RE: individual velocities

["Tim Goldstein" <tgoldste-at-mlh-services.com>]

>> From: Tim Goldstein <tgoldstein-at-mlh-services.com>
>> The problem with setting MAX_VELOCITY to 60 is any G0 move on the Z
>> axis will be implemented as a move at 60 ipm.

> I know it's hokey, but you could just not do G0 moves with the Z axis. As
far
> as I know, G0 moves are interpolated just like feed moves in the EMC. Old
> time controls could only interpolate toolpaths up to a velocity that was
> usually a fraction of the speed that the axes could be driven if you
weren't
> concerned about what path they took. Now, with much faster computers, we
can
> interpolate at rates sufficient to keep up with the capability of
available
> axis drives (the ones available to me anyway) at resolutions typical in
> machine tool control.

I have been trying to avoid this solution as it is a real pain to recode the
g-code. I am using Vector as my CAD/CAM program and it makes frequent use of
G0 on all axis's when not actually cutting.

>> Matt had mentioned that you are suppose to be able to throttle back an
axis
>> now (as of the Jan-2000 build) by putting a parameter in the individual
axis
>> entries under MIN_OUTPUT and MAX_OUTPUT. I have tried this on my Z axis,
but
>> it just seems to cause me follow errors that I can't get around.

> You'll need to set FERROR to a larger value. It probably won't hurt
anything
> since the following error gets really large only when you command
velocities
> higher than your lagging axis can achieve such as during rapid moves when
> (hopefully) you won't be cutting anything. Another unfortunate side effect
of
> using the MIN and MAX OUTPUT limits is that jogging in manual mode is a
> little dicey unless you set back the feedrate override before you move.
The
> reason is that the commanded position gets way ahead of the actual
position.
> I guess that letting up on the jog button should send down a commanded
> position equal to the current actual position, but that's:

I will try upping the FERROR setting and see what is does. Thanks for
pointing out the consequence in relations to jogs. That could be a big
problem if you are not expecting it.

> 1. Not implemented.
> 2. A feature that may have other side effects I'm not aware of.

> Anyway, maybe another solution would be to specify (in the .ini file) the
> velocity to use for G0 moves for each axis, and to stop interpolating G0
> moves. I think Fred has this issue on his "things to do" list, but I'm not
> sure what his current thinking is about how to handle this issue. As far
as
> acceleration goes, I don't know what amount of work on the motion system
> would be required to implement per axis acceleration values.

I think I would like the idea of an ini parameter for G0 on a per axis
basis.

> Matt

> P.S. The MIN and MAX OUTPUT limits are specified for stepper systems as
> units/sec so for 24ipm, you'd have them set to + and - 0.4, and 1.0 for
> 60ipm. You'll want to set these for X and Y as well.

That is how I tried it, but I did not change the FERROR setting.


Tim
[Denver, CO]

timg-at-ktmarketing.com <timg-at-ktmarketing.com>
http://www.ktmarketing.com