I am definitely experiencing the same issue described by some of the folks here.
Kickr Core (1.5.5) via BLE FTMS + Garmin CAD sensor via ANT+ (cadence reported by Kickr is too choppy because I’m a novice cyclist and my pedaling smoothness is nonexistent).
Happens only in SIM mode with Virtual Shifting ON – ERG mode is buttery-smooth.
Happens in any RPM, but most noticeable in low RPM “uphill”, and the lower the virtual gear the worse it gets – 60rpm in vGear 2 on a +7 grade feels like alternating between stomping and spinning out every few revolutions while the PWR display dances between 30 and 200W!
Real gearing is 24t front, 21t rear – I use a 2x8 MTB as my trainer bike. Zwift Cog that came with the trainer did not align out of the box and did not work with the provided 8spd shim, so I had to improvise and purchase a secondary 8spd cassette for trainer use.
No IRL shifting, that’s what virtual gears are for, right?..
Here is how a 10min flat-ish portion of a ride looks like, believe me, inclines are even worse. The fluctuations are not display-only, I can feel this constant power wobble through the pedals and it’s incredibly annoying:
I noticed there were at least a couple other reports of this issue under Bugs that went unanswered:
spiky-power-with-consistent-cadence/631744
kickr-core-virtual-shifter-bug/629376
pulsating-resistance/619048
P.S. The Start Chat button on the Contact Us page doesn’t seem to work, so I can’t even start a support case on this.
I’ve been paying closer attention to pacer rides. Typically a rider increases and decreases power to maintain a certain position in the pack, and because of dynamics/ churn, sometimes that needs to be a strong surge or a substantial reduction in power. What I’ve noticed is that when using manual shifting, the surge with any given cog is much more responsive and immediate compared with doing the same in any given virtual gear. So over the course of a ride, the additional (“unrecorded”) effort with virtual shifting seems to be much more than manual shifting. After an hour or so, the average reported power for the same pacer is very close, but the perceived effort is massively different.
You answered yourself. Your front sprocket is too small. Sufficient flywheel speed is required for stable operation. Your problem is getting worse, that’s understandable. The slower the flywheel rotates, the more power the trainer needs to hold it. The faster it spins, the easier it is to maintain power stability.
Happens in any RPM, but most noticeable in low RPM “uphill”, and the lower the virtual gear the worse it gets – 60rpm in vGear 2 on a +7 grade feels like alternating between stomping and spinning out every few revolutions while the PWR display dances between 30 and 200W!
Thanks, that sounds logical - I will try my large sprocket combo, which will be 36f/16r but it is much noisier than having the chain mid-cassette with the small sprocket in the front.
My point, however, was that I’m only experiencing this drastic power wobble in virtual shifting – I never had any issues with pedal feel when using conventional shifting, even when climbing a 9% grade (100% trainer difficulty) in the granny gear (24 front - 40 rear)!
Anyone found any resolution to this oscillating power issue? Has Wahoo replaced any trainers, and did that solve it?
Just got a Kickr v6 and seeing this exact same issue. Small ring 36 and 14 on cassette. Problems is less noticable in big ring 52, but I would prefer to stay in small due to less noise.
Hi Steve, I’m getting exactly this issue now on my Kickr V6.
Power drops to zero briefly when I stop pedalling then RPM spikes up and Watts indicate 5-20W for maybe 10-15secs then settle at zero. Rinse and repeat every time I stop pedalling.
Done one factory spin down which fixed it for one or two rides but I had now crept back in.
Has yours stayed good even with virtual shifting on all the time since Wahoo replaced the unit?
Hey Sam, my partner had the same problem. Basically, every time she did sprints in training she had to do another factory spin down… Wahoo have sent her a new one the other day!
Its got much better since being replaced, It happens rarely, and interestingly when my partner uses the trainer… very similar to Josh.
I wonder if we have different pedalling imbalances, but then again any trainer being shared such as in a gym, would exhibit similar behaviour.
Other than all my previous tests and attempts to isolate the problem, I also wonder if hard sprints knock it out of balance and causes some flywheel wobble.
If it happens on certain courses after cresting a KOM, it can be a real pain as it prevents super tucking and puts you at risk of being dropped.
I have heard that Wahoo are looking at it and have consolidated a few tickets.
I wonder how many trainers with tthis symptoms are out in the wild…and with no Duals to compare the trainer drift many users don’t know it and they think they get stronger.
Well I have a ticket open with them now so we will see. So far they asked for the normal, reset all connections, do factory spin down then complete the 7min ERG test in Wahoo App.
I sent all that back today; but with the comment that they likely won’t see anything wrong since it is straight after the spin down.
Next time I see the issue I will redo the 7min test before doing the spin down.
I don’t know if you can video yourself, that’s what I did and uploaded it to Youtube as unlisted. I showed both my pedals and the screen.
I recorded a sprint effort and then coasting and showed the watts went to zero but then jumped back up to 15 to 20 watts.
That helped, and Wahoo agreed to replace the trainer
There’s also the second issue of power oscillation when using Virtual shifting. It only happens in the small ring and at high power, I use a 46t front chainring so it doesn’t affect me but I’ve seen it.
For these reasons (and so I could verify outdoors), I bought a Assioma MX1 SPD power pedals.
Its left sided power only and I have a pedalling imbalance which favours my right side, so there is a small margin of error with the dual recordings. However, I was able to match and exceed my benchmark power outside, so fairly happy the KickR is accurate.
There is also an issue knocking about that when using virtual shifting & the trainer calibrates whilst the chain is in the top few gears it over reads by 30-40w - Think Wahoo have raised this as an internal issue and have an open ticket on it.
I just found that thread. I don’t use high mechanical gears and keep a straight chain-line. Is this in reference to auto calibration (e.g. spin down) when coasting in virtual gear 22/24?
This sounds unlikely, I can’t imaging many people spinning down from such a high gear (unless they have TD set at 0). I tend to sprint in 19/20 max.
I genuinely don’t know - to me, there seems to be a rather large number of users with suspiciously low cadence for mid to high power points knocking about which suggests high gear usage but that might very well be 2+2=685 on my behalf.
Yeah the overall issue is around auto calibration.
Not sure if this is relevant or of any use, but I’ve just skim-read this thread and also the “Wahoo Trainers with virtual shifting issue - free watts” thread. I’m seeing some distinct similarities I’ve been having since purchasing a Zwift Click in April. My set-up is a Core trainer, hybrid 2 x 10 bike, running Zwift on a Zwift-dedicated W10 PC. I’ve been communicating with Zwift Support for many weeks and so I’ll distill what was the issue and what we learned below:
Issue: When using virtual shifting, (a) resistance was swinging wildly especially on steeper climbs (approx. > 10% gradient) - maybe what you are referring to as “oscillating power”?, (b) average power estimates about 50W higher than normal (210W vs 160W). and (c) Kickr Core-estimated cadence dropping on steeper climbs (40rpm instead of 60-70 rpm).
Things learned: This did not seem to be a connection issue nor a device issue - problem was replicated regardless of Bluetooth pairing or Companion pairing, and regardless of whether Zwift was running on my PC, Android phone or Apple TV.
The latter lessons were learned through sheer perseverance and a long series of trials on the Grade.
Through happenstance, we eventually realised the problem occurred due to LOW GEAR RATIO. Although Zwift virtual gearing should be tolerant of a range of mechanical gear ratios, the problems occurred when I was on the small chainring and middle cassette cog (gear ratio ca. 1.5). The issues went away completely when I rode on the large chainring and middle cassette cog (gear ratio ca. 3.0.) This resonates with your comment about power oscillation when small chainring and high power are combined - this is exactly the circumstances that caused my issues.
I now have a Zwift Ride (which has a gear ratio of 3.0 - 42t/14t), and virtual gearing continues to be problem-free. No resistance surges and slippages, no power overestimation, and accurate cadence at all times, including the steepest gradients.
I could have written a 10,000-word treatise on all the testing I have run to try to solve this issue and hope this “brief” explanation isn’t too long. Happy to answer queries.
I can confirm using the smallest front gear is causing oscillating issue. My front gear is 48/38/28T as it’s a mountain bike. When I installed the zwift cog, I followed the picture on the manual that seems to indicate to use the smallest one in front. I think the picture has be done only for 2x crankset.
However, i found recently on this forum the official recommendation
I do have in front 46 and 38.
since I do not have the Zwift cog, for a straight chain line I have in front 38 and in back 19. With this I’ve this mentioned wobble.
Will double check virtual gearing behavior again with 46 / 14 or 15 as @David_Rowe3 suggests
