Comparing power meters on multiple bikes

Anyone have thoughts on how best to make sure the power meter on bike 1 and power meter on bike 2 are measuring within a few watts of each other?

I have a “trainer” bike with a Stages and an outside bike with a different model Stages. I want to make sure they’re really measuring close enough to each other to not impact either my inside or outside training or zones/TSS measurement.

Thinking of doing a ramp test Tuesday on one, rest Wednesday, ramp test Thursday on the other. Same Road Machine trainer, fan setup, etc. for both. Zeroing each first of course.

1 Like

Ive had issues with this. Have a 4iii on my winter bike got a stages for my summer bike and all of a sudden was breaking every power record I had. Complained to Stages they took it back to test, 4iii say there’s is right so I was left with a useless setup.
So created my own workout in trainerroad (mini ramp test less stressful) that repeated twice in the workout. I’d have trainerroad control my elite trainer and then record the power on my wahoo, switch bikes halfway through and find the difference. 5.65% in my case. With the 4iii you can scale this
I don’t like it I think for the money they should work but at least I have both my bikes matching.

Yeah, with a smart trainer I think you have benefit of that and its erg mode as a constant. For me with dumb trainer I’ll have to use my legs and I figure the full ramp to exhaustion is only way to normalize. Assuming of course I’d reach near enough the same peak two days apart.

I dont think I would trust my legs to be that consistent. I think you need to try and borrow a friends power meter or trainer so you can create a constant to work against.

Yeah, I get your point. Putting both bikes on a Kickr or similar would sure make this easy.

I used dc Rainmakers analyzer to compare my Neo to my left crank Zimanox. From that I know how the two meters differ. It’d be nice if they matched but there’s a significant difference so the Zimanox has gone back to have the zero calibration looked at.

https://analyze.dcrainmaker.com/#/public/a0efdb56-286a-4b86-5c82-a486288b18cd

Are the cranks the same length? I’m pretty sure all the shimano road crank arms of the last few generations are mechanically interchangeable, so you could swap crank arms and leave the rest of the trainer/bike setup the same. This would allow you to use both power meters on the same bike/tire pressure/trainer tension, and would keep the variables to a minimum. Since you won’t be able to run both power meters at the same time, I think doing a few steady efforts at fixed power on each would be the way to go, rather than a ramp, and do repeated efforts at each level. Something like Low-high-med-low-med-high for each power meter. You would need a speed sensor to track the speed/vs power for each of these steady efforts. This way you can compare each of the efforts with the same power meter against each other - this gives you some idea how repeatable your test is. Doing the crank swap twice would also get your more data to validate how repeatable your measurements are. If these aren’t really close, I don’t think you’ll get much interesting data out of this.
My guess is that it will be difficult to measure with greater than 5-10% accuracy, which won’t tell you that much. Also, since the numbers you are comparing are power and speed, you will need to know the power curve of your trainer to determine what the error actually is. I think unless you have a strong suspicion that they are really off it’s likely not worth the effort.
Maybe doing something really simple like picking a workout level that you have a good feeling for the RPE and doing it on both bikes (or with the crank swap) and comparing the RPE would give you a good idea of if they are far off. Doing an over/under workout could be a good choice, as you will likely notice a 5% difference there.
You’re just really limited in what you can do if you don’t have a setup where you can simultaneously run multiple power meters.

I’ve heard a lot of people say that there will be variation in power readings even between two power meters that are the exact same make and model. Rather than getting into the weeds about it I would just recommend treating both as equal until you notice a consistent difference between the two, at which point you can simply adjust manually when you’re riding on one versus the other.

Given that you are running stages on both, I wouldn’t even bother with it though… if you limit one to inside use, the difference between riding indoors vs outdoors alone may skew the power numbers anyways.

1 Like

I’m with you @mattymurda. But I have felt - I think - a difference between the two and I’m wanting to try to measure it indoor vs. indoor in order to quantify that first as variable, and then I’ll know whether I’m also maybe dealing with indoor vs. outdoor performance difference. Also, I like geeking out on this stuff!

And, in the conversation in another similar forum thread I actually came to realization that I could maybe try to use Virtual Power as a constant to measure both against. It may just work well enough. I’m going to try it to see:

I’m just in the middle of doing this. Kinetic to Rock and Roll with Stages on the bike. I did the ramp test using virtual power and recorded the stages onto my Garmin. Compared using the DC rainmaker tool and plan to adjust the wheel circumference to get them to match as close as I can. I’ll then swap the stages for my other PM (4iii) and see what the result is. You can alter the 4iii using their app by a %. Not sure what the outcome will be. One problem is that by altering the wheel circumference you are in effect moving onto a different part of the power curve which may throw it out.

1 Like

You’re losing me here. Why would you change your wheel circumference when doing this test?

It really doesn’t matter precisely what power your TR/Kurt/Virtual Power is doing at any given step in the ramp (at 200W it may be truly 213W or something), just that it’s doing the same thing at the same step next time also. And from what TrainerRoad has always said about reliable Kurt power curves I think we can assume it will be same each time (especially if you don’t touch your wheel, tire, pressure, drum tension in between). Then you just have to see the difference between your Stages and 4iii at each step.

When I put the correct circumference in the power outputs were way way out. Probably due in part to the difference between power at pedals and at the wheel. I also want to be able to use virtual power on the trainer so I can put the stages on one of my outdoor bikes. But to do that Ive got to be reasonable certain the power numbers match up

Ah, gotcha. You have the added goal of equating your actual Virtual Power numbers to both your Stages and 4iii. Good luck!

Thanks but I’m not holding my breath

1 Like