All this talk about Tufo’s being slow is driving me nuts. Maybe they are but maybe they aren’t. Here is what I am proposing. I have both a Renn disc (clincher) and a Zipp disc (tubular) and I am willing to spend a bit of time “testing”. In fact, anyone in the Los Angeles area that wants to help is welcome.
What I need is a test that will measure the real world difference in clinchers vs. a Tufo. Having gone to business school and lacking a technical degree I am ill equipped to design this test.
So as a group why don’t we outline a few tests that would be useful. Who wants to go first?
Borrow a SRM. Borrow a Kreitler fork stand. Get the small diamater Kreitler rollers. Find someone who can maintain 400 watts for a few minutes. Have them do 400 watts on each wheel and record the speed. I actually have these items but I live in Northern California.
I’m in on the “Great Tufo” test - I may even do it up by Mono lake and call it the “Tufo at the tufa test”.
You need a power meter or there is no way to compare runs in my opinion…
Anyway - I’m going to compare my 909’s with Tufo tubulars to my training wheels with michelin pro clinchers. I’ll do 4 repeats of 10 miles or so - alternate the 2 sets of wheels. Pump each up to 110 psi. I can ride the aero bars the entire time except I’ll set up 1 short hill. I’ll use my ergomo power meter - download and post the data and we’ll see how it comes out…it’ll be right by an airport so I can get some wind/temp readings for the detail-oriented.
I’ll try and borrow my friends Renn disc/hed 3 - both clinchers and repeat the test.
Now the different wheels will confuse things but my 909’s better not be slower than my training wheels…and if the Renn/Hed 3 wins I’ll be switching camps…
Now if only my new bike would arrive so I can set it up!
I think you would have to do this on rollers. It seems that lots of people have Tufos on their race wheels and clinchers on their trainer wheels. Even if you maintain the same wattage, the speed will be different simply because your race wheels are probably more aero. Rollers seem to cut down on this because there’s not as much wind. Also, how will you decide what PSI to run? If you do both you might get skewed results because a Tufo which is meant to run at say 180 psi might have higher rolling resistance at 110 than a clincher. So, do you run both at their maximum rec. ?? Seems like a lot of variables to take into account here.
From the data posted so far, they magnify the difference in rolling resistance since the major resistance to movement on rollers is tire rolling resistance.
How about using time over the same course as a judge? I have both a clincher H3 and a Tubular 404 to run up front. I was thinking about using the H3 and riding around the rose bowl (3 mile, flat loop, no stops) then changing out the rear disc and doing it again and again and again. Perhaps I can get a few volunteers and we can do this with different riders. Run a little regression on the data and see if wheel/tire is a good predictor… comments?
I don’t think anyone can maintain power even enough to measure the difference accurately. Any tests done on a roller are in error because the road surface is much different…the tire will need to conform to road irregularities, energy is transferred to pebbles that shoot underneath, shape of deformation is different, etc. (note that at high pressure pebbles shoot further and thus have taken more energy!)
I suggest a roll down the hill procedure. This could be done in an empty parking garage with no wind and a decent ramp to roll down followed by a long flat run out. But any hill with a good flat, or even uphill section after it will work as long as wind can be discounted. You will need two identical front wheels with the tires to be compared installed. To be consistent, you will need a clincher and tubular version of the same wheel, ie Zipp 404 or something. Make sure the hubs bearings are in good shape and lubricated. Ideally you would have a neutral party changing the wheels each time and the rider and pusher (if there is one) would not see which wheel is being used. Pressures can be whatever you want to compare, ie exactly the same or max rating.
Start with wheel A on front. You can use any wheel on the back because you are not going to be changing it (to do so would be including error due to hub friction differences, which can be larger than rolling tire differences). You can be sitting up on your bullhorns or brake hoods, no need to be aero unless you feel you can maintain aero in a straight line at very low speeds.
Have a partner hold you steady at the top of the hill and release you. Coast under your own power. If the hill is not steep enough to start coasting from a dead stop, have your partner give you a little push but make sure it is consistent. Have the partner start the stopwatch.
Roll down the hill until you stop. Make sure you stay fixed in your position, feet at 3 and 9. Try to keep a straight line but don’t hit any road irregularities. Come all the way to a dead stop/trackstand if you can. Stop the stopwatch. Record the stopping point.
Switch wheels and repeat. Do this as many times as you can stand, or until you see a pattern developing that you feel is consistent. Compare the time and distances. This is mostly a sanity check. Unless you are hitting terminal velocity, the times should come out pretty close to the same…the faster wheel covers more ground but does it faster so the affects offset. I’m not sure if they should be “exactly” the same, but a part of me trying to recall my 3 years of engineering physics thinks it might (similar to constant period of a pendulum)
Try comparing Tufo at 220psi vs Tufo at 100psi, or other variables of interest.
You could use this same test to compare the aerodynamic differences of different wheels, although a steep/long enough hill to hit terminal velocity would be best in that case (and I think minor body position differences will probably wipe out any wheel differences if you are comparing similar wheels).
Alternatley, take both wheels by themselves (not in a bike) and let them go down the ramp (inside the parking garage!) together…the one the rolls fastest and furthest wins. Of course this totally eliminates the extra deflection of the tire due to the weight of the rider, so to make it similar you might run them at low, but consistent pressures. Hopefully they will track straight! A heavier wheel would have an advantage here. Compare to a bowling ball if you want to prove it.
I hope you have the time for this…I don’t have the time nor 3 friends with that kind of patience. For some reason I took all this time to write this though!
I like the roll down concept. It takes the power variation out of the equation. As I mentioned the closest I have to two identical wheels is a Renn disc and a Zipp disc. Everyone says, “a disc is a disc”. Even if they are not identical if there is a 20 watt difference in rolling resistance it will translate in to a measurable difference in distance traveled.
One question… why do you think it has to be a front wheel and not a rear wheel?
I disagree. I’ve tried rolldown tests and had a very hard time getting consistency with the same equipment. Subtle changes in body position influence your Cda and hence the results. Do the 7-8% hillclimb. It’s the only way I know to get consistent results.
And you’re right in asking. Swapping the front wheel only is silly. Depending on your weight distribution, it accounts for only about 1/3 of the overall RR. Simply shifting your weight a bit could skew the results.
Well, I think silly is being a little harsh. But that’s the great thing about the internet, nobody is afraid to be straightforward. Being able to measure 1/3 the total is good enough if you can do it accurately.
The reason to do front wheel only is because there are significant losses in the ratchet of the rear wheel. You could remove your chain I suppose, which should solve that dilema and allow you to use the disks. You will then need to have either use identical front wheels or the same front wheel each time.
The consistency you get, or lack there-of, with the same wheel is your measurement error. If the measurement errors of the two wheels overlap, then you can look for a difference in midpoint, but it will take many runs for the data to be legitimate. If they are nearly identical, then you have just proven that the difference is pretty small. I don’t know where you got 20W from, but I’ve been riding Tufo S3 and use a power meter for every ride and I am quite sure that I do not see 20W of difference, that is huge. As I mentioned, I don’t think you could measure the difference in aero drag betweein similar wheels because your body aerodynamics are such a bigger factor. But for rolling resistance you don’t need high speeds, so aero can be less of a factor. And again, if you can’t hold a position consistent enough to show a difference, it probably means there is no practical difference.
I don’t think I could ride a 8% hill repeatedly with average power within 1%, on a good day I might hold 3% variation on a trainer for a short while. If you have done this and shown repeatability that is very interesting though. I personally think the difference is so small it would take a machine and laboratory environment to really measure it.
Well, over a 10 mile course at about 17-18 mph (a pace I could hold for several repeats without getting too tired and hold a fairly constant wattage) - the most I’d expect from my race wheels is about 1 - 2 minutes which at 130-150 watts would be about a 10-20 watt difference I think. I believe the website at AnalyticCycling claims that tubular rolling resistance losses over a 40K can be as high as 10-20 watts or up to 2 minutes…so if my tubular race wheels and my clincher training wheels are about the same - maybe race wheels are not any faster or maybe the tubular thing is real…
The Renn/H3 clincher vs. 909/tubular would be better as they should be fairly equivalent race wheels. I’ll have to see if my friend can loan them to me. I can also compare to a set of HED Alps/clinchers which should be a tad slower. It would be nice to compare 2 new sets of identical race wheels (1 tub - 1 clincher) but I don’t have that.
As to holding constant watts - may be a little harder with an Ergomo as you can’t do intervals and get average watts to display - but I should be able to get within 5 watts…on a Powertap I can average within 1-2 watts easily as I just glance at average watts every 1-2 miles and make a subtle change in my watt range - then you can look at normalized watts vs. average watts to look for variation - plus you can look at the watt graph…Then you can do some practice runs all with identical set-ups and see how close the times are…
Maybe it won’t work but it’ll be interesting…
I think the hill climb is worth a try. I have some long hills around here so maybe that will be trial #2.
Roll-down - I have a lot of doubts on that one but maybe trial #3.
Well, I think silly is being a little harsh. But that’s the great thing about the internet, nobody is afraid to be straightforward. Being able to measure 1/3 the total is good enough if you can do it accurately.
. . . I don’t think I could ride a 8% hill repeatedly with average power within 1%, on a good day I might hold 3% variation on a trainer for a short while. If you have done this and shown repeatability that is very interesting though. I personally think the difference is so small it would take a machine and laboratory environment to really measure it.
I retract the comment about it being silly. I shouldn’t post after 8pm…
I should also have been more clear about the hill test. You don’t need to hold the power completely steady for this. You just need to come up with the same average power for the climb. Unless you do something extreme, like micro-intervals in one test and not in the other, matching average power within 4 or 5 watts should not be that hard.
If you can’t do matching trials, there’s something else you can do though since you’ve had a PM for a while. Go back and look at the hillclimbs you’ve done and plug the data for them into analyticycling.com’s calcultors. For virtually every climb I’ve done on clinchers analyticycling predicts the power down to the watt. For every climb I’ve done on Tufos, I come up 10+ watts short.
Not really, but they’ll pretty much single out tire rolling resistance as the only variable.
Doesn’t road surface roughness effect rolling resistance? Wouldn’t you expect different tires to respond differently to road surface roughness? On rollers, a solid PVC tire would have less rolling resistance than a pneumatic tire. However, on a typical road, the up and down bouncing caused by a solid PVC tire would greatly increase rolling resistance compared to the pneumatic tire.
Tests that demonstate that clinchers have less rolling resistance that tubs on rollers do not demonstrate that clinchers have less rolling resistance than tubs on roads.
I think you’re going to need a treadmill. Rollers are round, not flat, and cause mor tire deformation than a flat surface would. This can cause up to a 15-watt difference and will greatly affect how much rubber is on the road. I work for Kinetic, where I have the tools necesary to measure the difference between a severely deformed tire and a regular one.
You do not necessarily need the test surface to mimic the road (the whole rough pavement vs smooth), you just need it to be consistant between trials. Because of this, the treadmill belt works.
There is a company that makes a bike treadmill and they are having obvious problems trying to sell their $60,000 monster. Perhaps one could convince them that it would be better used towards research of this sort.
