Hi all, I got the Zwift ride, play and core last week and have been plagued with this since. I installed the latest firmware straight out the box. I did a factory spin down on Thursday evening (braking factor 0.92), which cured it, then did a race Friday and was pleased to see it wasn’t an issue at the start of the race, although came back 10 minutes into the race.
Has the theory that the auto calibration is causing it, rather than curing it, been explored?
I read the article, and it was not because if this issue, but others. Att least my understanding
that’s not good.
We don’t know if its auto-calibration causing it, or auto-calibration exposing it.
I’ve done quite a bit of testing isolating variables. Including Belt tension, Virtual shifting, chainring gear ratio, pedalling imbalance, flywheel balance, software, device, wireless interference, temperature.
There are two conditions that I have found tend to trigger it
In both cases you need virtual shifting activated. Mechanical shifting des not trigger it.
My suspicion (as per the first forum posts that started to appear from Feb/March) is that under extreme conditions virtual shifting messes with auto-calibration and triggers an unresolvable miscalibration.
Unfortunately, we don’t know for sure. Only Wahoo will be able to understand how their firmware and algorithms work.
We also don’t know if other manufacturers are susceptible to the power offset, but we have no reports of ghost watts.
What we do know is that once the power offset is triggered it remains stored on the Wahoo KickR device and it is then present in all platforms (indieVelo, MyWhoosh, Wahoo app) even if you then use mechanical shifting
Hmm!
sounds familiar 
So after the news coming from ZWS should we stop using virtual shifting?
The latest Wahoo firmware did not fix the issue.
Apparently this is a Wahoo problem, but it should still be a priority for Zwift as well since it impacts the fairness of racing as well as the general accuracy of the user experience.
It seems like maybe there are tolerances within the virtual shifting processes that allow it to work fine in most situations but with gears/cogs on the extreme ends the calculations get off. The most recent firmware maybe expanded those tolerances a bit. But the free watt issue is not solved.
I tried the factory spindown. It’s impossible for me to do it directly from the Wahoo App (i tried tapping 10x many times but no access to the spindown process), but I can do it with the Wahoo Pro Spindown App (may be because I got a kickr Move which shouldn’t need any factory spindown?).
Anyway, the spindown works appently well but when I try on Zwift (with or without connecting the Zwift Play), the ghost watts remain.
I have tried a dozen time (before a race, after a race, with or without the ZP connected… actually in any possible configuration). The problem remain…
I’m not familiar with Wahoo Pro Spindown App it might not be available in my region.
here’s the link to the Wahoo’s instructions for the factory spindown. The KickR Move is included.
Make sure your device is not connected to any other app. - It might help to forget it in any other apps you are using.
If you still can’t get it to work, I would open a ticket with Wahoo. Their customer service is very good.
Thank you. I’ve tried many time and my device wasn’t connected to any other app. I’ll open a ticket…
In case anyone is interested, I replicated the issue today. 8W differential between my Assiomas and my Core when going up the Grade. This was with the latest Kickr Core firmware upgrade. (1.5.36, released October).
I am also noticing higher Wahoo Kickr Core readings with virtual shifting compared to my Assioma pedals. Very similar to what you noticed – traditionally, the Core would be 2 watts lower on average on longer efforts, but now I am seeing the Core read ~8 watts higher than the pedals on sustained efforts.
I’ve done a factory spindown + restart, but it seems to still persist.
I’ve raised the issue with Wahoo. They asked me to run a test with the trainer in erg mode run from the Zwift app and another head unit recording a different PM. I ran this once but my second head unit did not record any power?! I came down with COVID, so I’ll try to run that test when I recover.
Not sure if any fixes were released, but I am pleased to report that I have gone from getting ghost watts every race (after spin down each time) to no issues for the last 3 weeks, very happy!
I thought to share my solution with you… I saw around +20w ghost watts in Kickr Core data and assumed Kickr is having issues, but not my Favero Assioma Due. I was wrong… After a lot of spin down on Kickr I checked my Favero manual. It turned out I did not allow 1mm gap been the pedal/power meter casing and the crank arm. I placed a pedal washer and now kicker and Assioma are reading very close to each other.
Any solution/advice for this problem? Wahoo says they are working on it yet… Not doing super stuck on descends until 10-15 seconds.
Thanks.
Hello. The issue of “sticky watts” or “ghost watts” is a well-known problem with many (I don’t know if all) Wahoo trainers. Despite attempts to fix it with the latest update, manual respin, reset, etc., Wahoo acknowledges that the issue is not resolved and “is still working on it”, without specifying a timeline or if they will succeed. This is particularly annoying at the start of descents, where both watts and RPMs fluctuate and take up to 25 seconds to drop to zero, preventing you from performing a supertuck enough time to get dropped from the group in races.
For this reason, I believe that Zwift could make a small modification to the requirements for performing a supertuck.
Right now, one of the requirements is to have watts below 11. Well, there would be at least two very simple solutions for Zwift to implement:
Keep the 11 watts or, alternatively, require 0 RPM. This way, an external cadence sensor, which are very cheap, could be used, and the problem would be solved.
Increase the maximum required power to 40 watts or similar (no one races at 40 watts) instead of the current 11.
I believe either of these options would solve this problem, at least temporarily, until Wahoo fixes this issue that has been bothering Wahoo users for a long time.
Regards, and I look forward to your opinions, proposals, or alternative solutions.
Option 3: What about removing the supertuck or giving it no effect? Which would follow the UCI rules?
Not proposing any particular option.
I do have a wahoo trainer and do get impacted, but being light it is the difference between being a bit slower or a lot slower downhill than everyone else.
You’re gaining as much on the climb as you’re losing by not having supertuck. If you can’t remove the extra watts there’s no way that you should have the only disadvantage from the bug taken away.
I was able to fix it on my Core shortly after it was reported with update firmware / factory spindown, but I went back to my Kickr V5 as my primary trainer (gave the Core back to my wife) as I didn’t like virtual shifting and the Direct Connect was more important to me, so I don’t know if it reappeared.
Also - you do know that you can’t supertuck if you’re in the draft anymore? Supertuck or not should have nothing to do with being dropped from a group, only the lead rider should be supertucking.
You may be seeing expected behavior of no supertuck until you’ve dropped out of the draft given your description.
I suspect this is the case - with the latest firmware my Kickr Core can supertuck where it’s supposed to (when on the front, (but not in the draft) on a steep enough slope at a high enough speed), not sure that there’s a widespread wahoo issue with supertuck if you’ve got the latest firmware.
