Thanks for this. (Not the extra homework).
My left leg is already sore just thinking about the best way to test.
I’m about to start testing a 4iiii Left crank on my road bike. They have a nifty little option in the app that allows you to adjust the readout by a % margin. IMO, this is what Stages need to do in their app.
Having the option to scale the readout seems to be the easiest way to align different bikes/trainers. It’s not going to be perfect, but you can get fairly consistent data over the course of a ride (indoors, or on the trainer).
Now I need this stupid cold to go away so I can get back to training!
We are talking apples and oranges here. You are right that I should have added that the unit needs to be warmed up for the reasons you give. But that is very different from the necessity to compensate for temperature variations with strain-gauge-based power meters. Here, you are measuring tiny resistance changes, and the resistance does noticeably depend on temperature. The measurements of the optical torque sensor do not need temperature compensation.
What I want of a power meter is to accurately measure the power at that point in the drive train system. All smart trainers measure the power that arrives at the wheel, so just like a hub-based power meter, they will read lower than crank-or pedal-based power meters, since they measure “earlier” in the drive train. This does not make them less accurate, you just need to be aware of the difference. In thew world of cars, you have the exact same thing, you have wheel horse power and engine horsepower. So when you compare two otherwise identical cars, but one of them has all-wheel drive and the other one doesn’t, engine horse power will be identical, but the all-wheel drive version will have lower wheel horsepower due to the added friction in the drivetrain. Needless to say, there is no way you can accurately estimate power loss in your drive train.
Regarding the friction in the Direto’s mechanism, that does not figure in at all, because that is taken into account. Indeed, that’s what the calibration is all about where you spin up the flywheel and let it come to rest by itself. This measures the energy loss at a given rpm of the flywheel, which means you can account for friction.
My Kickr used to hit 40c easily during a solid workout. +50c isn’t unusual.
The flywheel was hot enough that you couldn’t hold onto it (50c+).
Looking back through things, the spin down times varied between 11 to 18 seconds after the unit was worn in.
IME, 10 minutes is not enough of a warm up for the trainer drive train. It’s probably not enough for your bike drive train if you’re being that fussy.
Moving away from that lot, I did some messing around with the scale function on the 4iiii unit today. Pretty happy with the results.
Tomorrow I’ll see what I can get from the Stages G3 crank length setting and report back.
I tested the Stages against the Neo 2 using different crank length settings in the app.
On a 175mm Gen 3 crank the 165mm setting lowered the readings from the Stages across the board a little.
This is to be expected. You changed the length of the lever arm which results in a diff torque and thus diff power value.
Torque = Length of lever arm x Force
Power = Torque x RPM
You should make sure the Stages crank length is set to the actual value, firmware is updated, and calibrated before riding. If your Stages is differing substantially from your Neo2, then you have a problem.
If you only the Stages to compare w/the Neo, I would say the Neo is the good one. If you have a 3rd power meter to use for comparison, that might give you a more definitive answer.
That is not a good idea. You shouldn’t change crank length — unless the crank length actually changes. The Tacx Neo has a good reputation as far as accuracy is concerned, so I’d just stick to that.
Thanks for the good advice guys. I typically do just that and stick with the NEO2 for training purposes.
Power match sucks IME with a left only PM, so no go for me there.
I should’ve mentioned that I’ve been testing things on different trainers and PM’s for nearly two years now.
Every time I do a test run, it involves various power efforts and individual leg drills to give me a reference with LR removed from the equation.
DCR analyzer for looking at the results.
During that time I’ve tested 2x gen 3 Stages and found them to be fairly consistent. The new 4iiii that I did some testing on yesterday seems to give similar results.
The Neo2 just does it’s thing. The Kickr (sold recently) liked being warmed up properly and needed it’s own scale factor to match the Neo and 2x Stages units.
My imbalance and drive train loss scales pretty well between the Stages and Neo. That’s made it easier to do the math and dial things down a little outdoors.
Intervals ICU allows you to scale data after the fact which is another good bandaid solution.
At the end of the day I’m more than comfortable playing around with crank length on the Stages. It has much the same result as adjusting the scale factor on the 4iiii. If it gets me data that’s aligned more closely with my Neo2, that’s what I’m doing.
After all, that’s the holy grail with multiple bikes and trainers. Numbers that align.
The Stages app should get with the program and offer a scale factor option. They missed out on another sale with me as I’ve been really happy with the hardware itself. Happy enough to pay the Stages price premium even.
In the meantime, changing the crank length in the back end of the app is easy to do and simple to reverse if needed.
(Or TR could offer a scale factor option along side the offset option).
After some more testing using Zwift today, I can confirm that changing the crank length in the Stages app is a worth a try.
My drive train and imbalance typically caused a 10% average increase in power numbers on my XC rig. Not a set 10% across my output range, but close enough for a left only unit over most rides.
Changing the 175mm crank to a 165mm setting gets me damned close to the Neo2.
I just added, to my Zwiftpower analyses (see above, post #3 in this thread), Stages @ 165mm vs my H3 doing Pettit. Basically a 4w difference in power.
My actual crank length is 175mm.
I still don’t like the idea of using a single-sided power meter, particularly for PowerMatch, especially given my perceived pedaling imbalance (so why did I buy a Stages in the first place… ) But I have to admit, now that I’ve got it down to such a small difference it seems pretty usable for outside workouts.
Edit: Sent my data to Stages support (who were very responsive) and determination from them was that my power meter is likely functioning properly, and attribute the 20w difference to drivetrain loss.
Here’s my experience…
Used Stages PM from 2016 - March 2019 then I got a Kickr18 and TrainerRoad and here is how the FTP varied when dual recording. My 1st Kickr18’ was very close (except in October when it had started making noise and I had to continually tighten a pulley bolt) to what the StagesPM read however the replacement Kickr18’ I received in February 2020 has shown 19-25 watts lower than the Stages PM.
