Power Match seems OK but not great - tips and tricks I'm missing?


#3

Thanks but that’s not me. Here a ride - https://www.trainerroad.com/career/stpatterson/rides/46568881-kaweah
It says power source CABLE which is how I’m syncing my Quarq.

My issues are not bad enough that I’d turn Power Match off, but bad enough I have a suspicion it could be better :wink:


#4

The key to a good ERG experience is keeping a steady cadence. Let the trainer give you the break, don’t slow your pedals to rest. If you need a hard rest stop peddling completely for ~5s, or even better back pedal. What happens is when you change your cadence gradually the trainer starts chasing the power that it wants to hold and the torque requirements get all screwed up. Even cadence and it all ‘just works’.


#5

If you go into an easier gear (36x25 or equivalent) to slow down the flywheel, do you find it tracks better?
If I wind my 2014KICKR up in a 52x13 or so I find that the flywheel is too quick and the power readings seemingly go to crap with power match.

The other thing is just keeping consistent cadence so that the program and trainer aren’t trying to hunt for optimal force to meet target wattage.


#6

Got it. So taking a look at your ride, I suspect the issue is to do with ERG, not Power Match.

In your working intervals the power tracks really well to target. It’s when you’re in the recovery intervals that things get out of control. I suspect what’s happening is that your trainer is trying it’s best to match your power target, but your power is all over the place and it looks like it’s all over the place because your cadence is going nuts.

But that can’t be right. I don’t think a cadence of 225 is real. So, then the question is why does your trainer think you’re doing all these wacky cadence numbers. Maybe other users with a setup similar to yours might have some good insights. The pattern is clearly when you start your recovery intervals, I’m guessing you just let off the power completely and slow spin for a bit. The best advice I can give is to not do that :slight_smile: I know that’s bad advice on the surface, but do everything you can to just ride at a consistent cadence when the watts come off, at least for 20-30 seconds until your trainer settles on the new power. If you still need to back it off, I think it’ll handle it better from a more settled starting point.


#7

This is a key point also. 50X16 will be spinning your flywheel very fast with a lot of inertia. When the power drops off it will take a while for that to settle in, combine that with your legs letting off the gas and I think you’re seeing some extreme yo-yoing of power in your trainer as it tries to find equilibrium. At @RONDAL suggests, try a test using low gearing and trying to keep cadence steady and see if that improves things.

Here is a thread on gearing in ERG mode:


#8

@julianoliver for what its worth, here is a zoomed in (14 minutes) portion of my Monday workout (Bluebell) showing 6 x 1 min intervals at 120%

53x15 gearing on Kickr 2017 and using PowerMatch with Stages LR. When dropping from target of 294W to 98W it takes 5-10 seconds of lightly spinning to catch up.

Overall training in 53x15 works well despite the faster spinning flywheel. Power doesn’t “go to crap” as per earlier comment, and I can quickly cover the jump from 98W to 294W. Sometimes I throw down a little too much power going from off-to-on, but you can see the 3rd and 6th intervals I did better at finessing the off-to-on jump.

Phone screenshot of overall workout here: What workout did you do today?


2016 Kickr Power Drift
#9

I suspect the issue here is bad cadence data, not choice of big vs little chainring.


#10

The fairly steady 101 on the work interval is real but the funky part you note is mostly a result of me coasting for several seconds then pedaling at a not particularity steady cadence. I’m also definitely having some issues with the CABLE and the cadence which show up in the rest interval. I spin but not 225 :wink: The cadence gets hung up and pretty regularly does not drop to zero when i stop pedaling and it really gets wonky if I coast for a bit then resume pedaling. I posted on that a while back and the official TR reaction was just suck it up as its too hard to optimize the program to work with “every” (i.e. 2 or 3 total) Ant/Bluetooth bridges out there :frowning:

Does the cadence reading itself get factored into power match? if so, that right there could be a big issue with my setup. Also, it occurs to me if the cadence is not syncing right, the power reading could be a little sticky too.


#11

I agree with this point - there are two problems being caused by this

The first is common and not worth stressing about - when you drop your cadence post interval the power will drop below the new lower target. This is fine and not a big deal

The second is semi-unique to your setup which is that something related to your cadence post interval is broken. The cadence (and thus power) being transmitted by your DFPM is very very messed up and that’s causing TR to chase all over the place. You can mitigate this by following the advice elsewhere in the thread of not dropping cadence and being consistent all the way through but really you need to find and address the cadence issue with your power meter


#12

both power and cadence look wacky for 1 minute in “this looks funky” section. Even in bigger gearing with even faster spinning Kickr 2017 flywheel, my 1 minute rest intervals at 98W are pretty clean.


#13

Totally agree. I have Stages in power match with Kickr18 and I find everything tracks really well regardless of gearing. For @STP things clearly aren’t working so well, so to try and mitigate the problem my suggestion is to remove as many variables as possible. 1) Keep cadence consistent so that power is more consistent 2) Try a low inertia gearing so that the trainer will find it easier to “catch up”. I’m not blaming the gearing as a cause, but rather as a step to help diagnose and isolate the problem.


#14

Ok, can’t hurt and I agree.

Doubt that is going to help with clearly bad cadence data, and it sure looks to my eyes that we are seeing bad power data too.

And thank you for confirming I’m not crazy - most of the “flywheel inertia” comments make it seem that physics have been repealed in my garage. I’ll step off my soap box now.


#15

I don’t believe it is directly factored. Indirectly, for sure. Power Match does what it sounds like, TR is telling your trainer to take and match the power reading from your power meter. So, if your Power Meter is giving crazy readings, your trainer will go crazy trying to track to it because of ERG. Cadence has a direct effect on power reading, since power is cadence times work. If cadence rises or falls and work stays constant, your power will track to your cadence. ERG will then register the increase in power and let off resistance to track to target power. So, you can indirectly relate cadence to your trainer making adjustments due to its influence on power readings.


#16

I’ve done no such thing :wink:

More seriously on the topic of gearing, I think there is a trade off to be made. High gearing will make hard efforts slightly easier since you’re getting the positive effects of inertia in the flywheel, but that same inertia means the trainer will also be less responsive to changes in wattage. I don’t think one is better than the other, but you can’t optimize both. What I am an advocate of is consistency, I don’t advocate high gearing for the ramp test and then low gearing for training for example.


#17

LOL, not letting me off easy hu?!

In my experience, high gearing also means the Kickr is applying the brake more often. Its not as easy as everyone makes it out to be.


#18

I’m definitely getting bad cadence readings on my iPhone using the TR app and the CABLE as a bridge but I’m certain the cadence readings are a CABLE/TR issue and the bad readings are some type of sampling issue introduced by the CABLE or the TR app, not an issue inside the power meter. I have never had an issue with weird power readings outside (or weird cadence readings either). (The CABLE works fine with Zwift and my Apple TV).

I’m definitely going to try getting out of the big ring and see what that does for me.

If the cadence thing then seems to be an issue that is affecting power match, I’ll pull out the laptop back out and ditch the CABLE. Cadence and power worked fine on that for the last couple years. I just prefer the recently discovered convenience of the iPhone app so I;m not excited about doing that.

Thanks everyone!


#19

It’s not the cadence readings, but your actual cadence that could be causing your poor experience.


#20

@STP good luck and let us know if switching to little chainring helps at all.


#21

Update - Tried the suggestion on gearing and that helped a lot. Out of the big ring now and using 34x16 and the power match is more stable. It definitely feels like it responds quicker. The lower fly wheel inertia also helps on the rest intervals as I don’t have to wait as long for the power to drop (it also helps when slowing cadence to stand). The feel is not quite as “road like” as the higher gear but still fine and I stopped noticing it after a few minutes.

Separately, my cadence reading issue got a ton better after a firmware update to the CABLE and the latest TR app update. Both updates ended up being installed at the same time so not sure which one did the trick but there has been a significant improvement. The cadence reading is more responsive and the app will now pause when I stop pedaling whereas before the cadence would “stick” at whatever the last reading was when coasting.


#22

Hey great news, I’m happy to hear that things have improved for you. Training is hard enough without feeling like you’re fighting your setup. Thanks for sharing your experience.