CompuTrainer inertia/flywheel...

Terry Ferguson said:

[font=Times New Roman]Been doing a lot of searching on the forums without really being able to find a bottom line.[/font]

[font=Times New Roman]I’m interested in increasing the inertia of my CT to provide enough ‘instantaneous resistance’ so that I can stand up without the pedal just flopping down. Does the heavy fly wheel work? If so, what is the best way to do this? Does the increased kinetic energy affect the spin down calibration? Any other questions I should be asking?[/font]

[font=Times New Roman]TIA,[/font]

[font=Times New Roman]TF[/font]

Contact Jens Heycke on the Wattage list or BTR.com. He has done exactly what you're talking about.

There is an on-going discussion about this exact topic on the RacerMate bulletin board (link below). As of yet, they have not been willing to divulge any information on the effect of the larger mass on the rolling resistance calibration. Link:[url="http://68.115.203.26/phpBB2/viewtopic.php?p=3154"][u]68.115.203.26viewtopic.phpOpen ↗[/u][/url]

Really quite frustrating. But then, I have never been impressed with their technical support. They are so concerned about "competitors" that they have abandoned their user base and won't address any technical question in any manner that actually helps.

My personal belief is that it will have a definite effect on the calibration. How much will depend on the ratios of the mass momenst of inertia, the surface interface, and the difference in windage (air friction).

If you happen upon any specific results, please post.

Regards,
Jeff

When I bought the Kurt Kinetic Pro trainer recently I emailed them to ask how the wattage profile was effected by a flywheel that was 3 times heavier ( 6 lbs vs 18 lbs) They informed me that additional flywheel weight did not change the power curve of their units. They just said it will better help maintain momentum. It would make it harder to speed up or slow down but at any given speed the wattage would be the same. I don't know if the CT is different because it has a variable brake.

JNFletcher said:

There is an on-going discussion about this exact topic on the RacerMate bulletin board (link below). As of yet, they have not been willing to divulge any information on the effect of the larger mass on the rolling resistance calibration.

Maybe somebody like rmur17 needs to apply some pressure. 😉

JNFletcher said:

Really quite frustrating. But then, I have never been impressed with their technical support. They are so concerned about "competitors" that they have abandoned their user base and won't address any technical question in any manner that actually helps.

What I found ironic was the juxtaposition of these comments by Smillard and Roger Moore, respectively:

"...break the coast-down into two phases, phase one measure the rolling resistance as they now do, and then in phase 2 apply a load with the generator to measure the total inertia. working with both measurements it seems they could more accurately measure rolling resistance with different wheels."

"I wish we could answer all your questions about this. This aspect of the product is something I think any of our competitors would just love to learn how we do it as well. I don't think we're willing to teach them."

Why is this so ironic? Because the approach that Smillard proposes is the one used by the Velodyne, but not by the Computrainer, despite the fact that differences in wheel mass (and thus inertia) have a much bigger impact on the latter than on the former (due to the differences in flywheel inertia).

tbowren said:

When I bought the Kurt Kinetic Pro trainer recently I emailed them to ask how the wattage profile was effected by a flywheel that was 3 times heavier ( 6 lbs vs 18 lbs) They informed me that additional flywheel weight did not change the power curve of their units. They just said it will better help maintain momentum. It would make it harder to speed up or slow down but at any given speed the wattage would be the same. I don't know if the CT is different because it has a variable brake.

You're talking about a different situation, however. That is, the Kurt trainer (like most others) essentially assumes that rolling resistance is constant, regardless of, e.g., how much air pressure you put in your tire. The Computrainer, OTOH, calculates the rolling resistance based on the results of the coast-down test, and then uses this value to calculate or control power. As you say, a significant increase in flywheel mass will cause the Computrainer to coast longer, which it will interpret as a lower rolling resistance. This will tend to skew the power data, because the resistance provided by the electromagnetic brake is varied so that it plus the rolling resistance produce the correct total resistance.

Jens had a machine shop make an 18lb flywheel to replace the existing one on the CT which attaches with a set screw. He is happy with the change.

Having said that, he was never able to get an answer out of RacerMate on the effects of the increased mass on the rolling resistance calibration (and therefore power display and control) either. Furthermore, he has not (to my knowledge) verified the power control and display accuracy after installing the heavier flywheel.

acoggan said:

Computrainer, OTOH, calculates the rolling resistance based on the results of the coast-down test, and then uses this value to calculate or control power. As you say, a significant increase in flywheel mass will cause the Computrainer to coast longer, which it will interpret as a lower rolling resistance. This will tend to skew the power data, because the resistance provided by the electromagnetic brake is varied so that it plus the rolling resistance produce the correct total resistance.

acoggan,
Yes, I have a feeling that this is true. However, RacerMate will infer (but not confirm nor deny) that the CT rolling resistance calibration algorithm handles the case of different rotary inertias. They won't provide any data to back up their (inferred) claims, nor will they provide any information on the relationship between changes in rotary inertia and power accuracy.

You must have faith to believe...

I would be satisfied with a simple range of acceptable inertias for a given power accuracy range (e.g., if the attached wheel inertia is less than or equal X times a standard bike wheel then the power accuracy is within Y %).

This sort of useful information doesn't appear to be forth coming.

If anyone is interested in collaborating on a set of experiments to explore this issue, let me know. I have pretty much everything required from a measurement and analysis point of view except a PowerTap or SRM.

JNFletcher said:

acoggan,
Yes, I have a feeling that this is true. However, RacerMate will infer (but not confirm nor deny) that the CT rolling resistance calibration algorithm handles the case of different rotary inertias. They won't provide any data to back up their (inferred) claims, nor will they provide any information on the relationship between changes in rotary inertia and power accuracy.

You must have faith to believe...

I would be satisfied with a simple range of acceptable inertias for a given power accuracy range (e.g., if the attached wheel inertia is less than or equal X times a standard bike wheel then the power accuracy is within Y %).

This sort of useful information doesn't appear to be forth coming.

If anyone is interested in collaborating on a set of experiments to explore this issue, let me know. I have pretty much everything required from a measurement and analysis point of view except a PowerTap or SRM.

Not having a CompuTrainer or a heavy flywheel to fit it, I'm not sure I can help you out in anyway, except perhaps to mention this: during the calibration procedure, the Velodyne asks that you input the exact mass of the tire, tube, and rim (spokes and hub are excluded, probably because they don't contribute much to the inertia). If that information is not available, you can choose between "light", "medium", or "heavy" wheels (with associated mass ranges, which offhand I can't remember). I could try lying to the Velodyne and seeing what impact that has on the calibration, but this situation isn't directly comparable to the Computrainer, since the Velodyne uses a 10 kg flywheel (such that variations in wheel inertia have a much smaller impact).

I thought is was the total mass of the entire rear wheel (wheel + cassette + tire/tube)? Doing so seemed to match my SRM nicely, although it may have been a fluke.

J

acoggan said:

Not having a CompuTrainer or a heavy flywheel to fit it, I'm not sure I can help you out in anyway, except perhaps to mention this: during the calibration procedure, the Velodyne asks that you input the exact mass of the tire, tube, and rim (spokes and hub are excluded, probably because they don't contribute much to the inertia). If that information is not available, you can choose between "light", "medium", or "heavy" wheels (with associated mass ranges, which offhand I can't remember). I could try lying to the Velodyne and seeing what impact that has on the calibration, but this situation isn't directly comparable to the Computrainer, since the Velodyne uses a 10 kg flywheel (such that variations in wheel inertia have a much smaller impact).

xcmntgeek said:

I thought is was the total mass of the entire rear wheel (wheel + cassette + tire/tube)? Doing so seemed to match my SRM nicely, although it may have been a fluke.

No, if you read the screen you'll see that the mass of the spokes, hub, and cassette are not included. It makes little difference, however, since 1) the spokes don't weight all that much, while the hub and cassette are much closer to the center of rotation, and 2) the 10 kg flywheel (which spins significantly faster) accounts for most of the inertia anyway.