Why is Vector 3 dual all over the place and Flux 2 fixed?


I just did a comparison between my pedals and my Flux 2 to see if I need to correct my ftp when winter comes. I have an issue that when using my pedals as power meter I get an delay in ERG mode with up to 15 sec which doesnt work with shorter interrvals. It seams like with the powermeter in the trainer this time gap gets smaller.

When looking at this picture, is this becasue trainerroad doesnt mesure as often as Garmin does or is it because it is less accurate?

The red is trainerroad and Flux 2S and the blue is Garmin Vector 3 dual wirh Garmin edge 520 plus

The pedals are showing the actual true power here and the Flux is showing that power smoothed. You will never pedal well enough to make a line like the Flux with your pedals. Nobody does or can.

For shorter intervals, many recommend resistance mode rather than ERG dude to the lag on some trainers. Also, the new PC app and iOS app (currently in beta) have improved Power Match that has resolved this for many, so I’d suggest trying one of those if you have access.

1 Like

Ah, that was what I thought. But does that mean that the flux never measures the actual power but rather the 5-10sec avg?

I did this workout in the phone today and last time directly on my Garmin. Will try the Mac desktop app next time and see if that works better. Thanks :pray:

I’m not entirely sure how exactly the flux measures this, but I know with the Kickr, it just reports whatever power TR is telling it to produce. So if it is 250 watts, it will hold that resistance with some sway up and down as normal, but when it reports the power to the head unit/phone/laptop/tablet, it will just report a flat amount equal to what TR is telling it to hold. In other words, what it reports is already smoothed out.

Most trainers have a switch on the software side that can remove ERG smoothing. This would be on the trainer side, rather than TR. Here’s a comment from TR about this specific thing for more info:

ah, that sounds like what is happening here. That was my guess, that the trainer smoothes things out.

That causes a problem also with the delay. If I go from 150W to 320W for 40SEC it will smooth out the first 15 seconds and the avarage will be lower than what I actually put out. Not a big deal on longer effort but really enoying on shorter once.

This is not the case with the Flux 2, and you can see it in the graph – there is variation in the power reported, just much less variation than with the pedals.

I use the Flux 2 without an additional power meter (and have done the same kind of comparison with an external power meter as above) pretty regularly and am very sure that it is not simply reporting the erg mode set point. Not only do you get variation in output power on constant intervals, but at interval edges you can definitely get power readings that are far from the erg mode set point.

It’s tricky to claim one measurement as “the real power”. They are different measurements, since they’re measuring power in different ways and locations. The pedals are also showing power readings that have some degree of smoothing, since the torque that the strain gauge measures varies across the pedal stroke.

The Flux 2 may be doing some software smoothing to its power output. (It’s not nearly as long as 5 or 10 seconds, though.) However, you will get a degree of smoothing, relative to the pedals, entirely from the mechanics of how they measuring power – there’s inertia in the drivetrain and trainer that damps the highly variable output of your legs.

Yeah that’s why I mentioned in the part you omitted from that quote that “I’m not entirely sure how exactly the flux measures this.” I based that comment specifically on my experience with the Kickr as a reference point.

All true. My point was less about absolute accuracy of the power and more about the true fluctuations in power that occur. Many riders come from a smoothed out display of power to pedal or crank-based power meters and ask why it is so much more up-and-down, when in reality power is just never really as smooth as most trainers report.

I’m not talking about accuracy, though.

The point is that singling out the measurements taken by a pedal or crank power meter and saying that is “true power” and is “reality” is not really correct. They’re both real power measurements, just on different sides of a high-inertia system. This is going to be true regardless of how the trainer chooses to report its measurements.

But to report the power that TR tells it to be in erg mode sounds really stupid. Then it doesn’t tell you at all what power you are actually putting out. Can that really be tru for the kickr?

It’s always measuring your input and comparing it to what it is outputting, but if you have ERG smoothing turned on it will report whatever the app is telling it to hold. You can see this on a lot of people’s workouts in the workout thread.

Here is a better and more clear explanation (in 2 minutes) from GP Lama using a Kickr Snap: