Pretty sure Kickr attempts to compensate for drivetrain loss. After all, it’s not a power meter. Might as well build in some compensation so that it’s more likely to reflect the work the rider is actually doing.
1 Like
I’m not sure they do Wahoo sight the reason for power meter differences to be drive train losses
I know with my Neo that I am better off to make sure the drive train is clean before a hard V02 workout, but that could just be in my head “I’ve done everything I can, I will get through this workout”
Blockquote Power measured by a direct drive smart trainer may be affected by losses from the drivetrain. A clean chain and clean gears can help minimize (but not eliminate) this loss.
The corollary of this is that small increments or decreases in FTP are silly. 5% minimum rule.
1 Like
I guess both could be true: my friends and I get nearly identical power with our waxed, indoor only drivetrains, but a dirty poorly lubed drivetrain could still account for another 5% loss.
But you’re right, I just referred back to a thread where I posited this and @dcrainmaker noted that none of the trainer companies say they attempt to compensate for losses. So I’m left with the anecdotal story that I know a handful of people with Kickrs that match their pedals almost perfectly.
2 Likes
I have a 2018 Kickr and a V5, I got the V5 because the 2018 on is a bit of a random number generator, it has terrible drift, and yes I tried everything to correct it but no luck.
2018 - starts about 20% out and drifts sometimes to about 15% but sometimes out to 25% which I can’t understand so along with some other reasons I bought a V5.
V5 - about 10 - 15% out but consistently so, I don’t really care though as I don’t use power from the KickR, as long as it doesn’t drift so I can us slope or resistance mode (or more rarely erg mode)
This 100% plus. Oh and in my case a very big ‘thing’ as I tend to abuse my drivetrain, I’d be on a chain a month to 6 weeks in the summer otherwise.
Agree totally.
2 Likes
Just listening to Episode 244 of the AACC and Nate and Jonathan mention that the 1.5% accuracy is (supposedly PM to PM etc) but that if you are calibrating correctly you should reasonably expect your PM to be consistent.
i.e. if the PM is 2% off, or reading 6 Watts higher, you’d expect it to consistently measure the same 6 Watts at 300W under similar conditions.
This aligns with my experience and what I’ve seen from reviews/comparisons online.
@dcrainmaker @GPLama is that completely incorrect?
That degree of power reading difference between different power meters is pretty typical in my experience.
Different units seem to have a pretty wide range of reported measurements for the same power - ie: calibration of commercially available power meters (and maybe precision) seems to frequently be off by 5-10%.
It’s part of the reason why I’ve moved towards pedal based power measurement. Otherwise, I just used a different FTP for each power meter.
GPLama has some pretty good Power Meter reviews in terms of which ones he’s tested and provide good accuracy, and which ones he doesn’t trust / recommend.
Personally, my Favero Assioma Duo and my TacX Neo2T have been within 1W on every ride I’ve tested.
1 Like
similar story to you. I went from a stages left only to a Quarq and my “FTP” went down 20w from 300w.
very humbling haha…
Minor nitpick: every smart trainer comes with a built-in power meter.
Same here, my previous 4iiii power meter and my Quarq have been very consistent with each other once I factored in leg imbalance. I rarely use my Suito’s built-in power meter, but when I compared them, also those were close. Perhaps a few watts lower, but exactly within the range I would expect.
Not correct… my old Elite doesnt.
Any smart trainer must have a power meter, it couldn’t function as a smart trainer otherwise. The only trainer I know of that doesn’t come with a built-in power meter is Wahoo’s Rollr. I am unaware of any Elite trainer that can dynamically regulate resistance but doesn’t measure power itself.
2 Likes
No. You are not correct.
It uses an algorithm to guesstimate power that is not the same as a power meter measuring power.
My Elite even points this out in the manual, estimated power, still did slope, resistance mode, erg etc. I had a 360w FTP with the Elite, that was fun on Zwift. Lol. Real FTP was about 290w at the time.
I bought a Kickr because I was advertised as having a built in power meter. The model after my Elite was advertised as includes a built in power meter, but Elite had annoyed me by then.
Current releases do, but the first Smart trainers may or may not have circa 2013 /14 (guess they werent really smart trainers but they were marketed as the first smart trainers at the time.)
1 Like
I had this recently I lost about 10-15w going from an infocrank double sided (im generally 45-55, being stronger on my right leg) to a left-only garmin rally XC100.
It kinda sucked, especially since i was so close to hitting my 300W FTP goal 
I swallowed my pride and just carried on. But I think if the difference was up to 30W I would probably change the PM altogether.
No, both measure power, just in different ways. In strain gauges you measure resistances, which are converted to forces according to a device-specific calibration curve. Most trainers use energy balance to infer power, although some of Elite’s trainers use an optical torque sensor. One method isn’t more real or fundamental than the other, in both cases you need compute power from the measurement of another quantity, taking things like temperature into account.
You can achieve comparable accuracies with both methods. Importantly, trainers are also calibrated device-by-device in the factory. That’s and the lack of temperature compensation what distinguishes them from mere speed sensors.
2 Likes
I know. I work in the industry. Most estimate, there is a technical difference, it is all marketing.
I’ll agree to disagree.
They aren’t precision instruments, nowhere near, which is the main point.
I’m a physics professor, I understand how these devices work, too. You have estimates and interpolation curves involved in both, strain gauge-based power meters and power meters that use different methods. These are just parameter-dependent functions that take in resistances or (differences of) speeds and spit out a number. Factory calibration is key for all types to measure power accurately.
Both have similar accuracies, mostly between 1.5 and 2 %. Some of Elite’s trainers promise 1 %, I think. So neither seems more accurate than the other, which makes sense. Higher precision would require more extensive calibration at the least and perhaps other compromises.
3 Likes
No trainer can directly measure power. They guesstimate from the measurements already mentioned, or for example (pedals or cranks), from torque measured in a strain gauge multiplied by measured cadence times known crank length.
I know what you refer to, e.g:
Now the key difference between the Elite Direto/X/XR and the Suito is the lack of power meter within it (called OTS by Elite). But keep in mind, having a power meter in it doesn’t really matter. What matters is whether or not it’s accurate. After all – neither the Tacx NEO nor the Wahoo KICKR series have power meters in them (in fact, Wahoo even ditched the power meter in theirs years ago due to accuracy issues). Of course, nobody is saying the Direto-X/XR isn’t accurate. It is. It’s incredibly accurate.
So this is really just one vendors’ marketing intentionally being confusing. What they’re saying is that the more expensive model has the OTS sensor (explanation here: https://www.elite-it.com/en/innovations/optical-torque-sensor-power-meter-integrated) that’s presumably more accurate than whatever method the lower end model is using to measure power. But that doesn’t mean the cheaper unit doesn’t have a “power meter”. That’s nonsense.
Thank you, I felt like I was banging my head against a wall… which I have given up on. Even the manufacture, Elite says that my trainer does not have a built in power meter, so the disagreement was really with the manufacture of the trainer. You’d think they know there own product 