You are here

un-commanded reverse pulse

6 posts / 0 new
Last post
maxhurtz
Offline
Last seen: 1 year 7 months ago
Joined: 2021-02-11 11:38
Posts: 1
un-commanded reverse pulse

Strange behaviour on my Trampa VESC 6 MKV.

I'm using PPM input with Duty Cycle control and when I am going forwards and brake suddenly (release the stick), it goes into reverse for about half a second.

If you look the data, the duty cycle ramps up to about 55% (in the opposite direction for which you were going).

Only happens in one direction e.g. if it happens when you are going forward and brake, then it doesn't happen when going in reverse.

Running at 12S (approx 50V), motor is a 6880 (16 mOhms).

The only way I have found to stop it happening is to reduce the Motor Current Max Brake e.g. from -75A down to -25A. But I would like more braking.

I have six VESCs. Same behaviour on each and same behaviour on a smaller motor.

Thanks for any help.

frank
Offline
Last seen: 1 day 21 hours ago
VESC BronzeVESC FreeVESC GoldVESC OriginalVESC PlatinumVESC Silver
Joined: 2016-12-27 20:19
Posts: 847

For vehicles duty cycle control is not a good method. Typically you would use Current Control only.

We can try to replicate that, but I would advise to switch to Current Control anyway.

 

canoeman
Offline
Last seen: 2 weeks 2 days ago
VESC Free
Joined: 2020-03-23 15:48
Posts: 34

Quite a few people have noticed this duty mode bug on FW5.3. It eventually caught the attention of the esc gurus over on the discord who replicated it and sent a video to vedder. Then vedder implemented a fix in the beta firmware, which seemed to resolve the issue. Here's one clip of what they were talking about to make sure it matches your problem:

dutymodebugdescription.PNG

I have myself replicated this uncommanded reverse pulse during sudden stop in duty mode on the stable version of FW5.3 numerous times. I then upgraded to the beta version and can confirm that the problem is fixed. To get the beta version you would need to download vesc tool beta as shown in the following clip.

betadownload.PNG

Then upgrade firmware on one of your vescs using vesctool beta. It will automatically grab the beta firmware that includes (among hundreds of other changes) the fix for the duty mode bug. Remember to completely redo your motor detection/configuration rather than just loading the motor configuration that worked before.

maxhertz
Offline
Last seen: 6 months 2 weeks ago
Joined: 2022-01-25 01:50
Posts: 14

@frank - Current control is not suitable for most vehicles. As the speed of the vehicle increases, the controller will try to maintain the same current, steadily increasing the power supplied to the motor.

Duty cycle is much more familiar - it accelerates towards a speed dictated by the input, similar to the behaviour of a brushed motor controller.

maxhertz
Offline
Last seen: 6 months 2 weeks ago
Joined: 2022-01-25 01:50
Posts: 14

Thank you canoeman, most helphul.

canoeman
Offline
Last seen: 2 weeks 2 days ago
VESC Free
Joined: 2020-03-23 15:48
Posts: 34

Hope this fix works out for you. I'd add that current control can work well for electric skateboards and ebikes, since smooth, controlled acceleration is critical to function. For multirotor drones and general robotics applications duty cycle mode is usually best. We do a lot of heavy ugv work, which is why I've been tracking this issue closely.