Announcement

Collapse
No announcement yet.

RBC .... or,

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • RBC .... or,

    Really Big Coils.

    I know that several folks out there have coils that contain many strands and are pretty hefty.

    This thread is about such coils, hoping that we can all gain knowledge about these things.

    My coil began life as a standard SSG coil, 117 strands of 23AWG, about 315 feet long. The wire size was used because I came across a large spool pretty cheap.

    The number of strands and length came from trying to impedance match the coil to a battery bank that I had at the time.

    The core was soft iron, and the coil had several strands of #26AWG wire in there for tgger wires.

    The wheel was made from a 5 foot diameter table top and had 16 6" x 2" x 1" ceramic magnets mounted. Shaft diameter was 1".

    The thing never worked very well, and, I didn't have good batteries either.

    So, I've ditched the wheel and made an air core out of the thing.

    I'm using 7 MJL21194 transistors per heat sink, each with 4 strands of wire attached.

    Firing comes from the PWM which allows both frequency and duty cycle adjustment. (The PWM is from Bob French and posted over on EF.)

    Several frequencies produce good radiant, with something like 1.8 KHz seeming to give the best.

    The machine is connected 'split positive' between a 24 VDC supply and 6 each 6 VDC batteries, series connected for 12 VDC.

    On my first try at this, I only had one 12 VDC bank at the bottom of the circuit. It quickly rose to 15.5 VDC, so I shut it down and am using 1 battery for charge from the SG and 3 banks of 12 on the bottom.

    This is working well, as both the charge battery and the bottom bank are above 14 VDC.

    Current from the 24 VDC supply is 3.2 amperes.

    What I want to do with this next is:
    1. Use JB's SG3524 PWM from his patent to control the base biasing transistor. There's a print around somewhere that shows how to vary the duty cycle. I've played around a bit but can't accomplish that.
    I can get the frequencies that I want, but not the duty cycle.

    2. I've had a picaxe setup around for some years. Reckon I'll start studying up and find out how to use that.
    I want to be able to choose about 5 somewhat tweakable output frequencies to send to the coil and I would like the frequencies to be 'rock solid'.

    I can live with the thing the way it is now, but would like to have a cleaner wave form.

    Feel free to chime in there folks, I'm sure there are others out there.

    glenWV
    Attached Files

  • #2
    Originally posted by GlenWV View Post
    Really Big Coils.

    I know that several folks out there have coils that contain many strands and are pretty hefty.

    This thread is about such coils, hoping that we can all gain knowledge about these things.

    My coil began life as a standard SSG coil, 117 strands of 23AWG, about 315 feet long. The wire size was used because I came across a large spool pretty cheap.

    The number of strands and length came from trying to impedance match the coil to a battery bank that I had at the time.

    The core was soft iron, and the coil had several strands of #26AWG wire in there for tgger wires.

    The wheel was made from a 5 foot diameter table top and had 16 6" x 2" x 1" ceramic magnets mounted. Shaft diameter was 1".

    The thing never worked very well, and, I didn't have good batteries either.

    So, I've ditched the wheel and made an air core out of the thing.

    I'm using 7 MJL21194 transistors per heat sink, each with 4 strands of wire attached.

    Firing comes from the PWM which allows both frequency and duty cycle adjustment. (The PWM is from Bob French and posted over on EF.)

    Several frequencies produce good radiant, with something like 1.8 KHz seeming to give the best.

    The machine is connected 'split positive' between a 24 VDC supply and 6 each 6 VDC batteries, series connected for 12 VDC.

    On my first try at this, I only had one 12 VDC bank at the bottom of the circuit. It quickly rose to 15.5 VDC, so I shut it down and am using 1 battery for charge from the SG and 3 banks of 12 on the bottom.

    This is working well, as both the charge battery and the bottom bank are above 14 VDC.

    Current from the 24 VDC supply is 3.2 amperes.

    What I want to do with this next is:
    1. Use JB's SG3524 PWM from his patent to control the base biasing transistor. There's a print around somewhere that shows how to vary the duty cycle. I've played around a bit but can't accomplish that.
    I can get the frequencies that I want, but not the duty cycle.

    2. I've had a picaxe setup around for some years. Reckon I'll start studying up and find out how to use that.
    I want to be able to choose about 5 somewhat tweakable output frequencies to send to the coil and I would like the frequencies to be 'rock solid'.

    I can live with the thing the way it is now, but would like to have a cleaner wave form.

    Feel free to chime in there folks, I'm sure there are others out there.

    glenWV
    Hi ,
    SG 3534 pwm Chip as per his patent is applicable not essentially for the ssg build that you have shown...its a cap dump similar to his comparator circuit..
    you may refer the Advance SG Book for this....
    Rgds,
    Faraday88.
    'Wisdom comes from living out of the knowledge.'

    Comment


    • #3
      Originally posted by Faraday88 View Post
      Hi ,
      SG 3534 pwm Chip as per his patent is applicable not essentially for the ssg build that you have shown...its a cap dump similar to his comparator circuit..
      you may refer the Advance SG Book for this....
      Rgds,
      Faraday88.
      Greetings:

      Yup, you are correct.

      I've built several variations on this cap pulser and have set them up to vary the frequency, but not the duty cycle.

      I found the print showing how to vary the duty cycle and put it together on a bread board and tested it. It worked, somewhat.

      Then, I moved the components to a prefab PCB and soldered it up. Today, we'll see if it works.

      Like the patent, optos will be used to bias a transistor or FET used to pulse the power transistors on the heat sinks.

      There are four heat sinks with 7 transistors on each.

      First, we'll try pulsing two transistor transistor banks from each opto. Reducing the duty cycle should allow the coil enough off time.

      If that doesn't play well, all four banks will be fired from one opto.

      If that doesn't play well, its back to the LM339 PWM circuit and the SG3524 will be the brain for another cap pulser.

      Just another bunch of experiments trying to find the right fit for this machine's personality.

      Thanks for responding, I appreciate it.

      glenWV

      Comment

      Working...
      X