Diminutive, a small EDM prototype

For the last little bit I’ve been working on a small EDM prototype for use on large parts, pipe sections or for use in the field. While they do make and sell these things, the vast majority of them require the use of vibratory assistance, aren’t terribly accurate and are relatively low powered. I intend to make one that is accurate and could potentially be used as a ram EDM in a pinch. Also, I’d like it to be able to use copper and graphite electrodes so polarity reversal is necessary.

This is the linear drive of the small EDM. It has 8 inches of travel and is surprisingly rigid and play-free. I was going to use a gear motor on the top with a rotary encoder but since the loads are quite small a stepper should be OK, even at higher speeds.
This is the linear drive of the small EDM. It has 8 inches of travel and is surprisingly rigid and play-free. I was going to use a gear motor on the top with a rotary encoder but since the loads are quite small a stepper should be OK, even at higher speeds.

I’m currently trying to come up with ways to mount this guy on magnetic bases but also allow for small amounts of fine adjustment so that if there is a need for precision work, the unit can be dialed in. Right now the coupling is made of two pieces of steel. I’m going to replace one of them with nylon to reduce noise. I was going to use a spider coupler with rubber in between but I didn’t have one kicking around on the weekend when I was making it so I just made my own.

It's all made from 7075 aircraft aluminum. Pretty strong stuff. The movement was initially pretty stiff because the screw, which is just hot rolled, was a little bit out so I had to run it back and forth to wear in the brass acme nut. I may put two acme nuts into the travel block simply to take up play, as it stands though right now there is less tan .0003" of play, even with significant force applied.
It’s all made from 7075 aircraft aluminum. Pretty strong stuff. The movement was initially pretty stiff because the screw, which is just hot rolled, was a little bit out so I had to run it back and forth to wear in the brass acme nut. I may put two acme nuts into the travel block simply to take up play, as it stands though right now there is less tan .0003″ of play, even with significant force applied.

So next things next, the prototype electronics. Right now I’m simply going to use what I have laying around, so a TM1638 display, a cheap chinese stepper driver with a toshiba stepper driver and I opted for an Arduino to drive the motor. I will be making a full fledged PCB that will nicely fit in a box but I need to get the pulse generator and sensory circuits working first, then I can integrate them fully on to one PCB.

Just a rough testbed for the motion control and a simple interface. I would like this thing to be nice and easy to use.
Just a rough testbed for the motion control and a simple interface. I would like this thing to be nice and easy to use.

Well, anyways, I have quite a bit more work to do for this project. Even a single axis CNC can be a little more tricky since it’s doing more than simply positioning. A ram EDM is a dynamic machine that reacts to the sensory input to keep the cut running smoothly and with appropriate voltage and gap settings.

More later.

A look at the waveforms from a CNC EDM

I decided to try out my new Hantek DS1060 and get some waveforms from the head of the EDM (Electrical Discharge Machine). The probe was hooked to the head of the machine and the ground was connected to the frame. I only took a few measurements but it may be interesting to some to see what the pulses look like when it’s cutting. This cut was burning at 75us on and 37us off.

This is the waveform while cutting at 75us on and 37us off. 8A current
This is the waveform while cutting at 75us on and 37us off. 8A current

You can see that the voltage doesn’t drop to zero, and there seems to be a dropoff after the ionization path has started to cut the material. These dropoffs after the spike seems to indicate that the path is ionized and current control has kicked in, providing 8 amps. Once the time is up, the current shuts off and then there is a ringing spike, perhaps due to inductance. From there the voltage doesn’t reach zero but probably sits at a continuity testing voltage for a little under the 37us. From what I can tell, 75us is started from when the ionization path is established, and 37us is the time it cuts off the current, event though there is still voltage (and maybe a bit of capacitiance).

A view of the measurements taken from the oscilloscope
A view of the measurements taken from the oscilloscope

The measurements seem to correlate to the settings a bit. The EDM itself is set for a 240V cutting voltage so the VMax of 181V is pretty close. I think those values are a quite flexible, I know that before it cuts at all, the voltage is at approximately 240VDC.

Anyways, I thought this would be interesting to post since some people are into EDM machines. I may take some waveforms from the wire EDM and two more different EDM’s. Here’s a final image of a zoomed in view of the peak.

A wider view of the burning process, the short ringing spike must be a check for proximity
A wider view of the burning process, the short ringing spike must be a check for proximity or from inductance