Too big differences in CdA

[aeropod.user]

Dear John,

I have an issue which I cannot explain related to significant difference of CdA readings from Aeropod. I'm updated to the most recent version of AP firmware, and my device is set to Profile 3.

I rode two TT rides, first one on 06.06.2020, second one on 16.06.2020. The same track, actually it's a racing circuit relatively flat and with good surface. The difference between those two rides is that during the second one I lowered front stack of my bike by 2cm, saddle height by 1cm and rode with different TT helmet then the one used during the first TT ride. Maybe too many modifications at once for CdA comparison, but it's just to give you the context of changes between both rides. With lowered stack my AP went also down by ~1cm comparing to its mounting at the first ride.

What I cannot understand are the differences in CdA readings and results I achieved.
First TT ride: AVG power 352W, NP power 358W, AVG speed 45.3 kph, CdA from AP: 0.246 (when using CdA Analysis tool from Isaac) or 58.5 * 4 / 1000 = 0.234 (when using AVG from HR channel)
Second TT ride: AVG power 342W, NP power 350W, AVG speed 45.5kph, CdA from AP: 0.347 (when using CdA Analysis tool from Isaac) or 81,1 * 4 / 1000 = 0.324 (when using AVG from HR channel).

Power and speed readings listed above come from my Garmin device.

Can you help me understand why although I achieved better result with less power during the second ride, CdA from the second ride is much higher then the one from the first ride? The difference (0.347 - 0.246 = 0.101 by CdA Analysis or 0.324 - 0.234 = 0.09 by AVG from HR channel) seems to be really huge in favour to the first ride (~101-90W difference in aero watts), but results are quite opposite.

Also, I'm not sure about the difference between numbers in CdA Analysis tool and calculation based on HR channel (with HR * 4 / 1000 formula). The differences are on the level of 12-23 aero watts. What number is in your opinion more reliable - the one from CdA Analysis tool or from HR channel?

I attach files from both days to this post, I also separated TT efforts to two separate files and attach them too.

Thank you in advance for your answer.

Bartosz

[aeropod.user]

... also two full files from both days.

[Velocomp]

Your ride files can teach a lot of good things about CdA measurement and analysis.

It looks like you did your testing in the same place. That is a good thing!

You did nice long tests; that too is a good thing!

So, what is different?

Take a look at the 6/6 ride. Wind data looks good, but perhaps a little bit low:

Screen Shot 2020-06-17 at 7.16.14 PM.png (205.33 KiB) Viewed 4667 times

Now, look at the wind data from the 6/16 ride. Wind data looks really bad (always a tail wind):

Screen Shot 2020-06-17 at 7.21.14 PM.png (173.58 KiB) Viewed 4667 times

Why the difference? I do not know; the profile appears to be unchanged.

Here is what I did:

1) For each ride file, I used the "Analyze/Check Calibration" option. This fixes any wind problem. I did NOT select the "USE DFPM" option.

6/16 ride file with wind corrected:

Screen Shot 2020-06-17 at 7.25.04 PM.png (164.59 KiB) Viewed 4667 times

6/06 ride file with wind corrected:

Screen Shot 2020-06-17 at 7.32.08 PM.png (162.97 KiB) Viewed 4667 times

2) For each wind-corrected ride file, I then used the Tools/CdA Analysis feature to determine the CdA for the test section:

6/16 ride with wind corrected

CdA = 0.227

Screen Shot 2020-06-17 at 7.27.08 PM.png (167.29 KiB) Viewed 4667 times

6/06 ride file with wind corrected

Screen Shot 2020-06-17 at 7.33.11 PM.png (166.68 KiB) Viewed 4667 times

CdA = 0.219

Given the huge differences in wind calibration, these are really close CdA results.

I don't know why there is a big difference in wind calibration between the two rides. But once it is accounted for, things are fine.

[aeropod.user]

Thank you John! That clears lot of things.

Do you think I should hard reset my device and recalibrate it because of this wind related issue, or just do next ride and check it again?

And do you think that this difference in CdA Analysis and calculation based on HR channel (with HR * 4 / 1000 formula) is also related to the wind issue? Or I'm completely wrong trying to compare both values?

BTW - I'm really impressed by your great work!

Regards,
Bartosz Zbierajewski

[Velocomp]

1) A hard reset does not alter any calibration parameters, such as wind scaling. If you want to do a new cal ride, this will adjust wind scaling

2) The HR/CdA values are based on what AP recorded during the ride. HR/CdA is not altered by Isaac CdA Analysis. Yes, the different HR/CdA values are caused by the wind problem