Motor angle fluctuation

This is where you talk about the EV3 software itself, installation issues, and programming talk.

Moderators: roger, gloomyandy, skoehler

ruperty
Advanced Member
Posts: 179
Joined: Sun Oct 14, 2007 11:21 am

Motor angle fluctuation

Postby ruperty » Thu Jul 27, 2017 5:21 pm

I am trying to model a single leg standing up guided by an accelerometer value.

As a test I am trying to hold the knee motor at a particular angle. It does hold it for a while, but then drops, as can be seen in this video, before returning to the angle position, within a few degrees. This repeats occasionally.

I am using the NXTRegulatedMotor with the rotateTo method with immediate return, sending the command every 20ms.

Any ideas for why this is happening? Am I just putting too much exertion on the motor with the leg configuration? Would it help if I had gears?

Or is it something to do with the way regulation is done?
Regards,
Rupert
www.perceptualrobots.com

User avatar
gloomyandy
leJOS Team Member
Posts: 5869
Joined: Fri Sep 28, 2007 2:06 pm
Location: UK

Re: Motor angle fluctuation

Postby gloomyandy » Thu Jul 27, 2017 8:14 pm

Firstly there should be no need to send the rotateTo command multiple times. The regulator will hold the position if it possibly can. If it is not doing so then there are two possibilities either the motor has stalled trying to get to that position or if you are trying to hold a position that constantly puts strain on the motor then the built in thermal cut out has cut the power to the motor. Looking at the video it is almost certainly the latter case. Asking a motor to hold a position like that in effect puts max current draw through it, which will probably result in the cut out triggering.

Gears will probably help, even better use a worm gear as this will hold position without loading the motor.
leJOS news https://lejosnews.wordpress.com/

ruperty
Advanced Member
Posts: 179
Joined: Sun Oct 14, 2007 11:21 am

Re: Motor angle fluctuation

Postby ruperty » Sat Jul 29, 2017 11:09 am

gloomyandy wrote:Firstly there should be no need to send the rotateTo command multiple times. The regulator will hold the position if it possibly can. If it is not doing so then there are two possibilities either the motor has stalled trying to get to that position or if you are trying to hold a position that constantly puts strain on the motor then the built in thermal cut out has cut the power to the motor. Looking at the video it is almost certainly the latter case. Asking a motor to hold a position like that in effect puts max current draw through it, which will probably result in the cut out triggering.

Gears will probably help, even better use a worm gear as this will hold position without loading the motor.


Ah yes, worm gears is a good idea. I'll try and re-engineer. Thanks.

And only send commands when the value changes.
Regards,
Rupert
www.perceptualrobots.com

ruperty
Advanced Member
Posts: 179
Joined: Sun Oct 14, 2007 11:21 am

Re: Motor angle fluctuation

Postby ruperty » Thu Sep 07, 2017 7:20 pm

Here's the prototype using worm gears and perpendicular gearage.

https://youtu.be/ofN-VrVyKmM
Regards,
Rupert
www.perceptualrobots.com

User avatar
gloomyandy
leJOS Team Member
Posts: 5869
Joined: Fri Sep 28, 2007 2:06 pm
Location: UK

Re: Motor angle fluctuation

Postby gloomyandy » Thu Sep 07, 2017 8:36 pm

That looks pretty good, nice and stable. Not often you see so many gears being used with a Mindstorms robot. Could you post a close up of how the gearing works?
leJOS news https://lejosnews.wordpress.com/


Return to “EV3 Software”

Who is online

Users browsing this forum: No registered users and 2 guests