No announcement yet.

EDM osilator

  • Filter
  • Time
  • Show
Clear All
new posts

  • EDM osilator

    I want to build the "osilator circuit", that Robert Langlois speeks of in his book. I have some ideas but want to bounce it off you guys, and see what ya think.

    My plan is to use a big out put capacitor. Then make an adjustable pulse generator to turn on and off a P chanel FET. The FET will feed the electrode.

    My thoughts; The capacitance, as long as it is large enough should not make a difference. The pulse generator will controle the number of sparks (pulse rate)and the duration of the spark (pulse width). For A fine slow finish, you would use lots of sparks with a smaller pulse width. For a fast cut and rougher finish, you would use less sparks but longer pulse width. If you don't have enough capacitance, then you will run into trouble if the pulse width is too long. Also your off time will have to be long enough to charge the caps.

    The pulse generator; I plan on using two 558 timers. I will share the simple circuit if needed, but don't know how to add a schematic in here. Any help with adding in a schematic, to my post, is also appreciated.

    All safty mesures will be followed, isolation ect..

    Well, tell me what you think.


  • #2
    Try Jon Elson's page, good help for start.


    • #3
      Consider a power mosfet circuit for this. They can handle huge instantaneous spikes in current safely (with heat sinking). If you need a circuit designed for you ask a local tech school or university engineering department if they would do it for you. You will have to tell them the maximum input voltage, current, fequency range, and polarity reversing ability if desired. The more you ad the more complex it gets. If you have some electronics background you can get sample circuits out of Motorola or IRC Power MOSFET Data books.

      MOSFET's are opposite to bipolar devices in that as they heat up they naturally throttle back to prevent the thermal runaway tendencies that bipolar devices experience when pushed.


      • #4
        I have never operated an edm machine, but I do know about capacitive discharge. Anytime a good capacitor is discharged into a shorted circuit, which is essentially what happens with an edm, the current pulse is very large, possibly hundreds of amps, and thousands of amps in some cases, depending on the size of the leads, and the solidity of the connections. I don't think that putting the fet in the discharge path is all that good, although in theory, you gain the control over pulse rate and duration. Why not go with a switching type of power supply, which, in electronic ignition fashion, will power up the discharge capacitor without requiring the network of resistors (read that as waste heat and power) currently seen in edm's. This type of power supply shuts down when overloaded, then recharges it's capacitor when the short is removed, precisely what happens when a spark is triggered, in the case of an electronic ignition, or when an arc jumps the gap in edm operation. Any input on this, anybody?
        I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-


        • #5
          More fun, three people three ideas.....

          This is made to order for an SCR, with the disadvantage that once on, you can't turn it off. There are "GTO" (gate turn off) devices that do allow that, but they have other problems.

          The suggestion would then be to reverse the idea, letting the capacitors charge up to a settable voltage before triggering the SCR.
          By changing the trigger voltage, you change the energy per spark, and the erosion per spark.

          Since there is a limited range available, you can switch in and out capacitors to range switch.

          SCRs are tough devices, and are available in various current ranges. The largest types are slower, however, so enough and not too much is best.

          I tend to agree that any devices in the way of the spark are trouble and potential failure points in most cases. One would want to avoid them if possible.

          An idea that may be already described somewhere is to use a high voltage short pulse/low energy spark as an initiator, with the main energy coming from the large low voltage bank as what the power company calls a "follow-on arc". Using a small inductor to isolate a high frequency initiator might be possible. This avoids a device in series with the main spark, just a small air-core inductor.


          • #6
            Oscillator circuit?

            Ever consider a powerful stereo amplifier driven by a signal generator? Both used and maybe cheap?

            A good stereo amplifier will put out a couple hundred watts at 20 KHz. A square wave signal, maybe an auxilliary blower whose blast is directed over the heat sinks, who knows; maybe an industry leader is born.


            • #7
              What frequency and current are involved? Are you talking pulsed DC or AC?

              Phase controlled triacs are used in stage lighting and some of them can handle 50 amps. I have reverse engineered and hot-rodded some of these dimmer packs and might be able to help.


              • #8
                Hmm, stereo amp as an edm power supply. Very few are capable of handling the high peak output currents involved without breakdown. Some amps have an enormous output rating, current-wise, and have been advertised as being able to weld from. These cost in the multi-kilobuck range and won't be found at steal prices at flea markets or surplus houses. Maybe from a little old lady whose audiophile husband had died and she's just getting rid of his 'junk'. I like the idea of using a high frequency, high voltage, imposed on the output from an edm machine to initiate the discharge from the capacitors. There is a lot of potential here. (pun not intended). I once made such a circuit for a welder to use with an existing machine, and he said it made it easier to start and control the weld.
                I seldom do anything within the scope of logical reason and calculated cost/benefit, etc- I'm following my passion-


                • #9
                  please send to my e-mail scematic Voltage controlled osilator wide-range 20Hz to 20kHz
                  i needed for eksperiment