Tilt good, then bad, then good again

[Norm]

After seeing some truly superhuman power numbers in the middle of my last ride, I checked out the "Analyze Tile and Power Advanced View" window and found the attached graph. Slope and elevation track fairly closely for the first 23 minutes, then slope starts climbing like crazy for an hour, then magically slope levels off and parallels elevation again.

I'd guess that shifting tilts during the ride would indicate a loose mount; but my mount is solid and it seems unlikely that tilt would be level, then grossly positive, then level again. Has anyone else seen this kind of behavior?

[Site_Admin]

Send me a photo of your mount... If it's a handlebar mount, that's probably the culprit. You need to use the double-sided sticky tape on both the top & bottom, and also set the bracket pretty tightly. The other thing this COULD have been, was a big change in temperature, but it's not likely.

[travispape]

Norm,

This is interesting data. The fact that the slope ended up being accurate for the last part of the ride indicates to me that when it was off in the middle it was because of high-vibration slightly over-driving your accelerometer rather than an issue with the iBike actually tiling away from the cal position and the magically right back to where it started. Normally the iBike2 tilt and power analysis function will catch such areas and mark them as high-vibration (and repair the data by substituing in redundant data); however, in this ride it only caught a portion of the high-vibration areas. For example, the region from minute 25 to 31 should have been marked as a high-vibration area.

By the way, in this feature there are two different types of corrections that are applied. For the regions marked as high-vibration, the accelerometer data is extracted (slope and accerometer-measured inertia) and substituted with non-accelerometer data (slope as derived from the altimeter and acceleration as calculated from the wheel data). The non-vibration areas are corrected by adding or subtracting a small tilt correction to force the integrated slope data to match the overall terrain profile (as measured by the altimeter). So in your ride, the tilt correction was applied to areas where you actually had vibration.

The good news both correction methods will do a good job making your average power accurate. The fast details of the power might be off some since some of your high-vibration areas were not detected; however, the post-analysis average power should be good even with the tilt-correction.

Travis

[Isaias]

Could a rear tire leak (2x) that reached a point where it quit leaking create such a graph?

[travispape]

Could a rear tire leak (2x) that reached a point where it quit leaking create such a graph?

The tire would have had to reinflate itself at the 30 minute mark and then again at 1:27.

If that 600 ft of tilt rise from the 23 minute mark to the 30 minute mark was at 20 mph (2.33 miles in 7 minutes), then the slope is 4.8%. If his bike's wheelbase is 3.5 ft, then the back tire would have had to deflate by 2 inches to account for the tilt error, which would get him down to his spokes.

So I'm thinking not. :-)

[Isaias]

Could a rear tire leak (2x) that reached a point where it quit leaking create such a graph?

The tire would have had to reinflate itself at the 30 minute mark and then again at 1:27.

If that 600 ft of tilt rise from the 23 minute mark to the 30 minute mark was at 20 mph (2.33 miles in 7 minutes), then the slope is 4.8%. If his bike's wheelbase is 3.5 ft, then the back tire would have had to deflate by 2 inches to account for the tilt error, which would get him down to his spokes.

So I'm thinking not.

Yes, of course, especially regarding the "magic" re-inflations that the slope indicates.

[daucella]

As Richard pointed out it may be the mount to the handelbar but it could also be the mount itself. I had this problem with the older wired mount. Try tapping on the front of the ibike and than the rear of the ibike. The tilt should change initially but return to the pre tapped setting within a few seconds. If the tilt does not return, I suspect a mounting or mount issue.

David

[leonsrocketcar]

...suspect a mounting or mount issue. David

can you or someone elaborate on the mount issue. The mounting issue is limited to 4 possibilities: 1. bad or no sticky tape on bars 2. bad or no stick on mount 3. mount not cranked down to the max 4. combo of 1-3

So what are the mount issues not related to 1 -4 above?

[Isaias]

For S'es & G's, you folks should try this one:

1 - Set tilt
2 - Verify tilt
3 - Remove iBike from mount
4 - re-install iBike on mount
5 - check tilt

I think you may find that even if your mount is glued to the stem (like mine using ShoeGoo) or to the handlebars that your tilt will be different after removing the computer from the mount.

Learned lesson: always, always, always do a tilt calibration if you remove the iBike from it's mount.

Anyone else see this?

[racerfern]

I think you may find that even if your mount is glued to the stem (like mine using ShoeGoo) or to the handlebars that your tilt will be different after removing the computer from the mount.

Sorry, I can't see this. I set my tilt, possibly put the unit outside to acclimate or plug it into the computer to download an interval, put it back on the mount and the tilt numbers match every time. I've had issues but this is not one of them.

[Isaias]

I think you may find that even if your mount is glued to the stem (like mine using ShoeGoo) or to the handlebars that your tilt will be different after removing the computer from the mount.

Sorry, I can't see this. I set my tilt, possibly put the unit outside to acclimate or plug it into the computer to download an interval, put it back on the mount and the tilt numbers match every time. I've had issues but this is not one of them.

Great for you!

That's a sample of 2. Anyone else have a data point they'd like to contribute?

[daucella]

can you or someone elaborate on the mount issue. The mounting issue is limited to 4 possibilities: 1. bad or no sticky tape on bars 2. bad or no stick on mount 3. mount not cranked down to the max 4. combo of 1-3

I'm not sure about the new wireless mount but on the older wired mount the top of the mount was glued to the bottom of the mount. I actually had the glue fail. It was a tough problem to diagnose.

David