Re: Re: EMC & Wire EDM?

On Monday 07 April 2003 10:59 pm, Dave wrote:

> Message: 24
>    Date: Tue, 08 Apr 2003 02:58:54 -0000
>    From: "turbulatordude" <davemucha@...>
> Subject: Re: EMC & Wire EDM?
>
> --- In CAD_CAM_EDM_DRO@yahoogroups.com, "inetd_conf" <benedict@h...>
>
> wrote:
> > I've been considering tackling a wire EDM.  I'd prefer four axis
> > control, but honestly I think I'd be doing good with two.
> > 
> > My CAM software will handle wire EDM just fine (yeah, I'm using
> > Vector).  But I don't currently have control software that'll do
>
> the job.
>
> > 
> > I was planning on building a power supply similar to the one Robert
> > Langlois described in his book published by Village Press.  In
> > particular, I liked that it had a signal line that would go high if
> > the wire drew too much current (shorted), and another that would go
> > high if it drew too little (stopped cutting).  In his book, which
> > described a ram EDM, these two lines controlled the ram.
> > 
> > A wire EDM is a little more complicated.  If the wire shorts, you
> > don't just retract the ram, you have to stop or move the wire back
> > along the toolpath.
> > 
> > Ideally I'd like to use software that's designed with an EDM in
>
> mind.
>
> >  I'd like it to be able to listen for high or low current draw, and
> > act accordingly.
> > 
> > If anyone here has made a wire EDM, I'd be curious what your
> > experiences were.  I'm not going to tackle this without a lot more
> > thought on the matter, so the more you can tell me the merrier I'll
>
> be.
>
> > 
> > Thanks,
> > 
> > Tom
>
> I've been thinking on the same lines.  what is needed is a software
> that will accept the input from the voltage monitoring circuit of the
> EDM to adjust cutting path speed.  the math of where to go is done by
> the software, but speed is dependant on the EDM circuit.
>
> Another feature you mention is cutting path reversal.  on a sinker,
> that is simple, one axis, but on a wire, it is 2 axis reversal.
>
> Some high end packages look ahead 150 steps of code or so, in order
> to maintain very smooth movements.  this would need look ahead
> and 'remember' the last "X" number of lines for that reverse path.
>
> Master5/Mach1, has recently introduced the first step but it was
> really designed for plasma cutting.  ie: the plasma cutting circuit
> moved the head up and down, but the software keeps tabs of the exact
> location.
>
> There is a need for this, and hopefully the answer is not in the high
> dollar software packages.
>
> Dave

Roland (Cardinal Engineering) and I have been working on a plan to retrofit
his two axis Agie with the EMC. We've got the idea that it might make a nice
article for HSM. EMC is easier to do this with because it's source is open
and the price is just right.

Monitoring two pins like you suggest and issuing feedrate commands would be
trivial with processes like tkio and iosh which are already in the code.
Watching an analog spark current signal is a bit more trouble but still
doable.

Reversing using the g-code is not trivial if you are using wire diameter
offsetting but it would be fairly easy if we used a toolpath and file routine
like tkbackplot uses.

There have been threads on control for this both here and on the emc list.

Ray Henry