On Sun, May 22, 2011 at 8:06 AM, Olin Lathrop w= rote: > RussellMc wrote: >> Conclusion: Best of following meanderings seems to be a sprocket in >> the top chain run which deflects the usually straight chain run >> somewhat and which can move verticlly wrt chain. Sprocket loads a load >> cell which allows chain tension to be measured from frame. > > Problems with this are that the chain moves laterally when gears are > shifted, and that the speed/force ratio changes as different back gears a= re > selected. =A0You need to know what gear you are in (at least in the back)= to > do the math. > Actually, if you know chain tension and chain speed, you have enough information to determine power. One of the more technically interesting system's is the Polar chain tension sensor. If the distance from front chainring to rear sprocket is more-or-less fixed and the mass of the chain in that section is known, it is possible to measure the frequency of the chain's vibration and calculate the chain tension. The system places what is essentially a bass guitar pickup near the top run of chain, and uses a modified derailler pulley to measure chain speed. This is reasonably accurate when set up properly. (disclaimer: I work for a bicycle power meter company.) It turn out that for training purposes, rather good measurement accuracy is required. A cyclist may improve only a percent or two from one season to the next, so a system with 10% uncertainty is not worth much. So the market is moving to systems that measure torque directly with strain gages. Even Polar is introducing a strain-gage-on-pedal-spindle system this year. Regards, Mark markrages@gmail --=20 Mark Rages, Engineer Midwest Telecine LLC markrages@midwesttelecine.com --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .