Out and back test protocol
Posted: Thu Apr 22, 2021 8:19 am
Apologies in advance for this rambling post! I live in a rural area, where road surfaces are highly variable, often potholed. It's really difficult to find a consistently surfaced road without heavy traffic and where you don't need to brake. If I want to ride a circuit, I have to brake during the laps. If I'm not going to brake during the laps, I need to ride out and back on a single stretch of road, and stop and turn between test laps.
I've had some success with testing on a 3km circuit which has fairly decent surfaces and only two places one needs to slow down or brake, but it's not possible to find a circuit where you don't have to brake at all. The first attachment is a test of two tri bar positions using this protocol and circuit. The runs were in the following order AAAAAABBBBBBAAAAA where A was the lower position. It surprised me as the higher position showed a lower CdA on average by about 0.01m2. The results lap by lap seemed consistent enough except for the first "B" run which read very low whilst riding and then appeared to correct as I pressed the lap button. The first set of "A" runs were from KM 2-21. The "B"s from KM21-39, and the second block of "A"s from km39 onwards.
I'd be happy to stick with this protocol but I'm afraid the need to slow down for some junctions may be introducing error. Traffic got lighter through the test allowing me to coast through the junctions, with far more consistent results in the final block of "A"s.
With this concern in mind, I decided to try an out and back system. I found a fairly well surfaced stretch of road, and lapped back and forth. I set a lap marker each time I passed the start or end of the stretch, so you have alternating laps on the test strip and at each end where I slow, turn in the road, and get back to speed in the opposite direction. The result are pretty poor: a very high variance in the calculated CdA depending on direction, and a high variation overall which I suspect may be due to the runs being too short. I'm wondering if the need to brake and turn in the road is messing up the calibration. When I stopped to switch configurations, the CdA readings were madly low with a consistent high wind speed displayed. I switched the Aeropod off and back on again and got the same outcome so I gave up and decided to seek advice. So, my questions are these:
1 How much will the need to brake to the extent needed by the first protocol effect the outcome? Does the Aeropod correct for braking to any extent. If I coast down to 15 km/h before braking will this help in any way?
2 Is there a better way to manage out and back runs, for example stop and pick up the bike at the turns? Should I be having a longer run in at the start and run out before turning?
3 What is going on with the wind speed and ultra low CdA in the third and fourth attachments? Is this caused by the aeropod's calibration changing whilst I alter the tri bar pedestal height. The Aeropod is mounted on the base bar using a QR plate. Should I switch it off or even when I stop to make alterations to the bike set-up? Because it is mounted on a QR plate I can take it off and put it back on without messing up the tilt.
I've had some success with testing on a 3km circuit which has fairly decent surfaces and only two places one needs to slow down or brake, but it's not possible to find a circuit where you don't have to brake at all. The first attachment is a test of two tri bar positions using this protocol and circuit. The runs were in the following order AAAAAABBBBBBAAAAA where A was the lower position. It surprised me as the higher position showed a lower CdA on average by about 0.01m2. The results lap by lap seemed consistent enough except for the first "B" run which read very low whilst riding and then appeared to correct as I pressed the lap button. The first set of "A" runs were from KM 2-21. The "B"s from KM21-39, and the second block of "A"s from km39 onwards.
I'd be happy to stick with this protocol but I'm afraid the need to slow down for some junctions may be introducing error. Traffic got lighter through the test allowing me to coast through the junctions, with far more consistent results in the final block of "A"s.
With this concern in mind, I decided to try an out and back system. I found a fairly well surfaced stretch of road, and lapped back and forth. I set a lap marker each time I passed the start or end of the stretch, so you have alternating laps on the test strip and at each end where I slow, turn in the road, and get back to speed in the opposite direction. The result are pretty poor: a very high variance in the calculated CdA depending on direction, and a high variation overall which I suspect may be due to the runs being too short. I'm wondering if the need to brake and turn in the road is messing up the calibration. When I stopped to switch configurations, the CdA readings were madly low with a consistent high wind speed displayed. I switched the Aeropod off and back on again and got the same outcome so I gave up and decided to seek advice. So, my questions are these:
1 How much will the need to brake to the extent needed by the first protocol effect the outcome? Does the Aeropod correct for braking to any extent. If I coast down to 15 km/h before braking will this help in any way?
2 Is there a better way to manage out and back runs, for example stop and pick up the bike at the turns? Should I be having a longer run in at the start and run out before turning?
3 What is going on with the wind speed and ultra low CdA in the third and fourth attachments? Is this caused by the aeropod's calibration changing whilst I alter the tri bar pedestal height. The Aeropod is mounted on the base bar using a QR plate. Should I switch it off or even when I stop to make alterations to the bike set-up? Because it is mounted on a QR plate I can take it off and put it back on without messing up the tilt.