AJA #45: Pump Accuracy & Cheap Carbon Parts

Hear why that “psst” when you pull the chuck is just hose air (not lost PSI), how to set smarter pressures for tandems and “big-butt” weight splits using a bathroom scale, and what tread actually grips on loose-over-hard. Josh also demystifies analog vs. digital gauge accuracy (and drift) and tackles the carbon-on-the-cheap question—when a bargain is fine, and when it’s a dentist bill waiting to happen.

TABLE OF CONTENTS

1:09 Understanding Air Loss with Presta Valves

4:54 Shimano's New Cleat Design

6:03 Variability in Silca Pump Readings

12:20 Exploring Gauge Accuracy in Pumps

28:07 Tandem Tire Pressure Challenges

34:42 Weight Distribution and Tire Pressure

45:35 Optimizing Tread Patterns for Grip

51:39 The Risks of Non-Mainstream Carbon Parts

 

In this episode of the Marginal Gains Podcast, we sit down with Silca's head honcho, Josh Portner, for an engaging "Ask Josh Anything" session. Our conversation dives deep into the intricacies of tire pressure management and pump accuracy, alongside some thrilling questions from our listeners. As an avid user of Silca products, I can't help but geek out over the Hero Chuck and my beloved collection of Silca pumps, including the Super Pista Ultimate, the Pista Digital, and more.

We kick things off with a question about air loss when disconnecting the Hero Chuck from a Presta valve. Josh clarifies that the sound of air escaping is just “hose air” and assures me that there's negligible PSI loss when disconnecting; what I've been hearing is simply the pressure release from the pump's hose. He elaborates on the mechanics of how Silca’s chucks function, emphasizing the design that ensures no air escapes from the tire itself when disconnected, reaffirming the precision engineering that goes into Silca pumps.

Moving forward, we discuss the variations in pressure readings across different pump models. Josh explains how manufacturing tolerances and gauge types contribute to these discrepancies. For instance, the Super Pista Ultimate features a Bordon tube gauge that aims for high accuracy typically within 1% of full scale. Our conversation unpacks the complexities of mechanical versus digital gauges and the importance of accurate calibration in providing reliable readings. As Josh connects the dots, he reveals how even the calibration procedures are designed to ensure accuracy where it matters most for users, particularly in the mid-range of use.

Next, we venture into the practical realm of pressure calculators, specifically for tandem riders. Josh addresses whether typical calculators can effectively determine optimal tire pressure for tandem riders due to the lack of data for dual-rider weight. His insights into the innovative design of the Silca calculator highlight how it leverages actual historical data to create helpful pressure recommendations—while acknowledging that some extrapolation may occur when it comes to atypical setups like tandems.

We receive some fascinating listener questions on grip optimization for different bike terrains including gravel and hardpack conditions. Josh dives into the science of tire tread patterns, stressing that there is no one-size-fits-all approach and reviewing the impact of various designs on grip versus rolling resistance—fun facts that can significantly enhance riding performance.

Furthermore, we tackle the subject of wear patterns on tires, notably the impact of bike handling and weight distribution on tire longevity. Josh educates us on the importance of testing individual riding positions and how certain riding styles can lead to inflated pressure and subsequent tire wear, offering a simple yet effective method for listeners to gauge their weight distribution on their bikes.

Lastly, the conversation touches on the sometimes contentious topic of carbon fiber components produced outside mainstream manufacturers. Josh provides a candid overview of the risks associated with purchasing cheaper alternatives, explaining how most reputable carbon fiber products come from factories known for quality control. He advises caution when it comes to prices that seem too good to be true, encouraging listeners to seek out products from credible factories that also manufacture for established brands.

TRANSCRIPT

Fatty:

[0:03] The marginal gains podcast is presented by Silca makers of the finest pumps waxing systems and other lubes cleaning systems and tools i'm fatty and i think i may actually be the uh Silca pump greatest fan or at least at least greatest consumer uh at least more than any non-employee or bike shop more on that shortly go to silka.cc to try to catch up to me if you can this is an ask josh anything episode of the show and we've got silka head honcho josh portner with us hi josh hey guys good to see you before i even let hottie step up to the mic i'm gonna line jump because otherwise i'm gonna get the hottie stink eye maybe getting it right now as you know i am a big fan of the Hero Chuck,

Fatty:

[0:57] which I use with both my Super Pista Ultimate and my Pista Digital.

Fatty:

[1:02] And I got to ask, how much am I losing when I unlock the chuck and pull that off the valve? Is there a best practice for minimizing that loss or at least making sure that it is a consistent amount? I'm always worrying when I hear that little pssst and pull off the chuck that I'm going from a, you know, 80 PSI to 78?

Josh:

[1:28] Yeah, that's a great question. We actually get this from consumers all the time and you're going to love the answer. It's you're losing nothing.

Fatty:

[1:36] So, that is just the sound of air coming out of the hose.

Josh:

[1:39] That is hose air. That is hose air that you're hearing. Yeah. None of the Silke Chucks hold open the Presta valve core when they're installed. So, there's kind of two schools of thought out there.

Josh:

[1:52] And I would say some brands, they intentionally, they bottom out on the top of the core and hold it open while you pump, which is like a Schrader valve.

Josh:

[2:00] And they do that because a lot of times, you know, we've all done this. Like your valve core is maybe a little bit sticky, you put the pump on and those first couple of pumps, you know, it goes to like 150 and then finally it clicks and you're, oh, it's actually only 60 PSI.

Josh:

[2:16] Some companies intentionally have the chuck push that open to prevent that overpressure because that overpressure on a stuck valve is the number one cause of gauge error and gauge failure because gauges aren't meant to be maxed out like that. Um, no silka chuck does that because the, the way a typical press is meant to operate, the air flows in and then the slightest little bit of backflow happens. Uh, it's just like a little pressure wave and that pushes the valve closed, um, with each pump. And so that's why when you're pumping with your silka pump, you get like a little click with each stroke and that's the pressure, the press to valve opening and closing, um, each time. So, yeah, when you release your Hero truck, that Presta valve is in its closed position. And the only air that comes out is what was trapped in the hose. And then there is like a circuit inside the pump itself that, you know, is open to the gauge and then runs back to the check valve, which lives kind of directly under the bottom of the piston there. So, you know, there's, it's not a huge volume of air, but, you know, you think that's a pretty hoses over a meter long, plus the little, the gauge air, plus that little circuit in the bottom of the pump.

Fatty:

[3:35] So, in the end, I know that that little sound is not me dropping back 2 PSI and that I need to overcorrect by 2 PSI or something like that. That is just air coming out of the pump.

Josh:

[3:47] That is just air from the pump. Yep. Whatever it reads when you stop pumping, that's what's in the tire. And, yep, you're good to go. And I was like, because someone will call it out. We've actually measured it. It's like hundreds of a PSI. Lower than the air that's in the tire, right? There needs to be a slight pressure differential for that valve to be pushed back closed. And so you get this sort of like wave form, you know, the air flows in and then it starts to, the pressure, you know, essentially comes up in the tire and then it starts to come backwards. And so, yeah, your hose air is probably, you know, it's less than a tenth of a PSI. It's essentially immeasurable. But yeah, Yeah, a few hundredths of a PSI lower in the hose than it is in the actual tire. But that little differential is what's allowed that valve to close and stay closed.

Hottie:

[4:40] Okay, that wasn't bad. We're only a few minutes into the show, Fatty. Well done. And I get to talk.

Josh:

[4:45] It's a good strategy for line jumping. I mean, you just don't even let the guy talk. Just go right at it.

Hottie:

[4:50] Don't even just shut his mic off. Don't let him talk until it's really his turn.

Hottie:

[4:55] I've been muted. This is the 45th edition of Ask Josh Anything of Marginal Gains. I am Michael Hutton, also known as Hottie. This show also supported by Shimano. Shimano's mountain bike cleats remained unchanged for almost 30 years. So it was a huge surprise when Shimano dropped a new take on its venerable off-road clip-in part. The new cleat has three ways it can be engaged. the way it's always worked, toe first. It also clips in from the backside or heel first. And the third and most exciting way, a rider can simply stomp straight down on the pedal and be clipped in. Holy cycle cross heaven, Batman. The new cleat is backwards compatible with all Shimano mountain bike pedals. That's great news. Or you can call those gravel pedals, by the way, whatever you want to call them. It's fine. Off-road pedals. If you need a fresh pair of cleats and you want that three-way engagement, look for part number CL-MT001.

Hottie:

[5:58] To see all the ways you can get clipped in, head to bike.shimano.com.

Fatty:

[6:03] So add read follow-up on that, Josh, is there going to be a Silca cleat that has essentially similar dimensions, but in titanium?

Josh:

[6:14] You know, we've actually got a set of these on order to take a look at them and try them out and see what we think. We'll have to see what the customer demand is for those.

Fatty:

[6:27] Speaking of ultimate and and pumps that I was talking about just a minute ago, I told you or hinted that I have quite a collection of Silca pumps. So my favorite pump, of course, the super piece to ultimate lives in my garage, which I use essentially every single day. I have the Pista Digital. It lives in the cab of my truck, and I use it essentially anytime I am starting a ride after driving somewhere. My Terra comes into play during fat bike season. It lives in my truck then because it is great for measuring low pressure, which, you know, I'm generally in my fat bike. I'm looking at four PSI or less. And my Electrico Ultimate lives in my traveling tool bag, which Hottie would have liked to know if he had Leadville when he could have checked the tire pressure on his Lexus. We talked about that last episode. Anyway, of course, I have all these pumps and I'm not even talking about my mini pumps yet. But there is a small amount of variability between the reading of each of these pumps on a given tire. And, Josh, I want to know, is that because I lose a little pressure when I connect or disconnect? Or is there a little bit of wear between, you know, or different amounts of wear on the parts of these pumps? Or is it something else?

Josh:

[7:51] That's a great question. So, yeah, you know, the different types of gauges all have slightly different behaviors from each other. And then you have the slight, you know, manufacturing difference, tolerance difference between them. And so I think, you know, we try to simplify that and industry has tried to simplify that with sort of this percent accuracy thing that we like to talk about. And, you know, Soka's, we're famous for speccing the most accurate, highest precision specs for a given price point of a pump. But then there's still slight variations and differences between them. So, you know, think about... We'll start with your ultimate pump. So ultimate pump is a plus minus 1% accurate gauge, and that's 1% of full scale. So that gauge goes to 160. So the total plus minus on that, your plus or minus 1.6 PSI.

Josh:

[8:51] But that's inclusive of slight nonlinearities in the flex of – inside there, there's what's called a Bordon tube, which looks like a, it's like a spiral of metal, copper in this, or it's brass actually in this one, a spiral of brass that pulls a little rack gear across a pinion, and then that pinion is on the shaft that shares the needle. And so, you know, it's like if you ever, you know, had your garden hose coiled up and you turn the water pressure on and it moves, right? That's the same effect that a Bordon tube has, that when you put air pressure into this spiral curled up thing, it wants to go straight. And so we're taking advantage of the end of that to draw a little rack gear across a pinion gear. And that's how most high-end gauges are made. It's expensive.

Josh:

[9:48] There's another one called a diaphragm style gauge. It's a little bit less expensive and a little bit less accurate. But in this case, it's Bordon tube. And so, you know, when they're specced at the plus minus 1%, that is that, you know, you're controlling the wall thickness and the accuracy and the weight and all of the manufacturing details of that Bordon tube curly cue within certain parameters. And then you're controlling similarly the accuracy and the fit of that rack gear and of that pinion gear. But you still, as the Bordon tube sees pressure and grows, it grows slightly non-linearly. And so like in that pump in particular, the way we solve it is that you actually, you know, the gauge is manufactured and it's a 160 PSI gauge. And so at the very end of it, the final assembly, we haven't yet put the needle on, right? So the shaft is there with the little rack and pinion gear and all the parts are together, but there's no lens and there's no needle. And they put it on a, you know, like 0.1% accurate gauge system and pump it to 80. And then we put the needle on with the needle at 80. And so in that case, like we're.

Josh:

[11:04] Of controlling the accuracy in a sense to say, well, most people are probably using this pump somewhere around the middle of its range. And so we're going to go for as perfectly accurate as we can at that one, and then let, we're going to allow the inaccuracy to fall at the ends. So, you know, that pump in particular is least accurate at zero and at 160, which makes it, you know, not the best pump then probably for your fat bike, right? Because we're, and so, and so you have to do this really with all of these technologies. And so you have to make a decision. That was our way of, well, our, we learned it from our vendor who said, hey, there's a bunch, you know, you, you can put the needle on it zero and let, let the inaccuracies build. You know, it's a lot more expensive to do it the way we're doing it because, you know, you can imagine if, if you're placing the needle at zero, you're just placing the needle at zero and you're never attaching it to anything. You're not pressurizing it. You're not, you know, you don't need special fixturing to get the needle perfectly aligned with 80.

Josh:

[12:07] And so, you know, that was a choice that we made to get the accuracy of that pump as good as possible in the range where customers are using it. You know, the kind of flip side of that would be the Terra, which is the least expensive of the Silka pumps.

Josh:

[12:21] And it has a nonlinear rack and pinion gear in it that, you know, essentially allows it to have multiple rates. So, you know, at low pressures, right, you have very large graduations between single PSI. And then as it goes up to higher pressures, those graduations become smaller. And so, you know, in that one, we do the same needle trick where you start at zero. That one goes zero to 120. But the midway point of that gauge is only 30, so the first half of the swing, right, is a quarter of the full scale. And so like in that one, we do the same thing and we put the needle on at 30, and so that's biasing it again kind of towards the middle, and you're allowing your inaccuracies to creep.

Josh:

[13:15] In at the extremes, in this case zero and 120. Now, digital gauges are a whole other really interesting thing that we kind of talked about in our last episode. You know, with an analog gauge, you are always getting the pressure differential, right, to atmosphere. So, we talked about, you know, is it absolute or relative pressure that you want? And a mechanical gauge is always, by nature, it's going to be relative. It's that needle's pointing at zero, and when the pressure's, you know, when there's no pressure in it, you take it up to altitude, it doesn't read any pressure, right? It's still at zero.

Josh:

[13:56] With digital gauges, there's different sensitivities, and, you know, they actually can, they can sense an external pressure change because when you turn it on, right, and it's now sensing force on the little, they're typically like, sometimes they're membrane, but sometimes they're a little like piezo. Ceramic, it can sense that force. And so typically for the lesser expensive ones, you have to set a zero somewhere, right? That little piezo ceramic is, you're putting a voltage to it, and then essentially reading the voltage that it gives you back, right? So you've got electricity connected to it, there's a little bit of signal going in, and then what's the signal coming out? And then you can use that differential to calculate air pressure. Those are typically set at the factory to a factory zero, which means that that zero is now zero. So that's why when you take some gauges and pumps to altitude, they won't read that initial pressure as zero because they're fixed. They have essentially fixed zero. You can do like what we did with the electrical ultimate, where you allow it to kind of reset its own zero every time and then go from there, right? And that requires a little bit of logic.

Josh:

[15:15] But the other thing you have to decide when working with electric types of gauges is they're also nonlinear. And the way the electronics think, you are linearizing that change in voltage in the system. And so if you kind of think of it, typically they look sort of like think of like a 45-degree line and now bend it down so it's got a little bit of a saddle shape, right? Like, you know, like if you were to hang a, you know, a power cord or something at 45 degrees, it wouldn't be straight. It has like a sag to it. And so that's going to be your output relative to pressure for the sensor. And then you're going to put a line to that. And, you know, essentially the way we've chosen to do ours is –, take a line that's basically a good curve fit in terms of, you know, say it's a 45 degree angle from, you know, zero to max pressure. So we would linearize that with a 45 degree line. And then you have to decide, well, do we want it most accurate at zero and at 200? Or do we want to move that line, right? Or do we even want to, do we want to change the slope of that line? Like if we change the slope of that line, we can make it much more accurate at the low pressure zone.

Josh:

[16:34] But then it's going to be much less accurate at the high pressure zone or do we shift the line, relative to that saddle shape so that you now say have two places on the line where it's going to be more or less perfectly accurate but then it's going to be, you know high at the low and very high pressures but it's going to read low in the middle range pressures right and so these are all decisions that have to be made when you're, you know designing for these things and so like in the case of the digital we've offset the line and tweak the slope slightly so that you are essentially, it crosses.

Josh:

[17:12] I'm trying to think of how to use words for this because I have a picture in my head. The slope is tweaked a little bit so that you're crossing at a low pressure and at a middle pressure. And then we're allowing the bulk of the inaccuracy to come at pressures over 100 PSI.

Josh:

[17:29] Because I think, you know, if like at your very low pressures and the other challenge you have with the digital pump, which is why you don't use it with your fat bike, is there's voltage sensitivity there. And so like that pump won't even read pressures less than about five or six PSI because it's, there's, there's just not enough change in the voltage of that pump, which is also why, you know, it's not, it's why you have to hard zero that particular one because it. It can't read that two PSI difference that you're going to get when you go from sea level to Leadville, right? There's there's not enough change. And so it's just like it turns on and it's like I'm at zero, you know, no matter what, no matter what that is. So, yeah, it's crazy. You know, people and we even see the magazines and the, you know, these YouTube channels do it. Oh, we're going to measure all of these against this, you know, $24 Topeak digital gauge because it's digital. And you're like, oh, God, that's people want to assume that digital means hyper accurate or, you know, and hyper precise. And that's not necessarily true. You know, I think the majority these days of the hyper accurate, hyper precise gauges are digital.

Josh:

[18:44] But it doesn't that doesn't go the other way around necessarily and so you know you can have like in the case of that you know a 24 digital gauge that you buy off the internet's probably the same three plus minus three or plus minus five percent accuracy that you know you are going to get from a lot of inexpensive mechanical gauges and then the the real question is what strategy did that company choose to linearize their data to? Because that's going to control where the thing is off, where the thing is most inaccurate. And so, you know, as an engineer, we have to learn this really painful, probably the least favorite thing I ever had to learn in engineering school, but it's called significant digits. And it just kind of breaks your brain and it's just not what you want to hear but it's the idea that like okay you know if your digital calipers go to you know like all the digital calipers we use here are Mitutoyo and they go to the hundreds place and.

Josh:

[19:49] But because of the variance in the caliper is within that hundreds place, it's like you essentially have to throw that digit out, right? So, you know, like, I mean, my customer service team or sales team will borrow calipers and they're like, you know, like, oh, that this seat post is actually 27.15. And you're like, well, you know, and then like, oh, but this one's, you know, 27.20. And you're like, yeah, technically those are the same, right? Like, I know that in our brains, we want to, you know, like to really know, be able to read that last place, you really need something that goes a place further. And then when you start, like, using multiple gauges in a system, those compound on each other. So, you know, essentially, you end up having to throw away multiple. And there's certain systems where, you know, you might have two or three sensors all feeding into one value. And you're like, oh, man, this sensor goes to the, you know, thousands place, this one goes to the hundreds place, this one only goes to the tens place. And in the end, you're like, yep, that you've got to throw all that out. And that's just your answer is a round number. You're like, but they all have decimals, you know?

Josh:

[21:06] Yeah, it's just terribly annoying. And but, you know, these a lot of these digital gauges, you know, I have 27.1 PSI in there. You're like, yeah, but it's plus or minus 5% accuracy on a 200 PSI scale. Like that's, you know, depending on how it was put together, you're off by PSI, right? I mean, you know, you're, you know, if it's a 5P or, you know, let's call it plus minus 5 and you're at a hundred, I mean, you know, you're really more like, depending on how it was linearized you're really more like well gosh that's it's not a it's not 100.1 it's like 95.1 to 105.1 or and that's my hardcore significant digits people are going to say that's not how the math works and it's not quite that simple but for conversational purposes it is kind of like it is like that so I think yeah there's there's a comfort in an assumption people have when they they see that decimal place being used but it it technically actually is just misleading you or tricking you into believing something that really probably isn't true so.

Fatty:

[22:20] As my pumps get older does accuracy degrade or does it stay the same my super piece to ultimate is i think 12 years old now. And I mean, it still looks essentially new. I mean, the thing is probably literally bomb proof, but I mean, it seems like it could become less accurate as time goes on.

Josh:

[22:43] A hundred percent. Yeah. And so what happens in mechanical gauges is that because you're putting this pressure in and you're flexing these internal components, you're subject to this thing called strain hardening. And strain hardening is what happens, you know, when you go to bend a wire coat hanger, right? And then you go to bend it back and it's a lot harder to bend it back than it was to bend it in the first place. That's a strain harden. And, you know, that's a pretty extreme example of it. But, you know, you think of, you know, thousands of, you know, flexes of this spiral over and over and over again, you are ever so gently strain hardening that material. And so over time, it is now taking more force to deflect the same amount.

Josh:

[23:33] And so it's actually going to start to read under the pressure that you're actually putting in it, right? Because you're going to have to put more pressure in it to get it to flex the same. And, It's even worse with diaphragm-style gauges where you have essentially, you know, a big diaphragm, like you think of like the head of a drum, and the air pressure is flexing that up and down. And then those, they push sort of a spiral staircase-looking thing that spins the needle. So those tend to have more deflection and over a big surface area. And so those tend to strain harden actually a little bit faster than board-on tube style gauges. And then that's before you even get to, if you're seeing tons and tons of use, you have this little rack and pinion gear in there. And, you know, anybody who's ever, you know, worked on old cars or anything like that or knows, those develop play or backlash, as we call it, over time as they wear, right? Because nothing wears itself bigger. Those surfaces all wear themselves smaller. Um, yeah, and so that, I'm not aware, and I actually haven't, I'm not aware of a mechanism that would cause a digital gauge to drift over time, but if it were like a membrane, it probably, there's probably some element of something similar. If it's like a piezo ceramic, there's, it probably isn't. It should be a good one.

Hottie:

[25:01] Well, how about sealant contamination? Will that cause it to drift or be inaccurate?

Josh:

[25:05] Not – no. Sealant contamination is much more likely to just plug it up and make it not read at all. And so, I guess, in a sense, that's kind of good news. You know, it's like if sealant contamination were making it all of a sudden, you know, read really low and you're like, oh, man, it's – you know, 20 PSI is actually 100. That would not be good. But yeah that's one where typically especially on these board on tubes I mean you're you know you're trying not to you're trying not to suck a ton of air volume out of the system, you know with any of these any of these gauges and so you know you're talking like, sub-millimeter diameter hole that's running through the center of this thing. I mean, very, very small kind of orifice. So, yeah, that's where the sealant, if it does get blown in there, it can be pretty quick to seal because it's very small orifice.

Hottie:

[26:04] Well, we're having so much fun here with tire pressure and pressure gauges. Let's continue with a new kind of take on tire pressure and the Silca calculator, Josh. Actually, we have a couple of questions regarding tandems. Now, I, several months ago, got one onto the Hottie Hotline here at Silco West and came from a friend of mine, Bob. He rides a tandem with his partner, Sarah, two great people. They have a nice titanium tandem, 29-incher mountain bike style. He estimated their total weight at about 350 pounds. He runs 2.1 mountain bike tires on it, But he said that the tire pressure he got was pretty low, like 21-21 front and rear. Now, I ran the same stuff, and I got about 25-27. So maybe he put too rough a road in there, rough conditions in there, or something like that for what he was doing. I checked with other calculators, too, and they got about 32-34 front and rear for a tandem with a total weight package of 350 pounds. We also got a question from Jared in Davis about a more general question about tandems and pressure calculators. And he just asked plainly, do calculators work for tandems?

Hottie:

[27:21] Does the reference range of the data behind the pressure model include double rider weight on effectively the same rolling gear? So, Josh, what should we know about tandems and tire pressure calculators?

Josh:

[27:35] Oh, that's a fun one. So, yeah, what makes our calculator unique and awesome is that it's essentially not a linearization, but a curve fit of real-world... Calculations, like real-world optimizations that we've done in the field. I mean, thousands, thousands of them. And so, you know, essentially you think of like if, you know, every year we go to and we do all the riders at Flanders, and then we do all

Josh:

[28:05] those same riders at Roubaix and across a couple of teams. And so we've been calculating that for, you know, years and years, and we're putting that data in. And we're doing it with all of our gravel athletes. I think there's 18 of them or something. And, you know, we've done that at Unbound for years and years and years. And so, you know, you can imagine those optimizations all start to become sort of a scatter plot of, you know, this weight, this tire width, this surface. And essentially what the calculator's doing is connecting, you know, like, okay, here's all the data for this surface and all the weights and then all the sizes. And then I'm going to, you know, essentially interpolate where the gaps are, to try to predict what the optimal pressure, you know, would be for the data points that we are in the holes. And, you know, the challenge with something like a tandem is, you know.

Josh:

[28:59] The core calculator is curve fitting similar circumstances, say, across weight or, you know, it's actually, it's kind of a cool thing when you get into the depths of it, you can essentially, you know take a rider on a tire size and like you'd be like okay well let's let's now calculate optimal surface roughness for this one and you can do some pretty cool stuff but uh.

Josh:

[29:21] We're interpolating within the data we have. And so as you get to something like a tandem, we don't have data out there. And so you're taking the end of a curve that exists within this cluster of known real-world data, and now you're just extrapolating it kind of off into the nethers, right? Out into places that we've never been or seen, and we don't have any data out there. And so, you know, it become – and all of these things, when you look at them, they're nonlinear, right? I think the curves end up being like fifth or sixth order. I mean, they're fairly complicated curves. And so, there's a big kind of black hole of knowledge out there. Like, well, does something change? Does it need to be a seventh order, right? Does it bend in a different direction? Like, you know, if it's generally, you know, increasing in its slope, does it keep increasing? Does it flatten? Does it bend the other way? We don't know.

Josh:

[30:25] The other calculators out there are all based off of these really old, I think they're 70s. I think they came back in the 2000s, or maybe 2000s, but they're called the Berto charts. And what they do is calculate what's called the 15% drop of a tire on a flat surface. And so, you know, 15% drop is essentially, you know, if the, you know, tires, you know, X millimeters tall at what load on a flat surface, is it, is that height reduced by 15%? And then that's the point that we're looking at. And, you know, and that's something that's actually, when you look at the data, it's pretty linear. And so it's very easy to linearize because it's already linear. And so, but again, all of the Berto stuff done way, way back, you know, was never done around –.

Josh:

[31:17] There was no efficiency testing being done. It was kind of being done as an optimal, like, oh, this 15% seems to be an ideal balance of comfort, rolling, feel, cornering ability, and so 15 is the number. Um, but there was no, you know, I, I've actually, we've kind of joked about building a, uh, sadly, we actually had it quoted. It was like $250,000, uh, uh, rolling resistance tester to, could look at this, um, to say, oh, does it, you know, could, could that kind of be simplified to, you know, is it 14, is it 16, is it something else? And then what you find on top of that, or certainly what we found in the real world is like our calculator tends to, we actually kind of synthetically bump the front pressure a little bit.

Josh:

[32:12] Because what you find is in fast descending, especially in the mountains, the 15% drop pressure relative to your weight on a flat surface becomes way too low and feels terrible. And so, yeah, like if you go to look at the Berto charts or, you know, the zip calculator or something and you put in your actual weight distribution, you know, which for a lot of riders is like 60% rear, 40% front, that 40 front is like that is not a pressure that you want to descend on um and and so we bump that up all of that to say um yeah i mean you know all calculators ours included are you know our starting points and if you you know don't like it you definitely change it um you know but but certainly for us um because it's built around such a, you know, really tight cluster of, yeah, and the, like 90 plus percent of, of these were done with pro cyclists, right? So, I mean, we're, we're talking a weight range of really probably a hundred pounds to a hundred and seventy pounds, right? I mean, that's it. It's, it's a pretty narrow range of people.

Josh:

[33:24] You know, we've talked to about, like, we, we know our, if anything, our calculator for road like road gravel pressures is probably slightly over what's optimal for most people because our our riders to create our data set are all incredibly low body fat right and you know because we're predicting that transition to hysteresis well the transition to history from you know casing loss to rider hysteresis loss will happen later in a very low body fat person than it does in a higher body fat person. But again, I just don't have a lot of that data. And so, yes, I would say if you didn't like what you got from us, see what the other guys are saying. And, you know, the answer might likely be something in between. I would say we tend to find.

Josh:

[34:21] That we're a little bit higher, I think, at road, kind of in that road pressure range than a lot of those calculators, but then we seem like we're a little bit lower in the very large gravel mountain tire range. I honestly have no idea what we are at the very high rider weight range.

Josh:

[34:43] And then the last thing I'll say with that is, you know, the most common, because we get so many of these literally every day, I think that, Pressure calculator runs like 100,000 calculations a month. I mean, it's bonkers. Like how many, how much you say, oh, it's crazy. It's crazy.

Josh:

[35:00] Yeah, actually, it makes our, when like consultants look at our web data, they're like, oh my God, you know, everybody at first is like, oh my God, you're, you know, your website performs terribly. You have all this traffic and, you know, and it's all to this one page and then people bounce like, yeah, that's the pressure calculator. Like it you know because we've said like we it's there as a service right like you know we don't there's no product link on it there's no thing to buy there's you know that's not what people want so um but the number one thing that happens when people are getting low measurements is typically that they're using sidewall um size and not measured size and and because we're the only calculator out there that wants an actual measurement. And the reason we do that, I think we covered it in some of the hookless stuff, is that, you know, this company's 2.1 is not the same as that company's 2.1. And then depending on what bead width rim you put it on, or if it's hooked versus hookless, or then that changes it even further. And even down to like different tires within a company, I think we talked about Continental last time, you know, depending on when that tire was released.

Josh:

[36:12] You know, they've changed over the years what their base inner bead width assumption is. So, you know, if that's a model of tire that's been around for 10 years, that 2.1 was based on a different inner bead width than the newest tire that they've just launched. And so, you know, people, I would say, typically find that their tire sidewall measurement is actually a good bit less than it actually measures. Because you're, you're putting a 2.1 that, you know, I mean, even today, a lot of the tire companies are still assuming like a 19 millimeter inner width or, I mean, there's products out there that are still on the market that were based on 17 millimeter inner width. And you're like, shoot, I don't even know if you can buy a 17 millimeter inner bead width rim, you know, anymore. You're, you're very likely using a 21 or a 24. And, you know, that tire is going to going to run millimeters wider on a modern rim than it than it ran on the rim. It was designed around five, six, 10 years ago.

Fatty:

[37:18] All right. Let's stay on this air pressure corner case stream for yet another question. This is a long question, guys. So, you know, settle in, mute yourselves, get a drink, but pay attention. Listener Daniel in Oregon texted us this to the hotline, which is 317-343-4506. He says, my question is regarding weight distribution and tire pressure, wear and rolling resistance. So lots of factors here. I got a new aero bike last year, and after 12,000 kilometers, I'm on my third rear tire, but haven't replaced the front. The rubber on the front is only now showing signs of age. despite the wear indicator still being very clearly visible, even on the center seam line. This makes me think that my weight distribution is heavily skewed toward the rear, at least more than the 4852 setting for road bikes in your calculator, and maybe even more extreme than the mountain bike setting. My position's comfortable, and the bike handle's great, but am I losing performance because of my large posture here?

Fatty:

[38:26] I imagine there's some non-linearity involved, so losses from added weight on one tire are greater than the gains from a lighter front, even if system weight is the same. Marginal, perhaps, but that is your business. So, he's asking, should I use the mountain bike setting for my aero road bike? Should I break the rule of 105 and put a fatter tire in the rear where it matters less? And he notes that he runs GP5K's tubeless, 28s that measure about 31 on reserve wheels, and a system weight of around 90 kilograms. I like this guy's style.

Hottie:

[39:03] Wait, this is pretty easy to consolidate, is it not, this question? I think the question is, if a rider has a large backside, right, should they make extreme adjustments in the pressure calculator?

Josh:

[39:16] I am adding this as a button to the calculator. Like, do you have a big ass? I mean, that is—.

Fatty:

[39:22] I have a big butt, and I cannot lie.

Josh:

[39:25] And I cannot lie. Oh, no, this one is great. Okay, so I think for starters, we can talk through the what ifs, but I would love that you can really easily at home get your weight distribution. And that's, you know, take a bathroom scale and, you know, you put it up against the wall, get your bike. You want to find a book that's the same thickness as the bathroom scale and set it down, you know, wheelbase distance apart, essentially. And I would put the bathroom scale under the rear wheel, but the math is the same regardless. But then you just hop on the bike and get in your riding position and, you know, do it like you're just leaning against a wall, right? Right. So you're not obviously falling over and see what the scale under that wheel measures.

Josh:

[40:22] And like I said, it doesn't matter front or rear. The math is the same and because the other one is the rest of it. Right. And then you, you know, get on the scale with your bike and see what the total system weight is. And boom, there's your your percentage of weight distribution. And you might. Yeah, you might be very surprised. um you know obviously with when we do things like say road or or time trial or or you know there's so many assumptions in there you know what what seat what seat angles your road bike is it is it a zero offset or a 25 millimeter offset post you know how how low are your bars because that starts shifting weight forwards um you know we we have triath like i think we we say time trial triathlon at 50-50, I've had triathletes that are like 56, 57% front weight.

Josh:

[41:15] And, you know, and you kind of like double take and go, holy crap, that's, you know, okay, didn't realize that. So, yeah, I think that would at least help us understand that part of the equation. The other thing could be that, you know, you actually have, because there's a couple factors in rear tire wear. So, you know, one of them could be that the weight distribution is actually a little bit more forward than you're thinking. But because of the calculator, you're running too much pressure in the rear and too much pressure will make a tire wear that flat spot in the center much quicker.

Josh:

[41:58] So when you have too much pressure for a given surface, you actually start to heat the tire and you're putting a lot of very high shear loads in the rubber locally. Because it's kind of, there's just too much pressure. The forces are too high. And then the rubber begins to break down due to kind of the breakdown, the separation of the little like rubber kind of polymer molecules being torn apart and sheer from each other. And so, you know, too high of a pressure is very often a cause of like too fast a tire wear or that, you know, accelerated flat spot, flat strip, really, wearing of a tire. The other thing is, you know, how you ride out of the saddle and how often.

Josh:

[42:50] Because especially on these modern, you know, very stiff bikes, and he says an aero bike, so, you know, I mean, a lot of those are, you know, they're pretty stiff these days. You think when you're climbing out of the saddle and rocking the bike, Your front wheel is able to go with the loads, right? And so you don't have a ton of lateral scrubbing, but you can hear, like that sound you hear when you're out of the saddle, that is rear tire scrub. I think we've talked about on the channel before, you know, I like to call it skating. Like so many of the modern bikes have gotten so stiff that they're actually less efficient climbing because you're putting so much energy into like lateral scrub in that tire. Whereas a lot of the older bikes had sort of a, we call it a modal flex. They had flex in a mode or a direction that allowed the rear tire to sort of almost steer, you know, the forces of the pedal stroke help kind of steer the tire in the direction that the forces are pushing it. And so that reduces the lateral scrubbing in the tire. And I call it skating if you think of like a cross-country skate skiing, right? So not the parallel type of skis, but you see them skate skiing and how like when you angle the ski out and then you put that force and that weight onto it, that lateral movement can translate into forward movement, right? With some efficiency.

Josh:

[44:16] You know, bikes that are less stiff seem to have this natural ability and can climb a little bit more efficiently, but you're also going to wear your tires out a lot less when you're out of the saddle climbing because you have so much less of that scrub. And that's something you can hear. You know, we've done a lot of testing around this and a lot of work with teams, companies, athletes. But it is kind of cool. Some of these old, like, you know, you guys know how much I love old bikes from the 70s and 80s. I mean, a lot of those bikes, like, you kind of don't hear any tire noise when you're out of the saddle climbing on them because they're flexy. And then you get on some of these modern bikes and you're like, whew, that's, you know, that whoosh whoosh sound that's coming up through your modern carbon wheels. You know, that is the sound of rubber being harmed. Um so i would say it could be any or or all of those things but yeah i i would say that let's get an actual weight distribution and then we can come back to this one so daniel when you hear this, Yeah, break out the bathroom scale, and I'd love to know, is it big, large posterior syndrome, or is it one of the others?

Hottie:

[45:35] Okay, more on tires here. This is an interesting question about grip, optimizing for grip, that is, and durability. It comes from Jeff in Northern California. Focusing on gravel and mountain biking, thinking about Northern California's loose-over-hard-pack conditions, is there objectively a tread pattern that would work better than others, or are they all similarly able to control acceleration and deceleration of the wheel? My riding is actually trail on the mountain bike, and my tire choice is usually towards the steepest and loosest part of the regular ride rather than the average. As a result, I am over-tired, in quotation marks. He's not tired, but he has too much tire, but don't want to be let down by my tires, only by my skills. Similar question on gravel and waffle versus file versus other tread. The marginal gain I'm after here is control rather than reduced rolling resistance. And the other thought is that the tread should work, in other words, continue to be effective at least into two thirds of the way worn through since I'm not sponsored and don't want tires to contribute to the throwaway culture.

Fatty:

[46:50] Yeah, I like his pragmatic, eco-conscious perspective, Josh.

Hottie:

[46:55] Well, I'll put an asterisk by the California style of riding, and that is with the loose over hard conditions here in California, it is really hard to find anything that does a great job of grabbing, in other words, gripping, getting through the loose stuff and grabbing into what's beneath it, which is just really hard pack firm. You're always going to have that skid, that slide, almost no matter what you run here.

Josh:

[47:20] Yeah, no, it's a great question. And, Loose over hard is such a, it's, I mean, it's one of the most difficult, right? Because it's one so varied. And like you said, it's so variable even within that context. You know, is it tiny pea gravel over hard? If so, how many, kind of, how thick? Is it larger gravel over hard? And if so, is it one layer of stone? Is it two layers? You know, like, oh, you get yourself in trouble here quickly. Um it's definitely a condition where you know the bigger bigger tire is going to give you more opportunity to get it right um because you're just putting down more contact patch, but typically the way i think of uh this problem is you know what's the size of the you know let's just pick a number and we're going to say okay we're we're average size six millimeter gravel, okay? And it's roughly in a layer, like as we build our mental model, it's one layer thick over hard surface.

Josh:

[48:30] You know, you are going to want a tire that's got a tread that's a little taller than that average size. And most importantly, you're going to want a tire with a tread spacing that is bigger than one gravel in size. And I think that's where you guys know have kind of been on this rant for a long time and well it's part of how dylan johnson and then damn near 80 of the um you know uh lifetime grand prix people ended up on conti race kings right i think the industry really did everybody a disservice with gravel tires for the first however many generations we've had right they they thought of gravel tire as like a beefed up road tire. And so you ended up with extra casing thickness, and then they're cutting into that. And the problem with that is, you know, you can say, oh, we've got five millimeters of tread depth, but we only have, you know, three millimeters of spacing in between tread blocks. And so not only have you made a tire that's actually pretty high hysteresis and less rolling efficiency and heavy because it's got all this added thickness, but.

Josh:

[49:46] You can never utilize the height of that tread because those gaps are too damn narrow, right? Like if you're rolling over six millimeter, you know, chunks, that tread does nothing if it's only three millimeters apart, you know. So something like the Race King comes along and you've got a nice height and then a lot of space in between them. And so you know that all of the the thin stuff thin part of the tire casing in between can kind of sit you think of as like a buoyancy problem kind of sits on top of the average gravel and then the random treads are actually able to get through because they can just you know you think to get through the gravel you've got to displace gravel sideways right and so you know that a tire that looks like that i think a tire that you look at and go well that's not.

Josh:

[50:38] That's not all that many knobs. That's probably the perfect tire here, you know, for what you want. And then you can kind of go about sizing, you know, well, how big is the gravel that I'm on and what that looks like. But yeah, you really need to think of, I've been on a big mission here, like we need to rethink tires as like, what's the buoyancy? And then how do we solve for traction within that realm of buoyancy? But I mean, I'll say it here and for all time period after, I think, you know, the industry really...

Josh:

[51:19] Went into gravel as like, let's make road tires durable. And they should have gone in and said, let's make cross-country tires even lighter.

Fatty:

[51:29] All right. One last question here. And I don't know if question's the right word for it. I should just call it what it is. Can of worms here.

Fatty:

[51:39] A listener, Mark out in Ontario, sent his voicemail into the Marginal Gains hotline.

Josh:

[51:45] Either marginal gains. It's Mark. I had a question about the thoughts or opinions of the show with respect to carbon fiber parts that don't come from the mainstream manufacturers. Most of the carbon fiber parts that we use are made in countries like Taiwan or mainland China. We know that these countries manufacturing infrastructure can produce very high quality parts. But, you know, outside of some very, very small examples, there's no company that I know of that hasn't had their share of failures. Are we really risking anything by going with some of these less expensive options? All right. Yeah. Are we risking anything? What makes this such a hard question is that you don't know until you know. Yeah, he's 100% right. The vast majority of the brands are making this stuff in Asia, Taiwan or China. Those factories are making, typically factories making for half a dozen name brands and maybe half a dozen brands you've never heard of. and then may be selling, you know, them on their own.

Josh:

[53:03] I don't think of the best way to say this. If you, there are factories out there that are known reputable factories that have their own product lines who also make for others. So, you know, Gigantex is probably the largest maker in Taiwan of carbon everything. I mean, you know, they were the company that put carbon cranks on the market, you know, for a bunch of brands, and then carbon handlebars, and they make frames for a bunch of people, and they make rims for a bunch of people. Yeah. So, you know, Trigon is another one. Great factory, really top quality product. They even have their own product line with their Trigon name. You know, if you can find, and the stuff is out there on the internet, I know there's YouTube channels that actually talk about all of these brands and do, you know, there's people who like cut them in half and look at the insides.

Josh:

[54:07] If you can find a name a known factory like that and buy parts that you know are coming from that factory you're you're probably fine or you you're at no more risk than you are with a branded part right i mean a lot you know sadly you know a lot of what you're paying for with the branded part is you're paying for like all the development all the r&d all the testing um you know i know my days at Zip, you know, you're typically into the 150 to 200 handlebars broken in testing before you get to a production layup, right? So you're cutting the tools, you're doing all the design, all of that engineering, you know, it's typically six to nine months for like a new handlebar product, right? So you've got an engineer, you know, 80, 90, $100,000 a year engineer spending six months of their life and company time on that, plus all the cost of all the factory layup. You know, you're typically paying a lot more for like your special layup people in the sample room, as opposed to the production floor people. You know, maybe you're buying or bringing in exotic materials that you're trying in these parts, right? And so, and then you're also paying for that backend support. You're paying for an email address that you can email if something does go wrong.

Josh:

[55:30] You're paying for warranty that you can take advantage of if something happens. And then, you know, probably most importantly for a lot of these products, a lot of what you're paying for is insurance.

Josh:

[55:42] That if something does happen and, you know, you do need a dentist, that there is somebody out there who can be held accountable for what's happened. And, you know, when you go to these like direct from Asia brands, you're essentially, you know, you just need to be aware that you're choosing to not pay for any of that stuff.

Josh:

[56:04] The challenge with this stuff and, you know, part of my time at the WFSGI years and years ago, you know, we had an arm of the industry working group that was focused on counterfeits and knockoffs. And, you know, I think there's a belief amongst a lot of customers that a lot of these products, and I guess I would start by saying, if it's on a website that begins with Ollie, or the price is too good to be true, run. I mean, just get the hell away from it.

Josh:

[56:36] The vast majority of those products are like counterfeit or knockoff products or coming from factories. Factors, quite frankly, a huge percentage of them are run by organized crime. And, you know, they do not give a shit about anything. I mean, you've seen parts, you know, a lot of you cut into them and it's like, you know, it's typically like a layer of newspaper to keep the bladder from sticking. Some fiberglass and maybe a layer of carbon, or maybe, I mean, we've seen parts that are just, you know, black fiberglass with black resin, and then they're painting it to look carbon. There's some really, really scary product out there. And so, you know, if it's an integrated bar stem set up for 50 bucks, there's no way. I mean, there's no way because, you know, that product costs more than $50 from the Taiwan factory, the reputable Taiwan factory. So there's no way anybody's selling it for 50 bucks. There are factories that are known to be the same factory that makes for name brand, makes for Synchros, makes for Reserve, makes for whoever, and they have their own brand. And those are very reputable factories. So if you can do some sleuthing and figure out who those people are, you're probably in relatively good shape.

Hottie:

[57:59] All right. Good wrap-up question there. A little safety announcement from Josh

Hottie:

[58:04] and some insights on carbon manufacturing. Hey, we want to hear all your questions, and here's how to submit them. Text or call that Marginal Games hotline, 317-343-4506. You can also comment on this or any episode at marginalgamespodcast.cc, which is where you'll see transcripts to the show. So if you want to look something up specifically, check out one of the transcripts, or just listen to the whole show, which is what we prefer.

Josh:

[58:29] Well, thanks for listening, and we'll be back soon with more questions and answers. Thanks for listening to Marginal Gains.