Hi all,
I coach 2 triathletes in my team, both share the same powermeter, quite a similar bike setup (in its quality) both sit quite good (IMO) and both weight the same at 80kg.
Thing is, in a test we did a few days ago, one rode 240W for 40K at 35.5kph, the other one did the same 40K and 240W at 36.9kph. (same course and everything).
Where? Why? The only thing i can think of is position… am I missing something? maybe the watts are not really the same and there is some error?
I will not tell you who is whom, so you will not be biased
Here are the photos: http://www.flickr.com/photos/95666807@N06/8728825916/in/photostream/lightbox/ http://www.flickr.com/photos/95666807@N06/8728825792/in/photostream/lightbox/
What about bike weight? Positions are different. Equipment is different. And even though they may use the same powermeters you have all your usual issues with them. Calibration, did you do an offset, normal percent error. You are really comparing apples to oranges here in my opinion.
Am i missing something? Different bikes, different wheels, i would say there are too many reasons to list why the results differ. I mean who knows how well each held their positions over that 40k.
…Where? Why? The only thing i can think of is position… am I missing something? maybe the watts are not really the same and there is some error? …
Yeah, position and specifically CdA is the most likely difference assuming nothing changed in terms of wind or temperature (air density) or passing traffic or other external factors.
But what exactly do you mean by ‘share the same power meter’ is it exactly the same unit and time passed while one rider finished and you swapped the PM to the other bike? In that case I’d suspect changes in air density, wind or traffic between the two rides as well as likely position changes.
Or do you mean the same make and model of PM? In that case were both meters properly torque zeroed before the ride and have both been recently calibrated or checked for accuracy? Do either drift excessively during use (especially if they’re using Quarqs). If that’s the case then I wouldn’t trust the power to the last watt between meters.
But realistically two riders on similar bikes with similar positions (and those positions aren’t all that similar) can easily have different drag coefficients and resulting CdAs and lower is not always less drag as body shape and things like elbow and shoulder width and how they hold their heads matters.
Yes, I happen to be a coach
Yes, equipment is diffefent, but do you think that if I swapped their equipment and fix current positions the speeds will turn over? I don’t really think so…
The guy on the Cervelo was actually faster.
Both held their positions very nicely, I was there, they both are very comfortable in it.
“same powermeter” I meant they own the same brand and model… I did not swap powermeter, but how bad could it be? I really do not know…
What changes would you suggest? if at all
…Yes, equipment is diffefent, but do you think that if I swapped their equipment and fix current positions the speeds will turn over? I don’t really think so…
Assuming there is a large CdA difference (likely) then it would almost certainly have a lot more to do with position than equipment. The eyeball wind tunnel just isn’t as good as a lot of folks think. If you take the time to validate the calibration (and zero offset drift) of both PMs you can do aero field testing to investigate this. There’s a reason time trial riders invest time and money in wind tunnel and field testing, it matters a lot more than some folks think.
… I did not swap powermeter, but how bad could it be? I really do not know…
Well factory specs if everything was torque zeroed and calibrated is on the order of +/- 3 watts at the powers they rode. If one or both meters were not properly zeroed before the ride or had excessive torque drift during the hour or so of riding you could see another 10 to 20 watts of error and if one or both meters was out of calibration by even as little as 5 percent you’d be talking about another 12 watts or so difference. So with somewhat normal variations you could be looking at 25 to 30 watts measurement deltas or more depending on how far off one or both units were during testing.
Did you zero the torque on both meters before riding and do your riders zero the torque before each ride? Have your riders paid attention to pre-ride and post-ride torque reporting to see if their meters drift which is a known problem with some models? Have both meters been checked for calibration accuracy by the factory recently or by field torque testing with known weights? If you want to have reliable power numbers to support field testing or to do comparisons like this then those things become important. If you’re just looking for decent day to day numbers to guide training then it’s not quite as important but if you really want to understand the differences or use power data to tune positions then you want accurate data.
AaronT I don’t quite get what you want to say? Stop being cinical please…
The guy on the trek is about 8cm higher standing, not sure how taller he is on the bike. But might that be all the difference?
Thanks for the great answer, I do suspect there might be some calibration issues, and I do not know if they did torque zero or not, but I will check that.
Both power meters are quite new and went out of the factory around 10 weeks or so.
We will be donig a re test soon and I’ll have them zero torque
I’m saying I can’t take you seriously as a coach because a person that gets paid to make someone faster should know to make the guy on the P3 faster without any change in training. The 8cm is significant as well; you also didn’t provide a shot of the guys from the front. What about tires/tubes?
Thanks for the great answer, I do suspect there might be some calibration issues, and I do not know if they did torque zero or not, but I will check that.
Both power meters are quite new and went out of the factory around 10 weeks or so.
We will be donig a re test soon and I’ll have them zero torque
BTW, you might import both race files into Golden Cheetah and evaluate them in Aerolab. Assume the same air density and see what the VE (Virtual Elevation) plot looks like for both riders. If they look the same but one is just shifted higher then it’s either power measurement accuracy, drag and rolling resistance differences or both. If there are vertical steps in one rider’s VE plots but not the other’s then that says things about how they rode the course like excessive braking followed by harder riding to regain speed.
BTW, do they run the same tubes and tires? Particularly does one rider run latex tubes and the other butyl or one ride fast tires like GP4000s or Vittoria Corsa CX and the other run heavier more flat resistant tires? That could explain some of the difference as well. Same thing for tubulars vs clinchers depending on the tubular glue job (light gluing increases Crr but makes tires easier to change out on the course).
The guy on the Trek has a big ass bottle on the frame vs behind the seat… The guy on the Trek has about like 8cm of risers on the bar and sits a little to far back - could go lower and maybe get more aero. I’m no expert either
…The guy on the Cervelo was actually faster. …
BTW, any chance the guy on the Trek was out of the water faster and passed fewer folks or had less incidental drafting while passing folks during the bike leg? Aerolab could give you clues to that as well.
What was nominal power vs. avg. power for both? VI? What was the course like? This could also have a big effect. But if you do not know that both PM’s were zeroed before the race, then there is really no point in comparing the data…
Why should you take me seriously as a coach? What difference does it make if i’m a coach or not?
That is me: http://www.slowtwitch.com/Interview/Life_of_an_Israeli_triathlete_3371.html and I am a coach. Maybe not as good as you consider yourself. I don’t get your attitude anyway.
I’ll try getting a shot from the front, that is interesting indeed. Tires for the P3 rider is conti gp4000 and trek rider has tubulars I believe they are conti sprinters.
Dave, the test was not done in a race but a clean 40K at a training ride.
NP was fairly close to avg, course was dead flat, VI was the same.
About the torque I’ll get that info soon.
