Re: Another EDM Approach - Ref Digest Number 1521

To the Wire EDM and Ram EDM enthusiasts - an offer

I notice some recent posts about EDM systems on the list

I've had this in mind for some time but, being primarily a mechanical type,
haven't done anything about the electronics - However, I do have a Wire EDM
machine and a Ram EDM machine available for experimentation - If anyone
would like to have a cooperative arrangement to develop these systems I
would be glad to offer the use of the equipment for testing purposes -

The wire EDM is an AGIE which served me well for nearly 20 years but
developed a serious case of "electronic aneurism" a few months ago caused by
a nearby lightning strike while running (not covered by insurance - I wonder
why I pay the premiums sometimes) -

The RAM machine is an older INGERSOLL unit with a vacuum tube power supply
that I haven't run for many years -

A few comments on the current and voltages that I have observed on the wire
EDM - The operating current averages in the 1amp to 5 amp range for 0.008"
brass wire and the voltage averages out to about 300 to 330 volts DC - A
scope on the wire shows about 100Khz operating frequency - The instantaneous
current may be quite a bit higher but not having the equipment I have not
attempted to measure it -

I'm sure that newer machines use higher average currents in order to achieve
the reported cut rates of 15 to 20 square inches per hour but this machine
would only get between 1 and 3 square inches per hour - Attempts at higher
currents would always result in wire breaks which were a real pain -

In some discussions with EDM service engineers I have been led to believe
that it is not the current or voltage that is the primary consideration in
keeping the wire from "exploding" and also in making a uniform predictable
cut - It appears to be more the ability to control the total energy in each
discharge spark which is obviously a function of both voltage and current
together - The higher voltages appear to allow more sparks per second which
results in more uniform and reliable cutting as well as higher speeds, while
close control of the total discharge energy per spark is apparently
responsible for preventing wire breakages -

Dielectric is deionized water - I know of no wire EDM machine that uses oil
or kerosene, although I am certainly not familiar with all of them -

The power supply on the wire EDM machine appears to have a dozen individual
power supply boards, each of which apparently charges a capacitor between
pulses - The boards appear to be sequentially connected to the wire using a
ring counter or similar circuit that allows the discharge frequency to be
varied more independently than using the charge/discharge time constant of
the capacitor circuitry -

Current, frequency, maximum CNC feed rate, and the on/off ratio are all
independently adjustable as of course, is the linear speed of the wire -

Resolution of the system is (was) 0.001mm -

It would be a fairly simple project to add an EMC or other CNC control to
the system but the problems appear to be A) - Control of the spark discharge
as discussed above - B) Varying the feedrate depending upon spark conditions
- C) Detecting a short circuit quickly enough to prevent exploding the wire
- D) Reversing along the programmed path to clear a shorted condition -
These problems are somewhat less critical on RAM type machines but they
still exist -

I'd be glad to provide the CNC control and the machines if the electronics
gurus out there would want to work on the power supply and interface between
power supply and controller - Once running, a totally homebuilt machine
would be the next step -

Roland Friestad

P.S. - Finally got my web page going - www.cardinaleng.com -
Would appreciate comments and suggestions off list -

*************************************

>Message: 3 - Digest 1521
> Date: Mon, 30 Jul 2001 05:34:52 +0000
> From: Jon Elson <elson@...>
>Subject: Re: Another EDM Approach
>
>vrsculptor@... wrote:
>
>> Just musing, I'm out of my depth here....
>>
>> I'm also interested in EDM. I had concluded, before the recent posts,
>> that useable home EDM was not feasible. Seeing the recent posts I'm
>> not so sure. How about simplifying the proposed circuitry by
>> offloading most of the control to a PC.
>
>HUH? The pulse on time that someone suggested was 1 - 9 uS.
>There's no way you can control pulses that short with the PC parallel
>port.
>
>> The parallel port already
>> controls my Z axis using a step and direction pin. Could the third
>> pin be tied to a one shot circuit to fire the pulse?
>
>The pulser circuit doesn't "fire" the pulse, the EDM action in the small
>space between the electrode and the workpiece does the 'firing", and
>the pulse control interrupts the discharge, and prevents another one
>from starting for a controlled time. This kind of circuit can be built
>with a dual timer chip, the LM556, and 2 resistors and 2 capacitors.
>
>> A trim pot with
>> knob on the one shot circuit could manually control the pulse length.
>> Determining optimal operation (no spark, just right, short) seems
>> like a much harder task. It could be done if the EDM interface board
>> could post the operational status using some of the remaining
>> parallel port pins. The software to drive such an arrangement would
>> be trivial.
>
>But, WAAY too slow. Letting the PC control electrode advance (or
>other axis movements) and figuring out when to stop is fine, but the
>pulse control is just a bit too 'real time' for the parallel port.
>
>Jon
>
>
>
>________________________________________________________________________
>________________________________________________________________________
>
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>