Maybe those mavic tubulars have butyl tubes?

Those Mavic tubulars are made by Tufo...so, technically they don't have a separate tube, but are actually a single, large butyl "hose"...'nuff said :-/

---

http://bikeblather.blogspot.com/

Oh good.... phew... I didn't notice the legend in the chart.

It's a shame they test tubulars at higher pressure than clinchers. My favorite thing about tubulars is I can race them at lower pressure than clinchers with less risk of flat, and in my testing on real world roads I have not found a situation where higher pressure is faster than lower pressure (tested between 60 and 120 psi).

Edits to fix auto-correct issues.

Anyone know if the new graphene tire is supposed to be more puncture resistent than Turbo Cotten? Doubt they will be much quicker than the 12-13 watts for the Turbo cotton :)

Not surprised by the result for vittoria graphene 25 mm...in my test on roller, corsa graphene 25 mm and gp 4000s 2 25 mm was very close ...and corsa speed 23 mm tlr was the best in rolling résistance, better than conti TT.

I wish I'd known about the Mavics when I bought a set a few months ago. Thought I was having a bad spot in training...

and corsa speed 23 mm tlr was the best in rolling résistance, better than conti TT.

Did you try a Supersonic? Veloflex Record?

It's a shame they test tubulars at higher pressure than clinchers. My favorite thing about tubulars is I can race them at lower pressure than clinchers with less risk of flat, and in my testing on real world roads I have not found a situation where higher pressure is faster than lower pressure (tested between 60 and 120 psi).

I'm interested in your protocol for that testing, and what your roads are like.

IME clinchers with latex tubes can be run at low pressure also. I've been running 70 psi on the chipseal, 23mm front and 25 rear, 170 lb.

Have you done any field testing to determine the rolling resistance on chip seal at 70 psi versus higher pressures?

IME they can't. I pinch-flatted in a road race at ~75PSI (172lbs.) And it was just a minor bump in the road.

Nope. Feels smoother though.

What tubes?

I don't run that kind of pressure in races because you can't maneuver around stuff. But in probably >40k miles on latex the only tubes I've pinch flatted are 2 times on the really thick Vredesteins that they don't make anymore. That was hitting sharp rocks with 20mm SS. I once smacked into a pothole hard enough to wreck both rims and nearly crash, but didn't hurt the tubes.

Vittoria. But I "misremembered". The road race pinch flat was an installation mistake early in my latex days. The low pressure flat I got was when I cornered hard enough for the bead to momentarily pull away from the rim, and the latex tube slipped through just enough to pinch flat. At least that's what I believe what happened based on the timing of the flat and the forensics. That's why I don't go under 90PSI.

i've done indoor Crr testing on rollers using Tom A's formulas (thank you). indoors, i can run things in a much more controlled fashion.

for the outdoor testing, i had to be a little bit looser. my competitions tend to be on chip-sealed road, so i picked a road with what i felt was representative/decent chip seal. (i race in CA, OR, UT, NV, AZ, NM, WA, CO, WY.....and less often east of the mississippi.)

i picked a steep climb (~10%) in a relatively protected area on a low-wind day. not all wind could be eliminated, but i figured the low speeds of climbing this grade would further minimize any effect of wind.

i'm pretty good at riding in a relatively narrow power band. i did 10' repeats at a pretty reasonable power (think it was 4w/kilo, something that was easy to keep consistent).

i started out at high pressure (something like 115 psi -- don't have my notes in front of me) and then let out air with each run...i think it was 15psi.

i simply noted how far i made it on each run. every time i dropped pressure, i made it farther -- and not just a couple meters. again, i don't have my data in front of me (this was 3 years ago), but the effect was fairly linear. small data set, sure, but the results for each step were pretty close if memory serves.

what i found was that there was no pressure too low where i did not travel further than the prior step -- even pressures that i would NOT race (~60psi).

i got the results i was interested in. for me, the message was to always race as low as i can get away with. this is a balance of road hazards (e.g. potholes--though i race tubulars so pinch flats aren't really a thing), the type of race (in a RR with 100 racers i'm more likely to hit stuff vs, say, an uphill time trial), and handling.

hope that helps.

I think that is a decent protocol for testing the effect of psi on climbing a chipseal road. Did you do repeats? If you start at high psi and only reduce pressure, and don't repeat, then you can experience an obvious bias. Could be wind, PM drift, changing temperature, changing weight, etc. Plus your power output will never be the same, and that needs to be normalized as well. I know it's more work and you'll need to bring a pump. What you did is at least as good as the "testing" that many hacks are posting on the internet!

I'd advise shorter intervals of say ~5 min or less, with distance held constant, 3 repeats, and calculate Crr from your data. You need to establish repeatability. If you want to test 60, 80, and 100 psi, do 60,80,100,60,80,100,60,80,100 for instance. If you (or anyone else) doesn't wish to reduce the data, I'll do it.

my protocol wasn't perfect; sorry. i do take exception to 'at least as good as the hacks' comment, though! :)

i agree i could have changed things up a bit, but i wanted to collect information while performing a workout. my power was within a couple watts every time (AP & NP), so, yes, there was a variance, but it was as minor as my indoor roller tests (keeping cadence and power within a tight band).

your point about an obvious directional bias is one i considered. it was a still day when i did it (intentionally), so whatever minimal wind there was was fairly constant. the low speeds reduced any impact of wind and aero--or so i thought. i also checked total system weight between runs.

i made it substantially further each time--there was no question in my mind that lower was better for my race scenarios. ymmv.

calculating Crr would have been nice, for sure, and i'd aim for that if i ever repeated the test. i'm not terribly inclined to repeat it, though, as unlike comparing brands of tires where the relative difference is important for me to know, the limiter here is the terrain, type of race, and handling. i *know* for my setup that lower is faster.

(i agree that more information, quantified, is always better, and perhaps you can coach me through it the next time -- just saying that personally for a TT when i'm choosing tire A vs tire B i want to know how much better X is than Y as there might be other trade-offs (aero, handling) that i'm balancing against, whereas regardless of which tire i select, x-n psi is faster than x psi for any tire i choose.)

in the 3 years since i did it, it remains the best test i've seen experimenting with varying pressures on a real-world surface. someone has probably done it better, but i haven't come across it.

This was quite a few years ago now...it was one of my first VE ("Chung") test sessions, and used the same tires AFM used in his roller testing (he sent them to me). As can be seen, the results (blue diamonds) match fairly closer to the roller testing up to a certain pressure, after which the "resistance to forward motion not attributable to aero drag" (lumped as Crr) dramatically increased.



One difference between this testing and your hill climbs was the fact that the "half pipe" course it was done on naturally featured a wide range of speeds...something I think is important in this type of test due to the higher vibration energy produced by a given road roughness at higher bike speeds. It's my suspicion that is what causes such the dramatic "break point" in the Crr at higher pressures.

---

http://bikeblather.blogspot.com/

If you check out the presentation slides that go with Dr Chungs virtual elevation method (I.e., the Chung method), he has a climbing based method that allows simultaneous CdA and Crr calculation. You do need an elevation profile for the climb, or at least surveyed endpoint a, but it would work well for this application. I wrote up R code to implement it if you are interested.

your test results, if I'm reading them correctly, are the opposite of mine. (i tested <=120 psi.) you found that Crr increased with a decrease in pressure, right?

it seems that varying speed and power is valuable in VE testing.

when i did the test i actually had a particular hill climb in mind, and for that i was able to run low pressure (no pack issues, no chance of hitting potholes, etc.) -- and while my all-out effort was harder (more w/kg), the duration was longer and the speeds were in the same ballpark. that is, i didn't test at 8mph and then race at 25 or 30 -- i was testing at 8 and racing at 10 (or similar).

i don't really know what the last sentence means, but it sounds impressive.

i don't have a surveyed course -- i'd be subject to using whatever barometric and/or bps-based elevation estimates are available.

seems that in the independent testing the vittorias CRR is even worse than GP4000sII; wonder where did such differences in results arise? anyone with the most recent ones?

The outdoor tests from Alban are out. The Corsa Speed open TLR is the fastest tire he ever tested being about 3 W better than the conti TT.

I hope he will come here and talk about his results. First question: are these results are aero + rolling resistance given that they are outdoor tests ?

Was that testing done with a inner tube?

both with an inner tube and tubeless with 30 ml sealant. Mind that on the table of he had to re-do the tests for the tubeless due to a large drift in external conditions (re-did all the test on the 19th)