Wheel-on trainer spin down calibration
davethebike
Posts: 7
I’ve got meself a 1st time, wheel on smart trainer (wahoo Kickr Snap) and just like a lot more people I’m sure, I’m fiddling around, to try and get the resistance etc as near to real road as I can. So now I’m into spindowns, and I think I’m actually getting somewhere near what my outside riding is like.
All of that though has got me thinking, about spindown calibration in general and I’m niggled now. Is this calibration method unnecessarily measuring, and compensating for something that just does not exist when pedalling?
I know that you ride the trainer up to a set speed then when you stop pedalling. The software measures how long it takes for the whole set up to slow down/stop. I can see how this would work, because the software would then compare this slowdown rat with its known slowdown rate on a reference trainer similarly up to speed but this time running totally free, ie with no bike wheel connected. I read that this difference can then be used to compensate for the effect of tyre + wheel bearing resistance. That’s fine, yes that resistance is obviously slowing things down and needs compensating for. But it all seems to completely ignore the other thing that unavoidably comes into play, but only when in spindown: the bike’s freehub drag. which obviously adds to the slowdown but which is irrelevant, as it’s never in play during actual riding. I know it won’t be a lot but all freehubs have some drag. So therefore the aggregated drag measured by the system is greater than what actually exists when pedalling and is therefore over compensated for.
It’s just something that’s niggling me 🙃
All of that though has got me thinking, about spindown calibration in general and I’m niggled now. Is this calibration method unnecessarily measuring, and compensating for something that just does not exist when pedalling?
I know that you ride the trainer up to a set speed then when you stop pedalling. The software measures how long it takes for the whole set up to slow down/stop. I can see how this would work, because the software would then compare this slowdown rat with its known slowdown rate on a reference trainer similarly up to speed but this time running totally free, ie with no bike wheel connected. I read that this difference can then be used to compensate for the effect of tyre + wheel bearing resistance. That’s fine, yes that resistance is obviously slowing things down and needs compensating for. But it all seems to completely ignore the other thing that unavoidably comes into play, but only when in spindown: the bike’s freehub drag. which obviously adds to the slowdown but which is irrelevant, as it’s never in play during actual riding. I know it won’t be a lot but all freehubs have some drag. So therefore the aggregated drag measured by the system is greater than what actually exists when pedalling and is therefore over compensated for.
It’s just something that’s niggling me 🙃
Dave
0
Comments
-
The spin down is for the power calibration, nothing to do with road feel etc.
0 -
I totally agree that it’s got absolutely nothing at all to do with road feel.
I’m saying that freehub (pawl) drag, which is obviously completely dormant when pedalling, gets unavoidably included in a non-pedalling, wheel-on trainer spindown. That creates unwanted and irrelevant information.
I’m sure the spindown boffs know, and consider that unless it’s a shopping bike that’s been outside for a year then this effect can be ignored.
But it’s still a good mind exercise.
I understand, and it makes complete sense to me that spindown is to get power accuracy, and making sure that the trainer reduces app induced braking by the right amounts, to compensate for power that’s getting lost due to what is considered to be tyre/roller resistance.
The manufacturers trainers have laboratory reference spin down times, between arbitrary start and finish speeds for a trainer with no bike on. These are then compared to an actual spin down time with an individually unique bike on. The difference, which is considered to be caused by tyre resistance at the roller, is used to calculate the power % offset needed.
My point was that the spindown test, because effectively it’s in reverse. ie it’s the roller driving the wheel this time, can never avoid including the freewheel drag, never mind wheel bearing drag. That will affect the calculations even though, unlike tyre and wheel bearing drag, freewheel drag is completely irrelevant when pedalling. That means too much power is assumed to have been lost, and therefore too much braking is removed from what the app is requesting at any given moment. So the stickier the free hub, the greater the error.
Basically, your ride ends up easier than it should be. That’s it 👍Dave0 -
You're over thinking it.
Your theory is only relevant if using erg mode, but even then, I'd imagine it would still be within the +/- 3% wahoo quote (assuming everything is set up correctly).1 -
Dave I think you're right, but that the freehub drag is probably negligible. If its a good freehub, and in good condition. Otherwise, when you actually ride the bike outside, it would feel like you were on the brakes when you tried to coast, right?0
