Wild change in Wind Scaling

[lorduintah]

100314_cal_data.ibcd4m

4.18 FW Cal Ride and CDs

(685.4 KiB) Downloaded 284 times

100318_cal_data.ibcd4m

4.20 FW Cal ride and CDs

(242.62 KiB) Downloaded 315 times

Just before FW 4.19 was released, I did a cal ride and CDs. I got some of the best results to date - or so I think.

Then 4.19 and 4.2 roll out - now my wind scaling has gone from 0.8 ish to 3.5, The cal wind is terribly unstable and things are really difficult to swallow. Same Bike, Same wheels and the mount has not been moved at all.

I am attaching the cal files from before (4.18 FW and after 4.20 FW). Maybe someone can make sense of this. I set up the tilt and wind cal the same way I always do before these calibration runs. The paths used were the same and there was only a mild to light wind.

Tom

[racerfern]

Tom,

A few questions to start off:
1) Is your iBike on the stem or the handlebars?
2) What made you do another set of calibration and CDs? The first set looks perfect.
3) If you're using the same wheels why does one cal ride say 2096mm circ and the other 2082mm?

I would consider the second calride not acceptable for use or analysis, between the wheel circ, the positive tilt and the varying CD numbers. Something just went really wrong. I'm inclined not to blame the FW updates since I've never seen FW updates do weird things like this.

[Velocomp]

There is absolutely nothing in the FW update that would cause this problem. We haven't touched the coast down portion of the code in years...

Your bad-looking cal ride shows an upward drift of wind speed over the course of your ride. This makes me suspicious that your battery was low.
I'd put a new battery in and try again.

[lorduintah]

Fernando and John -

The new calibration was prompted by the FW - and a corresponding inability to get a stable Wind Calibration - I was seeing large drifts as John noted on the second file. I could get a zeroed out reading and in a few minutes to miles - I was up to 8 mph headwind on an almost calm day or I was riding into a light breeze and seeing 10 mph tail winds. When I completed the last cal ride and went home, the wind was reading 10 mph in my bedroom - a drift for sure.

Just before riding the first cal - all batteries were replaced - I did find one in the spd/cad GSC-10 with a 3.45 V reading - a source of some very erratic pairing.

The mount is on the handlebar immediately next to the stem bracket. It is firmly clamped - the only way would be able to get that to move would be to unscrew it or break it.

I will take another read on the batteries as requested. When they were put in on the 14th all read OC > 3.2V and were "fresh".
The tire circumference changes are due to using the GPS values for the tires. As you may know, the Garmin has the capability to estimate this during the ride from rotations and usually takes about a mile plus to extract a value. I started with 2096 - standard 700 tires value and the second was after a couple of days riding and having the GPS settle down to the 2082 value.

Will post - if I continue to run into more issues on this. I know John means it when he confirms the CD portion of the FW has been untouched - but the wind scaling is not in that part of the calibration... rather the cal ride.

Tom

[racerfern]

Tom,

Don't trust the GPS calculation for tire circ. Do a roll out, here's how I did it and ideally have someone help balance you.

Lay out a long tape measure on a flat surface next to a wall that you can support yourself against. Set the valve stem straight down and roll out one revolution with you on the bike. Measure the distance and multiply by 25.4. Do it two more times to confirm. Make sure the tire pressure is set correctly. Balance yourself against the wall and have someone support you from the other side so you are as close to perfectly balanced as possible. The slightest lean can easily cause an error.

Take this number and punch it into the iBike and go to the Garmin and manually set it there also. This will be far more accurate than a GPS approximation that will vary from day to day by as much as 20mm. That's a huge difference.

[lorduintah]

I went for a short 6 mile ride today. 3 out/3 back.

In looking at the ride file, I had a wind of -6.1 MPH that had to be set to zero (there really was no change in the wind during the ride and it was a true retrace of the same path.)

The left a wind offset around -14.1 pascals.

This is what I have been experiencing since 4.19 FW. So I think either the iBike is not operating properly (yes, I have done several hard resets, yes I once again put in brand new batteries on the iBike and wireless mount.) or the FW change for the auto wind feature had something else slip in.

I sent a separate note to John on this, but private messages don't allow for attachments - I am posting the ride file with and without the wind correction.

Tom

[Velocomp]

I'm off to Spain tomorrow (a one-week session with Footon-Servetto) so I will forward this thread to Travis for analysis.

Your wind scaling factor should be about 3.2--a huge change from its current value of 0.896.

Again, there is NOTHING in FW 419 that would cause this change. Something else is going on; we will help you find out what it is.