# Electronic discharge machine

Discussion in 'The Projects Forum' started by shortbus, Mar 17, 2011.

1. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697

steveb - Could you please link to a circuit that can do this? Don't want to hijack this thread, so should I start a new thread linking to this one?

I have a project that would benefit from doing this, an EDM - electrical discharge machine, for my shop and have been trying to find away of doing this without buying a larger transformer.

2. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Suggestion, start another thread asking this question. Simple variance of resistors (charge resistor vs. discharge resistor) suggests itself.

Apr 20, 2004
15,815
283
4. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
Bill, this is what I am doing now, but the discharge resistance can't be changed, it is a spark gap. steveb was saying that there were ways of getting a higher discharge from a capacitor with out increasing the charge.

From all I can find if you put a certain amperage into a capacitor you can get a smaller amperage out,but not a larger amperage. The way my machine is now my transformer is 20 amps output, I'd like to get 40-50 amp discharge from the cap bank, and thought he might share how its done.

5. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
You are assuming using the same charge and discharge path, which is OK. Without a schematic I can only guess how your doing it.

There are circuits that might still do this though. I routinely see 100A Shottky rectifiers, if such were around a resistance using multiple Shottkys then the resistance for current going one way would be different than the other direction, and it could handle currents in the ranges you were referring too.

I think in terms resistances because that is the easy way to control a charge or discharge rate. It isn't the only way, you can use a bank of capacitors with switching circuits to multiply voltages, which will also cause variable current. Charge 2 capacitors from a source, then put them in series with each other. It is still related to resistance. More voltage, more current through a load.

6. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
Bill thanks for the reply, I'll draw up a schematic of what I'm doing and post it.

7. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
You should write a PM to steveb to point this thread out to him.

8. ### steveb Senior Member

Jul 3, 2008
2,433
469
That was a good idea Bill, as I wasn't aware of this thread. My ability to monitor and respond is limited because I'm on vacation now. My Internet access is now an iPod when I find wireless hotspots.

Yes, if resistance is not changed then current control is more limited.

Usually a charging method and path are different than the discharge method and path. A simple method might involve a diode to change the resistance based on direction.

Current control often requires transistors, pwm and current control feedback. If a simple resistor and voltage source is used, then current is not constant unless the voltage source is much larger than the cap voltage, but then much power is wasted in the resistor.

A forward boost reverse buck dc to dc converter is sometimes used to efficiently control current for both charging and discharging.

Note that I'm not an expert with practical circuits at hundreds of amps. My job requires me to model systems at hundreds and even thousands of volts and amps, but my practical hobby experience and work related hands on experience is limited to tens of amps. So it's difficult for me to recommend actual circuits and transistors (mosfets igbts etc). Also, some information I have access to I can't release because it is proprietary company information.

Sorry for brief answers here, but it's hard to type on an iPod. I'll think on this and post some more later.

9. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
Sorry it took so long to get back to this thread. Here is the schematic for the EDM that I'm building.

The pulse (square) wave that is turning the gates of the mosfet on/off is turning the charge switch of one capacitor bank on while the other bank is discharging. Thats what the square wave drawing is supposed to be showing.

The spark gap is actually a shaped graphite electrode that is used to cut that shape into a metal work piece.

The four position switch and resistors, in the upper right corner is used to limit the current for different size electrodes.

Now in the thread that this one came from it was said you could get more amperage from a fixed supply by just capacitors. Could/would some one explain how to apply that to this circuit?

The switched resistors are used to lower the amperage but what would raise the amperage and NOT lower the voltage? The voltage in the spark gap must remain the same to ionize the dielectric fluid and initiate the spark and removal of the metal in the gap. I'll include a picture of what is going on in the gap.

Any help is much appreciated.

File size:
92.2 KB
Views:
275
File size:
291.9 KB
Views:
58
10. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
It appears to me you already have variability, via the various switches on the caps and the resistors controlling the charge rates. Am I missing something?

Feb 5, 2010
3,854
968
12. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
In the thread that this was moved from Steveb said that getting more amperage out than put into a cap was simple and easy to do. I would like to be able to get more amperage for doing roughing type burns (the name for making a EDM cut) The only way I know of is to increase the amperage of the transformer.

Other wise, does this circuit look like it should work? In the real machine there are six capacitors in each bank, and the switches are DPST so what is on in one bank is also on in the other.

13. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
Have that saved in my favorites He started with the same plans I did (Langlois/HSM) It is what they call a RC discharge type, what I'm making now is a Pulse EDM. With the RC type the finish on the part is not constant, the particles of eroded metal sometimes make a lager spark happen and that can put a deep pit in the surface being machined.
With the pulse type the size of the spark is controlled by the charge and discharge switches and not by the resistance in the gap.

Here is a bad photo of the head that I built for my machine. Been a ongoing project that I'd like to finish before I die

File size:
204.3 KB
Views:
40
14. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Their are two ways to get more current out of a capacitor. Both are relatively simple.

Lower the resistance of the discharge path. The lower the resistance, the higher current and shorter duration the pulse is.

Increase the charge voltage. This will leave the duration of the pulse more or less the same, but dramatically increase the current.

15. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
Bill, this was my understanding of how things worked, but the other tread made me think maybe I was missing something. The voltage and discharge resistance are two things that in the process of EDM are pretty much fixed values.

If a second transfomer of the same size/type was added in parallel, that would work, correct? Say two 100V - 20A transformers (which I have) would give, 100V - 40 A to the circuit, correct? Thanks for your help.

16. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Are you going to use the extra transformer to increase the voltage or the current to the base power supply?

You may also want to consider adding some other MOSFETs in parallel to reduce the resistance on the discharge path. Or...

Using MOSFETs to switch a second bank of capacitors to bump the capacitors from parallel to serial, multiplying the voltage.

17. ### shortbus Thread Starter AAC Fanatic!

Sep 30, 2009
4,098
1,697
The extra transformer would be to increase the current, thus using them in parallel. The EDM process doesn't like higher voltages.