Is calibration required for A-B testing

[DaveQB]

Hi,

I think would have been asked, but a search showed nothing.

If I am doing A-B testing and then B-C testing the next session and then C-D the next, at the same venue but on different days, do I need to calibrate? As I don't really care about the actually CdA value, but more the relative difference between different runs.

Thanks.

[Velocomp]

I would do the following:

1) Prior to starting your testing, do a full calibration. If you are confident your current calibration is correct, then you can skip this step

2) Make sure your AP is firmly attached to its mount, so that it cannot rotate

3) After completing A-B testing, download your ride files, preferably keeping AP attached to your bike

4) Assuming you keep AP attached to your bike for downloads, for your B-C test, you do not need to do a calibration ride. Instead, ride for 5 minutes before starting testing.

[DaveQB]

Thank you John.

I would do the following:

4) Assuming you keep AP attached to your bike for downloads, for your B-C test, you do not need to do a calibration ride. Instead, ride for 5 minutes before starting testing.

I see you're highlighting the need to leave the AP attached. I would be taking it off between sessions as I would attach my Fly12 and it would be weeks or months between sessions, possibly. I assume leaving it attached means the result for B in the A-B test will match the results for B in the B-C test. But I don't mind if I get a different CdA result for B in the test results, as long as the results are consistent within the test. That is, I don't need accuracy, just consistency. Like a power meter, it doesn't need to be accurate to be used as a training tool, it just needs to be consistent.

So if B is faster than A and then in a latter session, I find B is faster than C, then I would progress to a B-D test next.

Of course, there might be times I want an accurate CdA value and then I would calibrate.

Am I making sense??
Thank you.

[Velocomp]

Thank you John.

I would do the following:

4) Assuming you keep AP attached to your bike for downloads, for your B-C test, you do not need to do a calibration ride. Instead, ride for 5 minutes before starting testing.

I see you're highlighting the need to leave the AP attached. I would be taking it off between sessions as I would attach my Fly12 and it would be weeks or months between sessions, possibly. I assume leaving it attached means the result for B in the A-B test will match the results for B in the B-C test. But I don't mind if I get a different CdA result for B in the test results, as long as the results are consistent within the test. That is, I don't need accuracy, just consistency. Like a power meter, it doesn't need to be accurate to be used as a training tool, it just needs to be consistent.

So if B is faster than A and then in a latter session, I find B is faster than C, then I would progress to a B-D test next.

Of course, there might be times I want an accurate CdA value and then I would calibrate.

Am I making sense??
Thank you.

Oh, I see what you are trying to do.

It's OK to remove AP between sessions, as long as you understand there may be differences in absolute CdA values. If you are removing your mount between sessions, then recalibrate each time you do a test.

[DaveQB]

Thanks John.

But if I just want to do an A-B test and don't care about absolute CdA, then I can skip calibration for that session?

[Velocomp]

Yes. Just make sure you ride for 5 minutes, without turns or braking, or stopping, so that AP will recalibrate itself. If you want to be super careful, make it 10 minutes.

[DaveQB]

Oh really? My testing area is a netball court. So it is like an outdoor velodrome in practice, without the banking. Thus turns are frequent. Is this not ideal?

[Velocomp]

Oh really? My testing area is a netball court. So it is like an outdoor velodrome in practice, without the banking. Thus turns are frequent. Is this not ideal?

How big is the track? If you ride on the track for all your testing you will get consistent results, but possibly not absolutely accurate results. I would ride for 12 minutes around the track before beginning testing.

[DaveQB]

How big is the track? If you ride on the track for all your testing you will get consistent results, but possibly not absolutely accurate results. I would ride for 12 minutes around the track before beginning testing.

It is very similar to a velodrome in terms of shape and size. A little smaller, so turns are a little more frequent. Quite a good surface as cars never use it. A slight rise in the middle that you go up and down. Perhaps 1% grade.

Ok. 12 minutes pre ride which would be quicker than a calibrate, I think.

Thanks John!