You all make fun...but then you also usually are first in line to take advantage of what all this type of discussion results in ;-)

---

http://bikeblather.blogspot.com/

yup. true dat. the skill that i strive (and often fail) to exhibit - my lifelong objective at which i routinely fall short - is to make accessible complex ideas. i must first hold a complex idea in my head (debatable how often i succeed at this); and then translate it effectively.

---

Dan Empfield
aka Slowman

My 2004 aluminum P3, Renn 575 and ancient Hed front wheel disagree. :p I've been riding since the late 80's and it's amazing to see the changes that have taken place. It's also good to keep on top of the latest and greatest so I have excuses for when people beat me on the bike!

Well...in this particular instance, we each were responding to questions of the other at a technical level that, in order to effectively make our points required using some words and concepts that may be slightly beyond the understanding of those not familiar with them. I was attempting to communicate to HTupolev only, just in a public manner. We could have done the discussion in PMs, but then that wouldn't allow those who ARE familiar with the concepts to follow along.

Now then, if you ask me (as you have in the past) to distill those concepts down into something more "easily digestible" to the general public, then I would do so...and most likely employ a measure of diagrams, graphs, and charts to better communicate the point. I think it was Feynman who said something along the lines of "You don't really know a subject until you can explain it to someone who doesn't have the same level of training as yourself" ;-)

---

http://bikeblather.blogspot.com/

Well...sometimes the results of those types of discussions/tests/analyses is "What you have is pretty good already" :-)

---

http://bikeblather.blogspot.com/

no doubt. and i wasn't indicting you; i was indicting myself. if there was a message in there for anybody else, it's just to remind us all that there are, and for 20 years have been, several hundred people on, logged in, and participating on this forum at any one time, day or night. this topic is of prime interest to them, because it directly affect their riding speed, riding experience, and riding purchases.

in fact, i was just discussing this with somebody a couple of days ago. just according to my observation the single biggest difference in how road and tri bike frames are made between, say, 8 years ago and beginning 2 or 3 years ago, is clearance. my road bike frames i purchased up until 2012 or 2014 were often not usable with tires wider than 25mm. over the past 3 or 4 years all my road bikes - for climbing, for general use, and my aero road bikes - have gigantic clearance. they're all rated for 32mm tires but they all take 35mm tires. i think this is a testament to where the industry is going. all today's relevant tri bikes can take 28mm tires easily.

this (they say) comports with their own testing, which often flows from the use of these bikes by their top cycling teams. i noted during the tour when pidcock rode away on that descent that the announcers (euro feed) commented on his tubeless tires, and one said that it was certainly the 25mm version of the tire (they could tell the make and model but not the width) i privately thought: not so fast. a fair number of teams are riding 30mm tubeless on stages with paves and that's almost certainly what got ridden during stage 5. if i were to guess i'd say pidcock may have been riding 28s during that mountain stage.

one difference is that other test paradigms besides drum testing are now used, and these highlight where drum testing is reliable and accurate, versus when drum testing is not the ideal tool. drum testing is hard to fault if you're testing compounds, for example. on the other hand, reliable differences in Crr among tires of different widths and pumped to different pressures appear harder for a drum rig to parse. but you know all of this better than i do.

---

Dan Empfield
aka Slowman

Aaah...I see. Yes the Wheel Energy load, unlike many other similar testing machines that use a simple mass somewhat "rigidly" attached to the wheel, employs an air cylinder to create the load force. So, unlike the nearly complete lack of damping in the 2nd SMD for those typical setups, the Wheel Energy setup as at least a small amount of damping present in that part of the system. In discussions with Josh Poertner on this subject, he noted that this type of air cylinder typically has only ~10% of the amount of damping as what one would need to simulate the damping if an actual rider was the "test mass".



Here's some observations and conclusions that might help make things more clear:

-When roller testing tires on a setup with a rigid connection between the wheel and the test mass, Crr tends to continually decrease with increasing tire pressure. This is true for both smooth and rough roller surfaces, with the rough surface data showing basically an offset, or bias upwards. This implies 2 things:

• The lack of damping in the 2nd SMD system we discussed above results in the lack of a breakpoint pressure
• At the tire level, additional surface roughness results in additional flexing losses in the tire only.

- When field testing with a human representing the 2nd SMD of the system, there IS an observation of a pressure breakpoint. The location of this breakpoint, and the shape of the curve overall is highly dependent on the surface roughness and the speed encountered, and thus the energy input into the overall system through the tire. The fact that below the breakpoint pressure, especially with typical road pavement surface roughness, the shape of the curve matches that of the curves observed on the roller testing with "rigidly" attached masses points to the change of the 2nd SMD properties as the source of that breakpoint observation.


- When roller testing with a setup that DOES include an appreciable measure of damping in the application of the load to the wheel, such as the Wheel Energy setup, a breakpoint pressure IS observed for a given surface roughness, but typically at a higher pressure than what one would expect for the same tire, load, and roughness for field testing done with a live rider. Again, this implies that the amount of damping in the 2nd SMD system is important in the observation of the breakpoint.


- Lastly, all of the above observations are consistent with the concept of the 2 Spring-Mass-Damper (SMD) systems in series, like the model shown:

---

http://bikeblather.blogspot.com/

Good stuff. My suspicion is that the parameter of interest is going to be the Power Spectral Density (PSD) of the road surface and speed combination, and roughly "higher PSD, higher resistance to forward motion (which is lumped in these analyses as Crr)"

---

http://bikeblather.blogspot.com/

I might be misunderstanding what the test setups are doing. The way in which the BRR setup sets load isn't obvious from the photography nor is described on the website's "The Test" page, and the appearance of the rails around and above the tire made me think that the wheel was being rigidly locked in place for the test after the load had been applied. That is, I didn't think it was representing suspension of a rigidly-attached mass, I thought it was removing suspension entirely (i.e. 2nd SMD system is infinite-mass and has infinitely-stiff spring).

In practical terms, that's really not true as long as you are making comparisons in the range below breakpoint pressures, which for TT/Tri purposes actually ends up being a bit higher than some have been assuming lately (especially for wheels with upper pressure limitations ;-) So, roller testing is a very valid method of making those comparisons for which you are asking.

Now then, if you're looking for roller testing to identify where the breakpoint is for a particular setup, then you're probably more efficient making that evaluation with a field test, in that even IF you have a roller setup with adequate damping, breakpoint pressure is still highly dependent on the particular road/trail conditions, the speed, and the "rider properties".

Once again, in regards to "optimum" tire pressures, 'tis far better to err on the side of too low of pressure than too high.

---

http://bikeblather.blogspot.com/

Yeah, the photographs aren't particularly enlightening on that front. Without putting in a direct question to Jarno, I'm only going by the description of the load applied being 42.5kg. I'm not sure why one would do the test in the manner you describe, with applying the load and then locking in the deflection. In any case, I think either situation would end up with effectively "removing suspension entirely", at least in a practical sense.

---

http://bikeblather.blogspot.com/

that for sure makes sense. while i wouldn't stipulate to what you write above without some corroborating (field testing) evidence, you have a lot of experience that gives you a voice of authority on this topic.

when i see a test on BRR, and the best tire is almost always: 1) the widest tire; 2) that tire at the highest pressure; the question then becomes: what is the break point pressure for each tire at each size? this you can't get from drum testing in any way that i can see. but once you get it then sure, it makes sense that drum testing should be reliable at pressures beneath that.

here's the one thing, tho, that i come back to, which drum testing can't measure: what the is the biomechanical effect of vibration? you and i and many others have experienced what i take to be the effect of of muscle vibration in performance when on a gravel bike. obviously riding over washboard but i think the effect is there with more subtle vibration. for me, it's anything from power loss to ass soreness to shermer's neck. i don't even know that field testing gives you the entire picture, unless your field trials involved distances so exceptionally long that the protocol would be impractical.

and i could go on but i'd become even more boring so i'll stop here. there's a bunch of stuff on muscle vibration outside of our industry and avocation, not enough on muscle vibration inside our industry. this is why i like drum testing for what it is and what it does; and i like field testing because of what drum testing isn't designed to give you.

---

Dan Empfield
aka Slowman

If the tire is loaded against the drum through vertical posts, it seems like letting the wheel "float" without troublesome play or stiction during the test might be tricky without a pretty decent linear bearing setup. I was under the impression that this was why some other test rigs suspend the wheel with an arm pivoting from the side.


Maybe. But it seems like vertical deflection transmitted through an extremely-stiff tire to a rigid load isn't necessarily going to tend toward returning to forward motion.

Think about the cartoonish extreme case of a wheel system rolling along a flat surface, which strikes a bump and bounces into the air, then lands on the flat surface away from the bump and continues on its way. That vertical deflection energy isn't returned to forward motion, even if there's almost zero damping in the wheel system; low damping just makes it take longer to dissipate (i.e. the wheeled system might bounce repeatedly). A squishier tire that produces a smaller bounce may lose less energy to vertical motion and have a higher forward speed after the bump.
Edit: And even if the system doesn't fully lose contact with the surface, a vertical lift in the system could give the tire less opportunity to push/roll off the back of the bump, even outside of considerations of hystersesis.

If you're talking about tires of the same manufacturer and model (assuming consistent construction across the model sizes), then yeah...but, that's just physics.

If you're talking across similar tires from different makers, then 1) above is demonstrably not true.

Also, I'm not of the opinion that trying to compare different tires across the ranges of sizes and pressures directly from that data is the most useful approach. Better is to avail oneself of something like the Silca pressure calculator which takes into account the system weight, the surface expected, the relative speed, and the measured tire width to come up with a pressure recommendation. Use that and then compare the tires AT the recommended pressures for each for a given application. That's going to make sure you're comparing them in the state they would most likely be used, and which is going to be well under any breakpoint pressure for the comparison. Make sense?


That is an excellent question...then again, it's long been my opinion that if vibration in the body is a concern (either for power demand or biomechanical issues) then you're probably not running a pressure low enough to prevent it. And if you are then concerned about tire/rim bottoming with the lower pressures, or pinch flatting, then you're not running tires wide enough for the pressures you want to run ;-)

---

http://bikeblather.blogspot.com/

two things: correct me if i'm wrong, maybe i dreamed this, but it seems to me that if you're looking at the same tire (conti 5000 S TR, schwalbe pro one, etc.) that the 28mm tire will usually demonstrate lower rolling resistance than the 25mm tire if they're both pumped to the same pressure. obviously, tho, that's not a guide for real life.

on that question of vibration. the typical thinking is that you're on a typical road, with typical vibration-causing features, you normalize for pressure based in tire size, and ideally you get something like a typical Crr out a 23mm, 25mm, 28mm tire. those tires might have 100psi, 82psi, and 65psi respectively (let us say). back to the typical thinking: that 28mm tire would be more comfortable. is this true? i don't know. i think so. it seems so. this is what my brain tells me i'm experiencing. maybe the deflection in the tire is the same. same sag rate. same spring rate. across these various tires. but perhaps the damping is better in the wider tire. i don't know. but i think what i'm explaining is what is typically thought to occur.

except, of course, that you may well elect not to stipulate to the idea that it's possible to get a Crr out of a 28mm tire for this typical road that you can get out of a 25mm tire inflated to its optimal pressure. but this is the question. zipp says it's not only equal, but better. again, on a typical road. polished wood velodrome? zipp would tell you to ride a 19mm tire.

---

Dan Empfield
aka Slowman

  
Bear with me.

Two points here,
a) you can do some good A vs B road condition testing
b) the contribution of nerds

Above I described the "smooth segment", "rough segment" testing and show very repeatable results
.253/.253/.251 on smooth, .264, .263, .262 on rough. Let's call it .252 and .263





The day before, I had done something similar except
2 longer, hillier longs runs on the smooth, 1 same route on rough segment
Different numbers : .264 vs .252 on the smooth
.271 vs .263 on the rough

(on 2 different days)

As above, orange shows road vibration




BUT.....

There was a 8 degree C difference between the two days

When you apply your correction factor for temperature, magic....

.254 vs .253 for smooth
.261 vs .263 for rough





All this to say that good, consistent, day to day results are possible, specifically for CRR testing and a big part of that is your work on correlation between ambient temperature and rolling resistance.

Yes, that's exactly what I said in my reply above. Across a particular model, wider (measured) tire is faster at a given pressure...no matter if the measured width increase is accomplished by a physically wider tire casing (i.e. tire sizes) on a given internal width rim, or by keeping the tire the same and increasing the rim internal width. Basically, "measured width is measured width" in regards to Crr and inflation pressures.

But, compare across the sizes at pressures "equalized" to the measured width, and the Crr differences become nil (for a given model).


Yes. This is true, because the majority of the "spring rate" in the air spring known as "a tire" is driven by the air pressure. Lower air pressure means lower spring rate (less stiff). However, the other part of that spring rate is the effects of casing tension, which for a given tire increases with measured tire width for a given pressure. That behavior is driven by geometry. Dropping the pressure to get the same deflection for a given load is mostly counteracting that geometric effect of casing width...and as HTupelov pointed out above, is mostly in regards to "flatter" inputs, since the Silca data points out that the smaller radius the interacting object, the less effect casing width has on spring rate of the tire (i.e. for small object inputs, it's ALL about the pressure)


It's not about the damping in the tire...the air spring contained by the tire has nearly none, and more damping in a casing means more losses, and thus a slower tire. If anything, you want to minimize the damping in the tire. What you're feeling as better comfort is the softer air spring from the lower pressure that the wider tire allows.


Now we're getting back to my original question in post #15, where the Zipp data shows the 28 and the 30 tires on the same rim having equivalent Crr at the same pressures, and then the 28 having better Crr than the 30 at slightly higher pressures (still in the 65-72 psi range). Both of those observations "go against the grain" of a lot of other data. I would expect the 28 to have better Crr than the 25 at the same pressures, and be nearly the same at width adjusted pressures, but that result also depends greatly on how "equivalent" in construction the 2 sizes of tire actually are. And, also depends on the definition of "optimal pressure" being used ;-)

---

http://bikeblather.blogspot.com/

Cool! Thanks for sharing that! I'm still amazed at how adding the Crr temperature compensation into my own VE spreadsheet made my results appreciably more repeatable. It's great to see it working out for others as well.

It's also amazing to me that it works so well since the value is basically from testing on a single tire. It would be interesting to see how (and if) the correction value changes over a wide range of tires...or if that original estimate is "good enough" :-) A project for another day, I guess...probably once I eventually retire and finally fabricate my own automated roller tester I've been promising myself for quite some time now <LOL>

---

http://bikeblather.blogspot.com/

We repeated the tests with several tires. We did not use your formula exactly, we used the data from our tests. but the correction functions are very close. There were at least 5 tires involved.

Since, we also tried using it across days in outdoor testing but to be honest I only know one fool willing to do same courses, same position, same equipment over and over.

zipp has a lot of charts and i just included one as an example. i'll include some of the others as i get a free moment. they actually show the 28mm to have a lower Crr, at normalized (lower) pressure, on their rolling road, and that 28mm is faster in general - all factors included - in field testing. now, on the rolling road they're not testing Crr, they're testing power. roll the road at 24mph (say), how much power does it take to pedal the bike and remain aboard the rolling road and not go flying off the back? many dozens (or more) of trials at different pressures and widths.

i ask myself what might account for this and i don't know. there are other factors at play such as the aforementioned muscle vibration. i just find it all interesting but beyond that i've had a running argument with the wheel brands who use hookless beads and want to put 25mm tires on them. when i add it all up, take in all the factors, i have no quarrel with the hookless platform and find a lot things i like about it; but i just don't see it as a platform that works for a 25mm tire. the required pressures are too high for a rider above, say, 160lb or 170lb. so, if zipp's testing is repeatable by other brands and others like you, and if 28mm tires on typical bike racing and triathlon roads proves to roll even equally as fast (we can then talk about aerodynamics) the news is that it unlocks the utility of hookless as a viable platform.

---

Dan Empfield
aka Slowman

Dan,
Would appreciate your advice if you are happy to share.
These wheels are hideously expensive down under here in Australia.
That being said, I can currently get the 808 NSW Tubeless DB wheelset for about $3500 AUD on sale. I am not sure how many generations of the 808 NSW's there have been, hence how old this technology is.
A tough question I know, but is there likely to be a massive difference in terms of time savings/watts with these wheels over the new ones you have talked about here?
I was an all Zipp person on my previous bikes, but since making the change to DB's have had to start from scratch again.
FYI: late 40's, 2:21 HIM bike split on my P3X running an 80mm Front and Rear disc by Parcours (2nd fastest age group split). Not sure it is worth it to spend anymore money on wheels as these certainly performed OK on the day.

there is not enough distance to sweat the time and money for the new wheels if the old wheels are considerably easier/cheaper. and there is one added value to the old wheels: you can use new or old tires; and you can use whatever pressure you want without consideration, which is not much of an issue with 28mm tires but it may well be if you're riding 25mm tires. you can ride 25mm tires with the new wheels, but i wouldn't recommend it because you may want to push that 72.5psi max and you can't safely do that, imo.

me? i would buy the new wheels, all things equal, because i'm done with 25mm tires, i'm all in on tubeless, i have scads of experience on everyone's hookless road wheels, and i have access to conforming tires for those wheels. but you live in a 4th world country down there (where i'm scheduled to visit in december, i'll bring provisions). i don't have confidence you'll have enough available tires. buy the old wheels.

---

Dan Empfield
aka Slowman

If that's the case, then the rolling road is the same as a roller test...just that the "roller" is flat (i.e. roller radius increased to infinity). They might be bringing in some of the rider damping that could result in breakpoint observations, depending on the roughness of the moving section. But, I think I'd need to see how the moving section is supported before being able to judge if that's something that could be "teased out". Even so, that chart doesn't seem to be showing that anyway (i.e. breakpoint effects) and I find the data presentation a bit awkward, with it being normalized to the 25 tire results. More insight could be gained from seeing the power vs pressure plots, I believe...normalizing to the 25 removes some information that might be interesting to see.

So yeah...seeing more of the info in your article might be helpful...what's taking so long?? ;-)


Well...combine that with the common opinion (one held by myself as well) that road tubeless for measured tire widths 30mm or smaller doesn't gain advantages over running a tire with a latex tube...and in some cases, disadvantages...then yeah, I think I'd join you in the opinion that hookless is a tough sell for for even 28s. Especially since hookless construction eliminates the possibility of running a non-TLR tire as well.

---

http://bikeblather.blogspot.com/

That is fucking insane.

i've been running conforming 28mm, 30mm and 32mm tubeless tires on just about every wheel made by every brand making hookless road (ENVE, HED, Zipp, CADEX). so far, i'm very happy. no spiking the football, not a flat since 2019 (when i began my hookless odyssey). but that's just so very n=1, and it's not hookless but tubeless that i credit for no flats. but... no rolldown tests. no chung. i'm JRA. i have no data, other than, you know, no flats. and, a pretty comfortable ride, to the limit of how comfortable a tire can make you.

---

Dan Empfield
aka Slowman