Aero Testing/ Aerolab

Thinking of doing some home aero testing for next year’s TT season and experimenting a bit with Aerolab on Golden Cheetah (also using MyWindsock, but I believe Bestbike split works in similar way).
Struggling to find a suitable nearby course/testing venue. Found a local business park which will be quiet on weekend mornings with a road with large roundabout at either end (a bout 2 mins to do a loop - say 1km). However on quite a slope so avg speed only approx 13mph in one direction bit 25-30mph other way…is this ok for testing?

Anyone else done much aero testing and have any tips? Can’t afford windtunnel/professional testing!

I’ve done a decent amount of aero testing on a 1k loop around a business park as well. I’m not sure how the slope would work though, I was under the impression that touching the brakes affects the virtual elevation and makes accurate testing impossible.

I think I read somewhere that if you do have to brake to do it hard which can give you an effective lap marker in aero lab but I might be remembering incorrectly :thinking:

No need to touch brakes at all as large roundabouts at either end. Just the low speed for half a lap means not sure how good results would be???

I image you should be fine then. Give it a go and see how ‘clean’ the data is.

Generally I’ve done 5 1k laps at a time comparing comparing two setups repeated 3 times thorough. The shorter repeated laps seem to make it easier match the elevation to the virtual elevation, so my testing would total around a 30 k ride.

I tended to go late in the evening to do my testing, around 10:30 at night as there was still street lighting and gives a good amount of runs to average out.

1 Like

Have you read Robs guide on Aerolab? He has some useful info on course selection.

I’ve spent quite a lot of time searching for the perfect place to try this but I’ve never been able to find anything quite right…so I’ve given up for now. Must have scoured every road within 10 miles on Google Maps.

I think the ideal type of road is dipped in the middle (half pipe type road), but circuits can work as well. With the incline I guess depending how steep it is, could be more difficult to hold position if you’re really crawling…but this probably has less impact at very low speed anyway. Coming in at 30mph (even to a large roundabout), in aero position, without touching the brakes sounds a bit sketchy to me though.

That’s a good resource. The Platypus Slowtwitch tread is worth trawling through as well. There are a few guides lurking in various corners of the interweb.

This is one thing I finally learned last year — how to brake (like a pro!). Nothing to do with aero testing per se, but get comfortable carrying speed for as long as possible and braking as hard and late as possible (e.g. TT turn around points).

Seems the pros do this only hard or only easy behaviour in a lot of different areas.

Many thanks all for replies and yes read Rob’s guide and even had few emails back and forth with him. He is so helpful, wish I could use his floataero service but money tight.
Will give it a go soon and see what results I get. Currently about as aero as a brick wall so hopefully some improvements (80kg, 6ft1, avg 265W for about 24min 10min TT).

Currently want to experiment with raising aero bars so up at more of a 25-30 degree angle as currently fairly flat/slight angle.

I understand about 5 laps per run is about right (do people bin 1st and last lap?), and repeat setup A-B-B-A to get decent idea

When I’ve tested a couple of things against each other (as long as they are easy to change between) I’ve gone ABABAB for 5 laps each. I’ve started before the lap ‘start’ and finished after the lap ‘end’ so have been able to average the figures from from each of the 5 laps.

I tend to go late evening and although everything there is shut there is a drive through MacDonalds on on of the corners so I have to discard the occasional lap because a car has gone past and the 5 laps leaves a bit of wiggle room.

1 Like

Do you do each run as a separate activity/ride and just press start/stop at start and end of the run?

I leave the Garmin recording for the whole session, if fact I have always used 2 Garmins just in case one file gets corrupted for some reason. I click the lap button each time I pass a particular lamppost - I can do that without moving as the Garmin is at the top of the aerobars where my hands sit.

The advantage of having one file its that you can see the whole file at a glance which can show you the macro trends and just zoom in and increase the resolution to a run of 5 laps and then each individual lap. I’ve never been interested really in finding out absolute CdA but comparing clothing or positions against each other.

This is a file of an ABAB test (the first 3 laps were getting a feel for the conditions)

By keeping the virtual elevation roughly level of the two ‘B’ sets of 5 laps you can see that the ‘A’ sets (1 and 3) slope downward which indicates a lower CdA You can drill down as far as you like by zooming in to get more accurate data but sometimes the macro level can tell you all you need to know. This was a test between a long sleeved triathlon speed top and a short sleeved trisuit.

The actual elevation (the green line) is irrelevant really as its a loop and by definition starts and ends at the same elevation. Checking the constant elevation box and getting a straight line for the elevation sometimes makes things easier to see.

Bear in mind I’m an enthusiast and not an expert who enjoys going fast off not many Watts so check everything I’ve said for yourself!


That is superb, thank you so much. I always planned on doing different ride for each run but from your screen shot I can totally see benefit of doing it all as one. Love your method, makes total sense, and easy to see which of A or B is more aero. I wpuld say though only need to press lap button after each run rather than each lap? Really helpful, many thanks. It does simply astound me the average speeds/low times that some riders get from fairly low watts. Just amazing. I average about 270W for a 10mile TT and my PB is 23.52…yet I see people go sub 20 of less watts!

1 Like

What sort of courses are you riding? A fast DC makes a huge difference compared to a club SPOCO both because of the road surface and the amount of traffic assistance which inflates times.

I’m fortunate to live near a pretty fast 10 course and I rode a 20:26 off 282W average and when I aerolab that ride it gives me a CdA of around 0.15 which would be pretty much impossible without the traffic! I am a small build and around 62-3kg so punch a fairly small hole in the air anyway.

There’s probably no need to lap each one but what it does do is help to spot the laps that have had interference from other traffic so you can discard them.

1 Like

I’m 6ft 1 and about 80kg so lot heavier. My TT are mainly done on a local race track/circuit that runs regular 10mile TT so no traffic at all. One hill (but obviously a circuit so some downhill too!). Also do local club 10 but that’s on small twisty back roads so virtually no traffic and couple of bends need to brake for.

That will make a big difference. Do some aero testing, get your times down so you can get rides on faster courses and smash one :+1:

I’ve a couple helmets and skinsuits I want to test. I read through the Fit Werx article…Just record the data (on suitable road) and use down load to Goldencheetah for analysis? Is it that simple?