Accurate, Affordable, Portable Power Measurement
https://velocompforum.com:443/
Actually, Crr has about the same percentage effect on speed whether you are climbing or on the flat. The big change would be that CdA has little effect on the climb while it dominates on the flat. Seperating out CdA and Crr on a coast-down is an inherently difficult thing to do accurately... especially when Crr is relatively small (like on a bicycle). If your Crr was low, then CdA was a little high to compensate. At typical riding speeds this wouldn't matter much since they would tend to cancel each other out, but on a steep climb where CdA becomes a small factor, the low Crr would screw up your result. That *is* a very low Crr number... what did you need to raise it to, to get better agreement on your climb?tommyturbo wrote:Despite doing multiple coastdown with profiles done on various surfaces, I was not able to get my power #'s to jive. Since my CdA #'s were fairly consistent, I suspected that my crr values of .002-.003 were responsible for lower iBike power readings than what the website was giving me for the same weight, speed, % grade, etc. Crr is more of a factor with slower climbing speeds than it is with speeds typical on flat or downhill sections.
What rruff is suggesting is different than the CdA & Crr adjustment that is available in the detailed tab of the post-ride analysis. What he is suggesting would be for a given set of coast-down data where the algorithm comes up with a value of Crr that is too low for whatever reason (say gusty conditions messed up the data a little) allow the user to override the Crr result increasing it to something higher and make the corresponding downward adjustment on CdA to optimize the fit to the data given the constrained Crr. It's a very good suggestion (and one that we have already thought about), but no, it is not available. We have to strike the balance between detailed, powerful features and intuitive UI/ease of use.rruff wrote:Actually, Crr has about the same percentage effect on speed whether you are climbing or on the flat. The big change would be that CdA has little effect on the climb while it dominates on the flat. Seperating out CdA and Crr on a coast-down is an inherently difficult thing to do accurately... especially when Crr is relatively small (like on a bicycle). If your Crr was low, then CdA was a little high to compensate. At typical riding speeds this wouldn't matter much since they would tend to cancel each other out, but on a steep climb where CdA becomes a small factor, the low Crr would screw up your result. That *is* a very low Crr number... what did you need to raise it to, to get better agreement on your climb?tommyturbo wrote:Despite doing multiple coastdown with profiles done on various surfaces, I was not able to get my power #'s to jive. Since my CdA #'s were fairly consistent, I suspected that my crr values of .002-.003 were responsible for lower iBike power readings than what the website was giving me for the same weight, speed, % grade, etc. Crr is more of a factor with slower climbing speeds than it is with speeds typical on flat or downhill sections.
Does iBike have the capability of automatically tweaking your CdA appropriately if you need to adjust your Crr value?
& how far off was your wattage before the tweak? +or-? Thanks for the infoThanks for the info! Good to see postings like this.tommyturbo wrote:My wattage was around 10 watts low compared to what I was coming up with on http://www.kreuzotter.de/english/espeed.htm
I actually had to go all the way to a crr of .007 to hit the # consistently, and I am now usually right on it, or at most 1-3 watts low. Tomorrow I'm testing another profile I tweaked that has a slightly lower crr, and a slightly higher CdA.
If you can, do a trip file that is only on a climb (being careful to do a standing start, and then coasting to 0 watts at the end, otherwise iBike2 will distort your file because it assumes every trip starts and ends with 0 watts)), plug in the speed, cadence, wind, etc. on the website, and then use iBike2 to tweak the ride file crr and CdA #'s to match. Then it's simply a matter of changing the profile #'s in the advanced menu until they match. Hit the road, do a climb, and then come home and check it out.
For what it's worth, it doesn't seem to matter if I do the tilt repair before I compare the #'s. Since the tilt repair corrects the average gradient, the wattage comparison to the website usually is about the same before and after the repair.
So I don't know if I am spot on or not, but I do have consistent readings that are very useable on the bike. I have used power on indoor trainers, and the RPE's seem to match up with what I remember, but power on an indoor trainer feels so different than climbing outdoors, that I can't really validate it. I have the iBike trainer key, but I have not spent any time on the trainer since I got it. If I do, I'll check out the wattage, and how it compares with RPE/HR, and post the results.
Tom
Agreed 100%. Consistency is the most important aspect for me. Of course it would be nice to know that my 235w FTP is absolutely accurate but as long as I reach 240w FTP in the foreseeable future and continue from there I will be happy.tommyturbo wrote: It's certainly not a perfect method, but for my purposes, I don't really care what my CdA or crr is. I'd just like to know that I am getting consistent results that are as close as possible to a known benchmark.
What I meant was that the iBike takes into account more real-world variables than the on-line calculator does... so using that calculator ( a known benchmark?) to calibrate your iBike doesn't make sense. It makes a bunch of general assumptions that might or might not be accurate for you. If you are riding on a very calm day and you know the slope precisely, then you should get a pretty good correlation... but the iBike measures slope and wind speed in real time. The only issue here is whether the Crr and CdA from the coast-down is correct. .002-.003 is quite low... .007 is probably a little high (but maybe not). But at any rate if you increase your Crr above what you got in the coast-down, you should also *decrease* your CdA a little to compensate... else your reading on the flat and downhills will be too high.tommyturbo wrote:It's certainly not a perfect method, but for my purposes, I don't really care what my CdA or crr is. I'd just like to know that I am getting consistent results that are as close as possible to a known benchmark.
Thanks for those details. I'm a little fuzzy on the calibration, since I don't have an iBike (thinking about getting one). I know that separating out CdA and Crr from dV/dt is a tough thing to do accurately at any rate.ahaile wrote:But if it comes from either a non-zeroed wind offset or from vibration or turns during the 4 mile ride, then the CdA will be correct even though the Crr is off.