CdA changing during the Ride

[raider]

Hi John,
I've recently noticed that during the rides the CdA that is used by the PP (and Isaac) to calculate power is changing and not what was defined on the profile (I'm not using a DFPM). I would like to understand why the PP is doing that, because it is doing exactly the opposite to what I'm really doing on the bike (decreasing CdA by moving to the drops). The attached file is an example, during the segment from the km 95 to 105 I was on a high tail wind and downhill (around -2%), during that time I was on the drops instead of hoods, so I was expecting a higher than normal power output, but it was too high for a small difference on CdA. When I opened the ride on Isaac and use the power meter comparison tool (or the data on the CSV), the CdA used during that segment was 0.55 instead of 0.35 from my profile. Is there a way to tell the PP not to do that and stay with the CdA on the profile? Is that change related to the Speed,Tail wind or what causes the PP to start wild guessing the CdA? I have seen that on almost all rides and if I tried to change the CdA, the offset doesn't go away.

Thank you for your help and for an excellent product.

[Velocomp]

The CdA used in your PP is established at setup (for you it is 0.350), and does not (and cannot) change at all during the ride.

Because you don't have a DFPM, there should be no CdA changes whatsoever in Isaac CdA Analysis... screen. What you're seeing is some kind of random noise artifact generated by Isaac. I have no idea why this is happening, but it is irrelevant.

Isaac is not altering the CdA or power calculations in any manner.

[raider]

Hi John,
Thank you for your fast answer.

The CdA used in your PP is established at setup (for you it is 0.350), and does not (and cannot) change at all during the ride.

That is what I was expecting, however, as I said in my original post, the power was way too high and my W' bal went too negative to be usable. What I did to reach that conclusion was to use the data from the CSV to calculate de power and use it to compare with the power reported by Isaac and PP.

• First of all, the power received on my Garmin is the same (at 60s moving average) as the power on the CSV, meaning that is the same power that I saw during the ride.

• The power graph below is a 60 sec moving average, so is easy to understand.

• As explained on the graph's legend, the orange line is using the CdA as reported by the CSV and the power meter comparison (the gray line on the background is the "variable" CdA). The yellow is the power calculated using a constant CdA=0.35 as my profile.

• I'd expect the numbers to be relative close (as in the left side of the graph), but there is a difference of 150+ watts when the random noise artifact appears on the CdA and the speed is high.

• The power spike at km 97 is a 6-9% climb (very slow) which match very nice between the three lines.

• The next 10km (tail wind, fast and downhill) is where the differences are more clear.

• I understand that the raw data from the PP is more accurate than the data that is exported to the CSV. The difference between the orange and the blue line (PP vs CSV calculated power) of about 50w is understandable, but if the CdA is constant as you stated, then what could cause a 150+ w differences between power pod and the calculated power (blue vs yellow lines) when the speed is above 40kph?

Thank you for your support.

[Velocomp]

It looks like you're using the CSV file exported by Isaac.

I do not understand why Isaac is reporting a changing the CdA value. If this is happening in the firmware, this is a bug we have never seen before.

It seems more likely that Isaac is creating the false CdA readings, but again, I don't know. We have never seen this before.

In the section you highlight, your power readings are also the highest of the ride. Note that your HR is also the highest anywhere during the ride. That is an independent measurement suggesting you were working hard.