The bit there that gets controversial is about how the transition occurs, whether... you have these hysteresis losses, or let's just, yeah, we can just call them you know vibrational losses.
01:02:09
Speaker
How do they show up? what does What does that look like when you start changing across tire pressure? When you start increasing tire pressure, how does it, you know, is it a very sudden thing or is it a very mild thing?
01:02:23
Speaker
And i'm I'm very much in camp. It's a very mild thing. So it's not as if we disagree that there is more resistance to forward progress or more power that needs to be applied to keep the going the same speed when things start shaking because the surface is rough. It's whether or not there is that sharp break point that we talked about earlier or if it's if it's much more gradual. That's the that's the debate then.
01:02:47
Speaker
Yes, exactly. so And I think there's, to me at least, that there's there's two separate things. There's the first one which you covered, which is about the relationship between roll and resistance between wide tires and narrow tires, whether there's something inherently quicker about wider tires rolling quicker. and there's There's different camps and I'm very much in the camp that it's they're quite comparable. Once you start equalizing pressures that you end up with almost an equal spring rate and equal rolling resistance. There's tiny differences between that. I'm not saying they're not there, but there's less magic there than I think has been made out.
01:03:29
Speaker
So so that's the that's initially in the tire width thing. And then the second one we're on now, which is about... break point, which would be kind of a very sharp, distinct optima to where you've got rolling resistance is decreasing with increase in pressure to hit a his certain point. And then suddenly it starts increasing again.
01:03:52
Speaker
and um And then they the second possibility is that it's a much smoother um transition. Mm-hmm. And let me ask you, you mentioned testing with with a bunch of teams. Now, without obviously disclosing anything, you you cannot disclose. have you And you've done the tire testing with them as well, yeah? Yes. Are you seeing similar results? Like that's that's where I'm going with this, as as to what you were what what Ronan was able to publish. So just even a little bit of a history on this was we'd done quite a few tire tests before Ronan approached this. The problem was that the tire tests were getting more and more, they were getting bigger and bigger. And so Ronan's test wasn't even the biggest tire test. It's not our biggest, I mean, it's in it's it's not our biggest tire tar set.
01:04:36
Speaker
So I didn't know about all this break point stuff. And so I was very... um I think Ronan was bit more unnerved by the results than me. And so i I then kind of looked up this breakpoint stuff. And all I'd say here is... is We don't see it and we've we've done quite a lot and we've even done quite a lot since since this result came out. I've never seen it.
01:05:04
Speaker
And the reason why i think it's a smooth optima is that all the data we have shows a smooth optima. And there is the question of how...
01:05:18
Speaker
how finely you measure those pressures. So for instance, you know is it you know are you measuring every sort of five PSI or every point? And we generally don't. We generally will so will sweep usually on about like one bar increments. So about 15 PSI increments would be pretty typical. But what we see is that we generally don't see big differences between there. So it's not likely that there's a sharp break point somewhere in between your your ah pressure set points.
01:05:45
Speaker
No, I don't believe just because we're seeing very, we we generally see small differences. What we see is that when things get very rough, the lower the pressure, the better. When things are very smooth, the higher pressure, the better. For a lot of roads that we're on, it's it's pretty flat.
01:06:03
Speaker
And so therefore you... What do you mean by flat? In that you can vary the tire pressure quite a lot and only see like one or two watts of difference.
01:06:15
Speaker
Because I think what's happening is that you've got two competing things, is that you have the... this the tire deflection is reduced as you go higher pressure, but then it's offset by the vibrational forces.
01:06:32
Speaker
And so if somebody is a strong advocate of the breakpoint, one of there's two things I would ask for. One is just what's the mathematical model for that? Because in order for the breakpoint to happen, you need like a sudden onset of energy loss at a certain pressure. So you you have to have no energy loss from vibration at the lower pressures, and then you suddenly have to have the energy. it has to suddenly kick it has to kick in very suddenly in order to produce that graph.
01:07:03
Speaker
Yeah, and become the dominant sort of a cause for energy loss. Yeah, which I don't understand in that when you do the tire pressure sweeps and you look at the vibration, the vibration just increases pretty steadily with increase in tire pressure. So why would the resistance suddenly from vibration shoot up like that? Because it needs to, if the graph looks like a hockey stick, that's overlaid on top of a decrease in the graph. Of course. So it's got to be like a wall. So it has to be like no vibration effect and suddenly a vibration effect. Mm-hmm. So that's just on the theoretical side to say what would be the mechanism for that. I think people, you could cite something like resonance for it, but I think it's... But wouldn't resonance be strongly like speed dependent? And then...
01:07:51
Speaker
Well, first of all, this stuff is speed dependent, but I don't buy, i I don't believe the resonance one just because you don't, i nobody experiences anything like resonance. You could tell. you would We've all had like, you know, ah speed wobbles. and Like we we we kind of sort of know what what resonance feels like.
01:08:11
Speaker
Yeah. And and the there there isn't, there isn't you you don't feel it like that. when I mean, you you just do the test yourself. Just just do low pressure and a high pressure on a road. And it just feels the higher pressure just a bit bumpier. Yeah. The only I can think of is like sometimes you hit a washboard section and on some like really bad gravel and then you might get some resonance effects there. But it's not like that washboard section is your whole ride.
Data Transparency & Conclusion
01:08:31
Speaker
That would be terrible. I would not recommend that to anyone. Yeah, so like like cobbles are an example where there's something. Or cobbles, yeah. Yeah, so there's definitely some nonlinear stuff going on there. And it's like, I'm not saying I have the complete picture, but I haven't seen the breakpoint in any of the data we do. And all I would say here is that we we collect a lot of data it
01:08:57
Speaker
I believe we've got ah far more accurate, we've got far better resolution than anybody who's tested before does. And I mean, we published this test with Ronin, which was done at night, which you said was kind of one to two watts.
01:09:15
Speaker
ah we'll We'll publish another test. We did a test with some on some gravel. I think we've got tests. We published a couple of tests on our site now with just a few athletes. And those are like one one watt tests on windy days.
01:09:29
Speaker
ah And we're publishing these tests. they They have repeats in them. You can see the repeats. All it would challenge is somebody says, okay, we saw the breakpoint. like Show us your testing methodology and show us how your repeats were for that day because we we're trying to be pretty transparent here. And like show I'd like to see these breakpoints repeated.
01:09:50
Speaker
I just haven't seen them. you know I don't understand the theory behind them and I haven't seen it the data. So as far as I'm concerned, you know the data we published on this escape collective, I sleep completely easy with that one.
01:10:04
Speaker
To me, it's like the burden of proof is on the breakpoint, not on what we're doing. Got it. John, this has been an awesome episode. As always, thank you so much for your time. If ah people want to learn more and if they, you know, if they're ah like ah a coach or a team who wants to inquire about your services, what is the best way for them to get in touch?
01:10:24
Speaker
um Yeah, I think you you can go on our site and there's an inquiry if you or just info at streamlines.aero. If you're a team or a manufacturer, yeah, get in touch. we're um We've got so basically you've got packages ready on that. If you're a fitter, we can We're not really supplying fitters right now. We're piloting some programs and we're hoping to get have a good system um that's economical for fitters to run.
01:10:53
Speaker
We're hoping for sort of Q3 of this year. and And then finally, like we do sell body sensors and speed sensors and the speed sensor, which we just did we actually made it for our own internal use just because... we had a lot of problems with existing speed sensors and we knew we could do something quite a bit more accurate. And that's been a surprise seller for us.
01:11:17
Speaker
Yeah. And I've used your former sensor. It's very cool. It's like it, it takes a lot of guesswork out of, out of position. I'm a, I'm a big fan. So that's one for me. Yeah. um Listeners as always, thank you very much for listening. I'm hoping to have a cadence of about an episode a month. I think that's, that's realistic now. So that's my commitment to you all.
01:11:38
Speaker
Thanks again.