On Sun, 16 Aug 2015 07:36:19 -0500, Pete Keillor
wrote:

On Sat, 15 Aug 2015 17:59:24 -0700 (PDT), "
wrote:

On Saturday, August 15, 2015 at 6:00:47 PM UTC-4, Pete Keillor wrote:



I'd still need to build the carriage, stepper drive, controls, etc. I
have a Beaglebone Black which could be applied to that job. I can get
a reasonable stepper for $11 from the salvage place in town, plus some
linear bearings and a bellows coupler for not much.

Any known reasons the welder is a bad idea?

Thanks.


There are some DIY edm plans on the internet. You might look at some of them for ideas.

Many years ago, somewhere around 1969 a friend and I kludged up a edm. We used his drill press for the feed by setting the depth nuts and then leaned on the handle to feed the tool a very small amount. As I remember we used a good sized maybe 500 VA transformer and I think a voltage doubler. A light bulb in series with the supply had low resistanece when things were working and a higher resistance when we had the tool shorting to the work. A kludge , but it worked and we could put square holes into tool steel.

We were charging the cap to about 300 volts. So a bunch higher voltage than your welder would supply, but the plans on the internet all use lower voltage.
So in answer to your qoestion , I do not know.

Dan

Thanks all. Lloyd, yeah I'm retired and I think it'd be educational,
plus I already spent an hour or two picking and pecking at it.

Good ideas, Ed. I might combine that with Dan's drill press idea for
a test. I'll try the welder at much lower voltage. If that doesn't
work, then I'll ge further into Lanlois' design using some big caps.

Pete Keillor

Here's another bit of inside info that may be good to know about --
not that you're trying to build a real EDM, but it may be useful to
know how they do it.

When they're using a hydrocarbon dielectric (kerosene, for all
intents), the voltage required to initiate a spark ranges from 90V to
300V. Keep in mind that their parameters are for producing a highly
accurate hole with minimum damage to the workpiece.

Around 1980, Agie had the world by the tail with a patented power
supply that initiated a spark by using a *string* of lower-voltage
pulses at high frequency. The voltage probably was around 90V. The
Japanese avoided the patent on their export machines. They didn't
worry about it on the machines they sold in Japan. g

Anyway, Sodick, for example, broke the process into three steps,
without the patented pulse circuit. There is more than one way to skin
that cat. First, a high-impedance 300V circuit polarized the channel
between the electrode and the workpiece, producing "stringers" of
ionized dielectric that reduced resistance for the second step.

The second step was a medium-impedance circuit at 150V. This one began
to turn the liquid dielectric into a low-resistance channel that
delivered enough amperage to turn the dielectric into ionized gas. As
it heated up, it became a plasma.

The final step was the money step. This one operated more like a
welder, at least in terms of delivering power. This was a very
low-impedance circuit (nearly a short circuit, with a big capacitor
supplying the energy) at around 15V - 30V. This step in the power
delivery used the plasma channel to deliver an extremely concentrated
spark, which was delivered to the workpiece at a rate of thousands of
amperes per square inch. The actual average amperage delivered through
the whole cycle might be 30A.

Anyway, that's how they did it then. I don't know how they do it now.
But it's useful to know that real EDMs employ *sparks*, rather than an
*arc*, because an arc is difficult to shut off and it can damage the
workpiece.

For a tap buster, that's less of a problem. You'll get sparks, and
also some arcs. You shut off the arcs the same way you stop it in arc
welding -- by pulling the electrode away from the work enough to break
the arc.

In real EDM, the retraction is only a few thousandths of an inch, at
most. That's not enough to stop an arc, so there is a lot of
power-supply and servo design directed at producing sparks but not
arcs. Again, that will be less of an issue for busting taps, so you
don't need all of that complication.

But it's good to know what's going on physically in those machines.

--
Ed Huntress