Difference in CdA calculations: Aeropod vs Isaac

Post Reply
aeropod.user
Posts: 20
Joined: Sat Jul 20, 2019 2:08 am

Difference in CdA calculations: Aeropod vs Isaac

Post by aeropod.user »

Hello,

I performed aero tests yesterday with the goal to test my current bike setup with 2 helmets and then the same with saddle moved back. I tested each configuration in 3 runs (out and back loop), starting with Aeropod calibration ride (after factory reset and installation of the most recent firmware).

I attached runs recorded by Aeropod to this post. Also a spreadsheet with results read from Isaac.

I need some help from you to understand why I get different CdA values from Aeropod during my field tests and from "Tools->CdA Analysis" function in Isaac and what results I should consider as valid. I see it's not only about the values in Isaac being higher than the ones from Aeropod, but also about missing correlation between those values.

For example for 3 runs within Test 1 (test_1.x_HJC_baseline_setup) I got following CdA values:
  • Aeropod CdA: 0,219 -> 0,224 (higher than first run) -> 0,222 (lower than second run, higher than first run)
  • Isaac CdA Analysis: 0,240 -> 0,238 (lower than first run) -> 0,249 (significantly higher than the first and second run)
Attachments
aerotests_20220605.zip
(436.84 KiB) Downloaded 138 times
Velocomp
Velocomp CEO
Posts: 7793
Joined: Fri Jan 11, 2008 8:43 am

Re: Difference in CdA calculations: Aeropod vs Isaac

Post by Velocomp »

Several comments:

1) AP has access to sensor data that is not available in Isaac; therefore, profile 4 AP CdA calculations are more accurate than those in Isaac. I don't pay much attention to the Isaac CdA-calculated values.

2) It's useful to combine test laps together so that you can see test results trends. Open the first file of your test configuration then use this command to add in the other files:
Screen Shot 2022-06-06 at 6.04.06 PM.png
Screen Shot 2022-06-06 at 6.04.06 PM.png (293.23 KiB) Viewed 3019 times
I've attached the combined files for all your tests. Here is the way your first test configuration looks (use 0 filtering for maximum resolution)
Screen Shot 2022-06-06 at 6.06.03 PM.png
Screen Shot 2022-06-06 at 6.06.03 PM.png (224.58 KiB) Viewed 3019 times
3)At the end of each test, where you click the lap button, your CdA for the just-completed lap is shown for 10 seconds. You see this value as a flat line in the HR file CdA = (HR+180)/1000 You can see these test results in the screen shot above for your first test configuration.

This means your spread sheet was "off" by one test; here are the results I get:
Screen Shot 2022-06-06 at 6.23.51 PM.png
Screen Shot 2022-06-06 at 6.23.51 PM.png (106.55 KiB) Viewed 3019 times
4) It's not clear to me what you're doing between tests, and between test configurations. It does appear that you stop between some of your tests. It's best if you can ride continuously between tests without stopping but if that's not possible then I would do 4 tests per configuration (you did 3).

5) When you start a new configuration, the first test has to "unlearn" any of the settings from the prior configuration. So, I do 4 tests per configuration.

6) I'm not sure how you derived your composite test results; If you do 4 tests per configuration, use the last 3 tests. I used the final 2 tests from each configuration to estimate your CdA (the first test configuration has an outlier in the 3rd test, so I used the first 2 tests)

This is a great first start; hopefully these comments will help you dial-in even more your testing.
Attachments
TEST 4-COMBINED_06_05_2022_0314_12_km.ibr
(84.12 KiB) Downloaded 149 times
TEST 3-COMBINED_06_05_2022_0246_12_km.ibr
(83.74 KiB) Downloaded 150 times
Test 2 ENDURA-COMBINED_06_05_2022_0223_12_km.ibr
(84.36 KiB) Downloaded 146 times
Test 1 HJC-COMBINED_06_05_2022_0203_12_km.ibr
(84.04 KiB) Downloaded 147 times
John Hamann
Post Reply