RE: [CAD_CAM_EDM_DRO] Re: high-performance EDM power supply

Hi Graham -

I can see open loop and slow feed can simplify things, but for the
requirements "wish list", I'd like to list the characteristics of
a state-of-the-art modern commercial system, and then see how
close we can get. A few electronics guys have already expressed
interest, but they need to know more about the process
requirements. For starters, I think we should assume the goal is
a "pulse"-type power supply, as opposed to the RC-type. With this
in mind, can anyone elaborate on the feedback loop? In my
ignorance, I can imagine a few approaches (am I even close?): (1)
constant gap, (2) constant voltage, (3) constant current, or (4)
constant power. I would guess that the g-code would have to
specify some primary parameter to control the burn rate and
surface finish. Would this just be a power setting, or are power
and surface finish controlled independently for a given material?
I suppose it's time for a Google search...

Regards,

Carlos Guillermo
VERVE Engineering & Design

-----Original Message-----
From: Graham Stabler [mailto:eexgs@...]
Sent: Friday, July 18, 2003 7:56 AM
To: CAD_CAM_EDM_DRO@yahoogroups.com
Subject: [CAD_CAM_EDM_DRO] Re: high-performance EDM power supply

>What I found out during 6 months of collecting and reading

articles

>is that the spark (desirable) and arc (undesirable) sensing
>circuitry can be quite sophisticated. Well, that was 30 years ago
>and there are integrated circuits and solid state devices now

that

>are far better. Still, the sensing circuits are very important

and

>these have to be connected in some sort of feedback mode to the
>stepper drivers.

That is certainly true but it MAY be a bit like using steppers.
They
have no feedback but if you go slow enough you don't have trouble
with lost steps. I suspect (guess) that the same may also be true
for EDM in as much as for faster feed rates these controls become
more important. We are hobbiest so we might be able to relax it a
little.

>It used to be that the power supply would charge up a capacitor
>which would discharge causing a spark. This capacitor doesn't

have

>to be large, but it will be expensive as it has to accomodate a

very

>high cyclic discharge rate if you are going to have any speed at

all.

That is for an RC type set up. If you are just wacking short
voltage
pulses on the workpiece then the capacitor can be much bigger and
only used as the smoothing of the DC supply. This pulse type
supply
seems to be used on all commercial EDMS.

Graham