Re: [CAD_CAM_EDM_DRO] EMC Following Error
Posted by
David Micklethwaite
on 2002-09-15 04:16:29 UTC
Jon,
encoders, 8tpi leadscrews and 2:1 ratio because I was concerned about the
motors ability to move the table). I am making new pulleys to get rid of the
2:1 ratio. If the motors can move the table I will halve the steps/inch and
increase the speed.
steppermod and hould be commented out. I have done that. Is it required?
that only EMC runs, screen updates and mouse tracking are very slow. I guess
I need to reduce that 32000 steps/inch. I reduced the velocity to 0.5 ips
and also dropped the acceleration to 1.0 ips**2. This appears to have got
rid of the following errors and the table moves around at reasonable cutting
speeds (rapids are slow).
following error. This puzzled me because it has no feedback. I didn't think
that it would stop because it couldn't handle the step rate. I guess I
expected it would reach some speed (less that I asked for) and run at that.
Regards,
Dave.
David Micklethwaite
> David Micklethwaite wrote:drivers
>
> > I am trying to get EMC working with a small mill. I have Gecko G320
> > and am using steppermod. I can jog the axes back and forwards by theamounts
> > I expect and in the right directions. However, I occasionally getfollowing
> > errors and do not understand why.get the
>
> Steppermod does not simulate a servo system, as freqmod does, so it can't
> following error from anything having to do with the acceleration, PIDgains, etc.
> But, it DOES have a limit of how fast it can put out pulses. Your inputscale is very
> high, 32000 steps/inch. For 1.2 IPS (Inch/Second) velocity, that requires38,400
> steps/second to be issued from the computer.I knew it was high - it is the result of using what was available (2000 cpr
encoders, 8tpi leadscrews and 2:1 ratio because I was concerned about the
motors ability to move the table). I am making new pulleys to get rid of the
2:1 ratio. If the motors can move the table I will halve the steps/inch and
increase the speed.
> You don't report what your PERIODI thought I had read somewhere that the period parameter was not needed by
> parameter is set to, but it needs to be REALLY high.
steppermod and hould be commented out. I have done that. Is it required?
> I think it needs an absolutevelocity a
> minimum of 38.4K x 2 = 76800 per second, or 0.00001303 seconds/interrupt.
> But, this will make the jump between the next lower velocity and top
> 2:1 jump (from 19200 steps/sec to 38400) which will drive steppers nuts.an 800+
> Maybe the G320 will not have a problem with this, but it might still cause
> problems in getting a steady speed, as the steps will be jumping around in
> timing. Anyway, that period comes out to 13 microseconds, meaning the CPU
> gets an interrupt for the stepper task every 13 uS. I think you'll need
> MHz Pentium III, minimum, to run at this rate and still keep the rest ofEMC and
> Linux functioning.Hmmm. Don't have an 800+ MHz P3. This is a 700 MHz Celeron. I have noticed
that only EMC runs, screen updates and mouse tracking are very slow. I guess
I need to reduce that 32000 steps/inch. I reduced the velocity to 0.5 ips
and also dropped the acceleration to 1.0 ips**2. This appears to have got
rid of the following errors and the table moves around at reasonable cutting
speeds (rapids are slow).
> If by following errors, you mean that the Gecko drives trip out, then youracceleration
> parameter is WAY too high. 20 Inch/sec/sec is More appropriate for ahigh-end
> pen plotter than a machine tool!No, the Geckos don't trip out. EMC goes into e-stop and pops a window saying
following error. This puzzled me because it has no feedback. I didn't think
that it would stop because it couldn't handle the step rate. I guess I
expected it would reach some speed (less that I asked for) and run at that.
Regards,
Dave.
David Micklethwaite
Discussion Thread
David Micklethwaite
2002-09-14 16:49:45 UTC
EMC Following Error
William Scalione
2002-09-14 17:44:19 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
Tim Goldstein
2002-09-14 19:34:33 UTC
RE: [CAD_CAM_EDM_DRO] EMC Following Error
cadcamcenter
2002-09-14 20:21:52 UTC
Re: EMC Following Error
cadcamcenter
2002-09-14 20:32:09 UTC
Re: EMC Following Error
Jon Elson
2002-09-14 22:28:02 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
bsptrades
2002-09-15 00:41:02 UTC
Re: EMC Following Error
David Micklethwaite
2002-09-15 04:16:26 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
David Micklethwaite
2002-09-15 04:16:29 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
David Micklethwaite
2002-09-15 04:16:31 UTC
Re: [CAD_CAM_EDM_DRO] Re: EMC Following Error
wanliker@a...
2002-09-15 09:13:04 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
William Scalione
2002-09-15 13:25:48 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
Ray Henry
2002-09-15 17:05:18 UTC
Re: RE: EMC Following Error
Jon Elson
2002-09-15 22:08:56 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error
David Micklethwaite
2002-09-16 04:17:01 UTC
Re: [CAD_CAM_EDM_DRO] EMC Following Error