CAD CAM EDM DRO - Yahoo Group Archive

Re: Re: Re: emc and following error problems

Posted by Ray Henry
on 2002-11-29 08:39:41 UTC
>    From: Jon Elson <elson@...>
<s>
> In fact, deadband is a very dangerous thing, as it introduces a
> non-linear transfer function
> inside a control loop.  I would recommend setting deadband to 0.0 in
> all cases.
>
> I have a separate thing in the code that essentially provides a
> deadband to prevent servo hunting.
> It puts the rate generators in stop if the "voltage" is commanded to be
> less than an absolute
> value of 0.001  The simulated DACs go from 0 .. 10 V, so that is .0001
> of max velocity.

Your code writing here sounds like a good thing for your stepper and
servo setups. I can see how the traditional deadband also fed in to the
position loop would work this way.

Just to be clear, It is still a valuable thing for traditional open loop
stepper setups. I was on a project recently where the stepper driver
would not throttle back when idle because it was hunting between two
steps. Both the motors and amps got quite warm. The hint is a noticable
buzz from the stepper motor.

Ray

Discussion Thread

got_milk_eh 2002-11-27 09:46:08 UTC emc and following error problems Askew, Jason 2002-11-27 13:23:38 UTC RE: [CAD_CAM_EDM_DRO] emc and following error problems got_milk_eh 2002-11-27 13:53:17 UTC Re: emc and following error problems Tim Goldstein 2002-11-27 18:39:53 UTC RE: [CAD_CAM_EDM_DRO] Re: emc and following error problems Ray Henry 2002-11-28 06:41:12 UTC Re: emc and following error problems alex 2002-11-28 07:28:37 UTC Re: [CAD_CAM_EDM_DRO] Re: emc and following error problems Ray Henry 2002-11-28 18:13:22 UTC Re: Re: Re: emc and following error problems Jon Elson 2002-11-28 22:28:39 UTC Re: [CAD_CAM_EDM_DRO] Re: emc and following error problems Ray Henry 2002-11-29 08:39:41 UTC Re: Re: Re: emc and following error problems alex 2002-11-29 09:28:23 UTC Re: [CAD_CAM_EDM_DRO] Re: Re: Re: emc and following error problems