CAD CAM EDM DRO - Yahoo Group Archive

Re: RTLinux lessons, and EDM's

Posted by Ian Wright
on 1999-10-19 11:15:35 UTC
Hi gang,

Jon, Wire EDM, I'm fairly sure, works only by monitoring arcing current
between the wire and the work and adjusting the feedrate in the X and Y axes
to maintain a constant, preset, current. This obvioustly involves a separate
monitoring circuit which feeds back into the CAM software at some point
halfway between the file input and the output to the machine. I am not a
programmer, but I suspect that it would not be too difficult to hack a
version of EMC to do this. If anyone has a commercial Wire EDM machine with
circuit diagrams I would be very interested to see scans or copies of these.
Other than this monitoring, a wire EDM is just a 2D mill with a continuous
vertical wire feed, spool to spool, in place of the cutter head. Whether the
situation on a wire EDM machine changes dependent on the thickness of the
workpiece I am not sure as I have never actually driven one. It would
obviously have to advance more slowly on a thick workpiece than on a thin
one but, unless the feedback current is averaged over a short period of time
rather than using instantaneous readings I cannot see how this would work -
presumably, at any one time you only have one spark passing between the wire
and the workpiece and this will give an instantaneous current reading, but
how the machine knows whether you need 1000 sparks to cut through a 1" thick
workpiece or 10 to cut through thin sheet I do not know. Perhaps someone
else can explain?
The vertical plunging EDM is an altogether simpler animal. The simplest
versions use a solenoid to push the ram down and contact with the work
shorts out its drive voltage and makes a spark as it retracts and breaks
contact - just like the 'interrupter' on an old sparking coil or magneto. A
stepper driven one can be easily made from the stepper out of an old 5.25"
disk drive powering a screwed rod driven 'ram'. This is controlled by
something like a SAA 1024 driver IC where the 'step' input is a continuous
train of pulses from a 555 IC oscillator and the 'direction' input, normally
pulled 'high' by a resistor, is grounded when the ram touches the workpiece,
so reversing the direction of drive. This is a bit simplistic as there needs
to be some circuitry to prevent the spark voltage from turning the SAA 1024
into a smoking pit. The more complex commercial machines do use a current
feedback system similar to the Wire EDM machines to control the quality and
dimensions of the holes produced but it is usually done completely within
the machine itself as it does not rely on an outside computer for commands.

Ian
--
Ian W. Wright
Sheffield UK


----- Original Message -----
From: Jon Anderson <janders@...>
> I'm not sure how wire EDM works, but suspect it tries to maintain a
> constant level of arcing. Part of the retract cycle in a ram EDM is to
> facilitate flushing, not a problem with wire.
> I wonder if an EDM module could be setup to work either way. Speaking as
> a total electronics dimwit, I see the circuit as need to discriminate
> the level of arcing to a much finer degree to accomplish this than for a
> simple ram setup. Also, in a wire setup would the parameters change when
> cutting 1" vs .125 stock?

Discussion Thread

Brad Heuver 1999-10-19 06:06:09 UTC RTLinux lessons, and EDM's Jon Anderson 1999-10-19 08:02:16 UTC Re: RTLinux lessons, and EDM's Ian Wright 1999-10-19 11:15:35 UTC Re: RTLinux lessons, and EDM's Jon Elson 1999-10-19 11:54:20 UTC Re: RTLinux lessons, and EDM's Jon Elson 1999-10-19 12:03:46 UTC Re: RTLinux lessons, and EDM's Jon Elson 1999-10-19 12:25:19 UTC Re: RTLinux lessons, and EDM's Ian Wright 1999-10-19 15:42:21 UTC Re: RTLinux lessons, and EDM's Paul Corner 1999-10-19 17:34:44 UTC Re: RTLinux lessons, and EDM's batwings@x... 1999-10-19 07:56:34 UTC Re: RTLinux lessons, and EDM's Jon Elson 1999-10-19 21:13:45 UTC Re: RTLinux lessons, and EDM's Ian Wright 1999-10-20 12:06:24 UTC Re: RTLinux lessons, and EDM's