Tacx Neo 2T Vibrations At Moderate Power Due to TrainerRoad?

Equipment
1

When using TrainerRoad on my Tacx Neo 2T in erg mode, I feel a strong vibration / pulsating resistance during the down stroke on each pedal at moderate powers (150+W) and typical cadences (80-100rpm). The vibration goes away when using the Tacx Training App’s Power Workout and manually setting the power. Only a slight vibration is felt at 250W. Is it even possible that TrainerRoad’s erg mode would control the trainer differently than Tacx Training App’s erg mode? If so, can TrainerRoad be set up differently to address the vibrations?

More Details:
During TrainerRoad workouts, the vibrations are hardly felt at power under 100W. But as the power increases, the vibrations become more pronounced particularly during the down stroke on each pedal. Increasing the cadence at a given power reduces the magnitude of the vibrations a bit. At 200W and 95 rpm, the vibrations are very annoying and become the focus of the workout. Also, shifting to a lower gear helps a bit. So it seems that higher torques are a driving mechanism.

I have tried the following to reduce or eliminate the vibrations without any improvement:

  • Changed chains keeping chain on small chainring and middle cog
  • Changed bikes going from a 10spd 105 mechanical to a 12spd Ultegra Di2 drivetrain in the process (others having this issue have mostly been on 10spd drivetrains)
  • Cleaned the interior of the trainer by removing the disk with a Garmin extractor tool
  • Replaced the main bearing assembly
  • Replaced the cassette body
  • Took off the motion plates

After all of these failed, I tried the Tacx Training App. The vibrations practically went away, even up to 250W. Since the Tacx app is on my phone which connects to the trainer via Bluetooth and TrainerRoad is on my tablet which connects to the trainer via ANT+, I tried connecting to the trainer via Bluetooth from TrainerRoad on the tablet and the vibrations still occurred. I am going to compare the Tacx app to TrainerRoad again tomorrow to make sure the behavior is consistent as well as try resistance mode through TrainerRoad.

I am going to follow up with Garmin support if there are no ideas on how TrainerRoad could be causing the issue.

Any suggestions on what else to try?

Thanks,
Jim

References:
Anyone have the Neo 2T?
Vibration in the pedals when in under load?
NEO 2T Issues with cassette alignment and low pitched rumble/grinding in high gears
Vibrations with Tacx Neo 2T trainer

2

I have encountered this issue in the past when using my road bike (11 speed shimano) on the NEO 2T with Trainerroad. I tried my gravel/commuter bike (also 11 speed shimano) and it was perfect so I can only assume it is a slight misalignment of derailleur and the NEO cassette. Very annoying and difficult to diagnose, I went down a rabbit hole of swapping cranks and cassettes eventually gave up and only use the gravel bike on the trainer.

1 Like
3

Have you tried the small/big ring up front? Have you tried every gear on the cassette? I ask because you’re in erg and you mentioned that shifting made a difference. My guess is that it’s something like mentioned above where something is misaligned. That won’t explain why it’s different between the two apps, but it may solve the issue.

4

Interesting. Back when I was on the sewer named FaceBook, and in a variety of bike related groups, I remember a poster making the argument that there was ‘different flavors’ of connection to a trainer. They swore that under one app, they had vibrations and noises, and under a different app those issues went away.

Technically there wouldn’t seem to be any difference between the app connections. It’s a simple Bluetooth (or ANT+) connection, the trainer shouldn’t give a hoot, it just (tries to) carries out the commands, it’s what it does. Predictably, they were shouted down as being silly and ridiculous. I thought it could be the flywheel speed that the app insists on during the demand that the rider is doing. Riding ERG means the system will achieve the demand at the current setup. Too low a gear, the flywheel is likely to be churning at a higher rate. Ahh, but the 2T uses a synthetic/simulated flywheel. That’s when it all falls apart. Could the 2T be running a simulated flywheel at a higher rotational speed as it hits the power set point? Why would it? Does the 2T physical flywheel actually change RPM during the changes in demand? Someone suggested that it was harmonics created by the magnets and magneting potential across the many magnets in the physical flywheel.

I have the Neo Smart Bike, and have never experienced any vibration that could be linked to magnetic resonance.

But on the 2T, is it up to the latest firmware?

5

I can feel what I think OP is describing on my OG Neo when I get into the smallest rings in the back. I’ve wondered before if it has something to do with magnetic frequencies. No data, just sharing that I can make my Neo feel “different than it feels to ride on smooth road” when I start grinding.

1 Like
6

This kind of sounds like the Neo’s road feel feature or whatever it’s called. Did you try turning that off in the Tacx app? Or maybe on if it was off just to try something different?

TR has mentioned that different apps do control the trainers differently. So doing ERG mode in TR workout may actually feel slightly different than if you were doing the same TR workout in Zwift.

7

Why would TrainerRoad drive a trainer differently than Zwift, or another app?

8

I don’t know. Response times, ramp rates, compensation factors?

From the TR/zwift integration FAQ:
Can I still do ERG mode workouts in Zwift?

Yes, though ERG behavior may feel slightly different than in TrainerRoad.

Why does power ramp up slower in Zwift compared to TrainerRoad?

Zwift’s power transitions are more gradual than TrainerRoad’s. This is normal and won’t significantly impact your workout quality or Adaptive Training.

1 Like
9

Isn’t/wasn’t there a ‘Zwift factor’? Like Zwift fudges the set point/watts exerted? I seem to remember something about that when I started riding Zwift. Now that I think about it, I’m not sure what the issue was then, and maybe it’s not there anymore.

10

Thanks for all of the suggestions. Here are my responses to the questions asked along with an updated set of tests.

The alignment of the drive train (chain, chain ring, cogs, derailleur) seems to be fine on both bikes that were tested. I don’t see any interference between the chain and gears or hear any rubbing across the gears evaluated (I avoided cross chain and the large cog which causes the derailleur to touch the trainer disc). Also the “vibration” term is probably not the best descriptor as the feeling at the pedals is more like grinding / pulsating, not pleasant. That said, I will spend some time aligning the drive train this weekend just to make sure they aren’t at least partial sources.

Note that the two bikes that both reproduced the issue are very different from one another.
Bike 1: Aluminum, quick release, mechanical 10spd 105, rim brakes (I used the suggested spacer ring behind the cassette per the Tacx 2T instructions)
Bike 2: Carbon, thru axle, Di2 12spd Ultegra, disk brakes (No spacer ring behind the cassette but used the spacer on the other side per the instructions for disc brakes)

I have tried small ring, big ring, and most cogs (avoiding cross chaining). Generally, the lower the gearing, the less the issue or the higher the power required to elicit the issue.

The Tacx 2T is using the latest firmware per the Tacx Training app, 0.0.38.

The Road Feel is set to 0% in the Tacx app. I don’t think TR enables road feel even if set.

I agree that it doesn’t make sense that different software communicating with the trainer would produce different behaviors, but my (limited) observations is that the issue does not arise with the Tacx app whereas it does with TR, though not all the time as explained below. My understanding is that in erg mode, the software just sends a command to the trainer to set a certain power target. It is then entirely up to the trainer to control how that power is maintained. In this case, which software sending the command shouldn’t affect the response of the trainer. The only thing I can think of is that the Tacx app is also providing a set of control parameters. Most likely the trainer is using a PID or similar controller. It is possible (though unlikely) that the Tacx app is sending a specific set of PID/control algorithm parameters than the default parameters that are used when other apps connect. Interestingly, I tried the following this morning

  1. Started a workout in TrainerRoad on the tablet in erg mode and continuously increased the intensity to push the power up during the initial warm up. After passing 160W, the vibrations could be felt and become more intense going up to 250W. Stopped the TrainerRoad workout, discarded it and closed the app.
  2. Started a power workout in the Tacx Training App on the phone and continuously increased the power to 250W. The pedal feel was smooth as one one expect for a trainer. Stopped the workout, discarded it, and closed the app.
  3. Opened TR on the phone, connected to the Tacx 2T (I think I had to disable Bluetooth on the tablet to connect), started a workout, increased the intensity so that the power reached 250W and no vibrations :thinking: Stopped the workout, discarded, and closed TR.
  4. Opened TR on the tablet (I think I turned Bluetooth back on while the app was running), started a workout, increased the intensity to 250W and no vibrations :exploding_head:

So now I am quite confused. WAG: perhaps the control parameters were set by the Tacx app that prevent the vibrations and they were not overwritten when subsequently opening TR as I don’t recall stopping pedaling during the testing. I need to repeat these steps and take detailed notes next time to try to isolate whether the app or connection order or connection type is having an effect. Of course the differences in the apps / connection could just be a red herring. I will revisit on Sunday or probably Monday as I need to rest for a 100+mi ride tomorrow with a group of younger and stronger riders that will definitely result in a red day for Sunday.

1 Like
11

Hey @jimshep :slight_smile:

Thanks for all the details! Please give us the update if you’ve tested things already.

I have a feeling that maybe resetting the BLE on your tablet might have cleared any corrupt data, and that’s why you felt no vibrations after turning it back on, because if the TR app was working well on your phone, then the issue seems device-related instead of app related.

I don’t know if you’ve tried it yet, but you can always try:

  • Deleting and downloading the TR app again on your tablet.
  • Checking our Minimum System Requirements
  • Make sure your tablet is running on the latest software upgrade.
  • Make sure you’re not running on low battery or have a power save mode on.
  • Make sure to close off any other open apps on your tablet so the TR app is able to run smoothly.
  • Turning off and on your table can sometimes get rid of corrupted data that is preventing the TR app from running smoothly.
  • Checking you have Permissions enabled on Android:
  1. Open up the settings on your Android
  2. Go to the “Apps” section
  3. Find TrainerRoad and select it
  4. Click on “Permissions”
  5. Make sure the Location permission is enabled

Note: These are the generic steps for changing Android permissions, so this process may vary slightly on your device.

12

It looks like the primary issue was the TR setup on the tablet. Running TR on the phone, the “vibrations” don’t occur, even at high power. And after reinstalling TR on the tablet, the vibrations are gone when running TR from the tablet. So the issue seems to have been resolved.

Last week when investigating the issue further, running TR on the tablet presented weird behaviors I haven’t seen before. When connecting to the trainer via ANT+ (which is built-in the Samsung tablet), the power from spinning was only around 85W and would only increase a little with increased cadence even though it was in erg mode and the workout power was set to over 120W. Switching to resistance mode didn’t allow the power to be controlled any better through cadence or gears. Switching back to erg mode and increasing the resistance level from 100% to 150% had no impact. I then disconnected the ANT+ version of the trainer device then connected to the Bluetooth version. The power was then stuck at 250W regardless of cadence or resistance vs erg. 250W was the previous power setting I used on the Tacx Training App on the phone (which I disabled Bluetooth to make sure it wasn’t interfering). Very strange. I eventually deleted TR from the tablet, restarted the tablet, then reinstalled TR as Caro had suggested. Now TR performs as expected on the tablet (as well as the phone) and the “vibrations” are still gone.

I am shocked that the issue seems to have been software related as the issue felt very mechanical. Guess it’s a potential side effect of software controlled electromagnetic trainers when the software flakes out. At least now all of the bearings are running very smoothly and the drive trains of two of my bikes have been optimized which I probably wouldn’t have done otherwise.