Re: emc for wire edm control

[Dave Engvall <dengvall-at-charter.net>]

Michael,
Oh, Ouch! If an '8' can do it then it doesn't take much processor power.
That is 12 bit code done in octal. Not too many instructions.
I might still have a book on an 8 around here someplace.
PDP/16 code is pretty straight forward.

Dave Engvall

On 3/13/03 8:43 AM, "Michael Coffey" <mdynac-at-prodigy.net> wrote:

> 
> Ray Henry wrote:
>> 
>> Marcel
>> 
>> I have done a bit of work here and with Roland Friestad planning for such a
>> retrofit.  He has a two axis Agie wire EDM that is a few years beyond factory
>> service.
>> 
>> We had planned to use an analog signal for wire current but your approach
>> also sounds interesting.  This could simply be a voltage up signal
>> optocoupled to a parport pin and read by tkio.
>> 
>> One way to achieve the reverse path is to simply record the positions of each
>> axis for each motion loop.  This is how we ran the first backplotter.  The
>> second plotter does the math to make a vector from a whole series of points.
>> 
>> With a file like this, when the parport pin goes low/high the grapical
>> interface could abort the running program, switch to mdi mode, and issues
>> each point back in the recorded position file as a g0.  Repeat this until the
>> pin clears.  Once cleared, we could restart the cut at the activeline when
>> the short happened unless we crossed position into a previous block of code.
>> If that happened we would need to decrement the activeline by 1 and restart
>> there.  
>> 
>> Cutter comp may cause a few problems but a few tests with the backplotter
>> should show a good approach here as well.
>> 
>> Ray
>> 
>> 
>> On Wednesday 12 March 2003 12:18 pm, Marcel wrote:
>> 
>>> Hi
>>> I want to know if its possible to change emc to control a two axis wire edm
>>> machine. A wire edm machine need to be able to go back when the cutting
>>> voltage is low (shorted) and only start to go back to the cutting direction
>>> when voltage is high Regards
>>> Marcel Gombault
>>> 
>>> This email was scanned by the ICL anti-virus solution.
>> 
>> 
>> 
> Not all Wire EDM's "back up" during a shorted condition.
> For example, most Japax wires go into feedhold during short,
> and then continues along the contour after the short is cleared,
> usually by flushing,
> Sensing the voltage at the gap is the best way to achieve this.
> A simple, yet precision comparator/opto circuit works nicely here.
> Setting the circuit to output at, say 20 gap volts, is a good starting
> point. Then tweak it for best performance...
> In my opinion I would get the machine up and burning using this
> approach, then delve into the more complex task of actually making the
> machine backup in a shorted condition.
> 
> 
> I am currently working on a retrofit for wire/sinking/hole popping EDM's
> Both servos and steppers.....
> 
> I have several old 2-axis "blue" Andrew 330's and 630's with servos,
> these macines do backup during short, a DEC PDP8\a is the heart of the
> axis motion on this beauty.
> Does anyone know how to dis-assemble pdp8\a a code???
> 
> and an old Japax 4-axis machine with steppers, this machine feedholds
> during short
> 
> Hope this helps....
> Michael
> 
> 
>