Re: Re edm software

Hi Tom,

that was exactly, what I wanted to say. The control-software from
this plans (called testware) uses a constant feedrate. This is very
inefficient from my point of view. If I program the feedrate too
slow, it just could go faster. But more often I put the feedrate to
fast, so the wire will run into the workpiece making a shortcut.
This is recognised by the feedback-circuit, which just put a signal
to high, so the software notice this. It then retracts the wire a
few steps and try again.

My idea was, not to program a constant feedrate, but evaluate the
voltage of the edm-capacitors. If it goes too low, the process
should be slowed down. If it goes too high, the "feedback-
calculated" feedrate shoud accelerate.

Alf

--- In CAD_CAM_EDM_DRO@yahoogroups.com, TomP <tomp_tag@i...> wrote:
> Alfewerken
> If this is any help for those on this thread...
> I think he wanted to say (excuse me for butting in...)
> the CNC software he has does not 'servo' (in EDM terms)
> his CNC software merely has a feedrate.
> It is possible to run EDM and use a feedrate, but it is very
inefficient.
> Feedrates are not good for EDM. EDM is a dynamic process that
envelopes
> the CNC, not the other way around. If the CNC is in charge, the
process will
> suffer. The process has to be primary. The ideas like Lag are bad
for EDM.
> Merely stay on course, but the position on that course is entirely
process
> governed. Retraction along the course is common. The progress
along the
> course is more like a cha-cha than a feedrate.
> So there is no way to determine the 'correct position' at the next
time frame.
> That's why Lag is irrelevant.
> I know of no open source EDM CNC software. If you are capable of
hacking the
> segment que in EMC, I'd certainly be interested. I think others
would be too.
> For Wire EDM the approach is very different than for Senkung. I
think it
> would
> be easiest to implement hole drilling ( single axis ) before 2, 4
or 5 axis
> wire.
> Once that was accomplished, Senkung would be next, then the
additional
> synchronous stuff in wire could be done.
> regards,
> TomP
>
>
> --
> No virus found in this outgoing message.
> Checked by AVG Anti-Virus.
> Version: 7.0.300 / Virus Database: 265.6.9 - Release Date: 1/6/05