I ride in Tucson, Az. On many rides, I see 30 and 40 deg changes in temperature during the ride. I've concluded, after looking at the post-ride calibration, that the iBike tilt sensor readings vary dramatically with temperature. This, of course, impacts the real-time power reading making it less useful during the ride. ... Let's see. Is it hotter than it was when I calibrated the tilt sensor? That means the real output power is somewhat higher than the reading says...
Is there any chance that a future version of firmware could enhance the accuracy of the tilt sensor over temperature? The unit already reads temperature in order to compensate the barometer...
Charlie
Temperature and tilt
-
travispape
- Site Admin
- Posts: 392
- Joined: Fri Jan 11, 2008 8:52 am
- Location: Durham, NC
Re: Temperature and tilt
40 degrees of temperature change during your ride? It must be hard to figure out how to dress for a ride like that!cwwees wrote:I ride in Tucson, Az. On many rides, I see 30 and 40 deg changes in temperature during the ride. I've concluded, after looking at the post-ride calibration, that the iBike tilt sensor readings vary dramatically with temperature. This, of course, impacts the real-time power reading making it less useful during the ride. ... Let's see. Is it hotter than it was when I calibrated the tilt sensor? That means the real output power is somewhat higher than the reading says...
Is there any chance that a future version of firmware could enhance the accuracy of the tilt sensor over temperature? The unit already reads temperature in order to compensate the barometer...
Charlie
While it is possible that temperature could cause some change in the tilt cal, if you are seeing something "dramatic", it is probably something else such as a loose mount. Keep in mind that a mount that seems rock-solid in the garage has to hold its position with all of the hard knocks on the road.
In any case, this is one of the major improvements in GenIII. The new generation fixes any tilt problems on-the-road. The processor in the GenI and II units is too small to handle the calculations needed--we've already packed everything that we can into the FW for the older generations.
If you really are seeing any temperature-depedance over your large temperature range for your GenI or II unit, remember that you can stop and redo your tilt cal. Note that if you had any strain gauge based power meter (e.g., PT, SRM, or Q) you would have definitely had to redo its calibration a few times during the ride.
Re: Temperature and tilt
I haven't noticed a temperature dependance on tilt... there is a pretty big temperature effect on the wind offset though (fixed I hear in Gen3). Are you sure this isn't what you are seeing?
Re: Temperature and tilt
I've moved to a Gen III, so this issue isn't so prevalent anymore.
Yep, it's tilt and it's temperature. It's not a loose mount.
I know it's tilt because it shows up in percent grade - the display I leave on when climbing. I know it's temperature from many, many rides through the summer and fall. The most convincing is to leave home (unit "warm" at tilt calibration), as I ride in the cold, it reports a steeper-than-real grade. As the day warms, it starts to report a gentler-than-real grade. Any ride with a 1-way temperature movement from tilt cal through the ride sees higher or lower gradient. I can always correlate the incline reading offset relative to grade by the temperature change from when I calibrated.
As I said, the Gen III runs a Kalman filter (or equivalent) off the altitimeter to estimate the drift of the inclinometer and that resolves the issue. Only had it for four rides, but it clearly agrees with my mental judgment of grades very closely. I've ridden Mt Lemmon about 40 times per year for the past 10 years - it's 20 miles ranging from 3% to 8% - and I can closely correlate the tilt readings on the Gen III with the visual and perception (effort) of road gradient. I could easily observe the offsets with Gen I or Gen II units.
Charlie
Yep, it's tilt and it's temperature. It's not a loose mount.
I know it's tilt because it shows up in percent grade - the display I leave on when climbing. I know it's temperature from many, many rides through the summer and fall. The most convincing is to leave home (unit "warm" at tilt calibration), as I ride in the cold, it reports a steeper-than-real grade. As the day warms, it starts to report a gentler-than-real grade. Any ride with a 1-way temperature movement from tilt cal through the ride sees higher or lower gradient. I can always correlate the incline reading offset relative to grade by the temperature change from when I calibrated.
As I said, the Gen III runs a Kalman filter (or equivalent) off the altitimeter to estimate the drift of the inclinometer and that resolves the issue. Only had it for four rides, but it clearly agrees with my mental judgment of grades very closely. I've ridden Mt Lemmon about 40 times per year for the past 10 years - it's 20 miles ranging from 3% to 8% - and I can closely correlate the tilt readings on the Gen III with the visual and perception (effort) of road gradient. I could easily observe the offsets with Gen I or Gen II units.
Charlie