Testing of helmets - success reproducing results across sessions

Post Reply
RolfK
Posts: 26
Joined: Fri Nov 02, 2018 4:47 am

Testing of helmets - success reproducing results across sessions

Post by RolfK »

I'd like to share my recent experiences with the Aeropod and testing of CdA out in the field.

A little background: I bought my Aeropod thru the kickstarter campaign, and initially I was disappointed with the reliability of the tests I did. I wrote about my findings on this messageboard, one thing lead to another, and I ended up being beta-tester of firmware. I've been involved in that for about a year now. I've done many test sessions and I have shared a lot of files with John Hamann from Velocomp.

While not perfect, I will say that with the current firmware (7.15) it is possible to do some solid testing. Testing outdoors is never going to be easy or fault-proof, there will always be the random, hard to explain outlier lap that one just has to discard from a session. I have found though, that there is a fair amount of tolerance towards wind conditions, if the test course allows for consistent riding across test laps.

Some numbers from two, recent sessions to illustrate this.

I am comparing two helmets (Giant Rivet TT vs Met Drone). I'm riding on my tt-bike, but with training wheels and in training kit, so the absolute CdA-numbers aren't impressive. However, I'm able to reproduce a difference in CdA of about 0.002 between these two helmets across two sessions on different days, in different conditions. This is something like 3 watts at race pace - enough to matter, but a small difference. I think that is pretty good for outdoors testing.

Session 1, 2020-06-27:
Session 1 2020-06-27.png
Session 1 2020-06-27.png (32.9 KiB) Viewed 2703 times

Session 2, 2020-08-28:
Session 2 2020-08-28.png
Session 2 2020-08-28.png (128.04 KiB) Viewed 2703 times

As you all probably know, the most recent version of the Garmin ConnectIQ data field records CdA, wind speed and slope in the .fit-file, making it possible to analyse results in WKO5.

I have attached the .fit-file from the last session, if anyone is interested in looking at the numbers themselves.

/Rolf
Attachments
5451912468.zip
(189.14 KiB) Downloaded 194 times
RolfK
Posts: 26
Joined: Fri Nov 02, 2018 4:47 am

Re: Testing of helmets - success reproducing results across sessions

Post by RolfK »

Just to follow up, I think I've figured out how to use Aerolab in Golden Cheetah (v3.5). Using weather station data from the day to calculate air density and the Aerolab settings in the picture below, I got the following CdA estimates:

Screenshot 2020-09-08_19-50-03-178.png
Screenshot 2020-09-08_19-50-03-178.png (147.33 KiB) Viewed 2550 times
Screenshot 2020-09-08_19-51-22-012.png
Screenshot 2020-09-08_19-51-22-012.png (80.77 KiB) Viewed 2550 times

As you can see, the overall conclusion re: the two helmets is exactly the same, although there are variations at lap level (I had to manipulate Crr to match absolute CdA levels, but that doesn't concern me). For me this further validates the Aeropod numbers.
aeropod.user
Posts: 20
Joined: Sat Jul 20, 2019 2:08 am

Re: Testing of helmets - success reproducing results across sessions

Post by aeropod.user »

Interesting post, thanks for sharing this.

I have a questions to your post: how do you find a realiable value for Crr in your tests if they are done not on the same day or maybe at different locations? With typical settings my Aeropod assumes Crr=0.00455. I ride Vittoria Corsa Speed G+ 2.0 TLR, which according to bicyclerollingresistance.com is of Crr=0.00225 at the tire pressure I typically use. When I tweak Crr in Isaac to 0.00225, my CdA increases by something like +0,026 (i.e. the difference is around 26 aerowatts if I calculate it right), so I think precise Crr value may influence what you read as CdA value significantly. Just wanted to know if you care about real Crr value in your tests, or just stick to defaults and care more about the difference in CdA between tests done on different days.
RolfK
Posts: 26
Joined: Fri Nov 02, 2018 4:47 am

Re: Testing of helmets - success reproducing results across sessions

Post by RolfK »

Thank you. I just make an assumption about Crr, I have no way of testing it. The way I test, I'm effectively solving one equation with two unknowns, Crr and CdA. So yes, I'm looking for relative gains; as long as I'm improving my CdA, the absolute value matters less.
Post Reply