LM339 PWM help

Discussion in 'The Projects Forum' started by JasterMake, Jul 9, 2011.

  1. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    I purchased the following PWM and it serves it's original purpose well but I need to change it.

    http://www.kit4diy.com/images/pdf/MX067.pdf <--Best I could do from here



    I can not drive an ignition coil with this PWM. I assume it is because the Hexxfets never shut off completely and I would think that the problem is with either R9, R10 or R11. but I am a self-taught and don't really know what to do here.

    Do you think it would be possible to modify this to drive an ignition coil?
     
    Last edited: Jul 9, 2011
  2. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,346
    Hello,

    There is NO schematic shown.
    What is the intention of the use of the ignition coil?

    Bertus
     
  3. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    I am working on getting the picture on the question. I need to find it on the net again or find a place to host it. I don't post very often.

    but the ignition coil is hopefully going to be my 5-10kV power supply for high voltage learning. My first set up is going to be a Tesla Induction coil. Then attempting to recreate the 5 discharges described by him. Then follow his path through 1892 impedance phenomena. If I could find a 10kV DC dynamo I would prefer to use that.

    But I need a high voltage power supply. And I hope a ignition coil will work.

    I have it self triggering for now but I would like some control over it.
    :eek:)
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You attempted to post the image by linking to where it is on your hard drive; you are the only person that can see it.

    Edit your 1st post, then click the "Go Advanced" button, on the next page that comes up scroll near the bottom and click the "Manage Attachments" button.
    It's pretty much self-explanatory from there.
    Your image was in "/Users/E443255/AppData/Local/Temp/msohtmlclip1/01/clip_image002.gif"
     
  5. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
  6. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    The kit you linked to is a motor driver. It is probably destroyed if you used it to drive a ignition coil. When the ignition coil pulses there is no longer just the 12V (or what ever voltage your using) on the primary winding. Because of back EMF there is some where around 300-400V on the primary.

    Here is a link to a couple of circuits that work and are made for driving a coil;
    http://www.rmcybernetics.com/projects/DIY_Devices/homemade_ignition_coil_driver.htm

    http://www.rmcybernetics.com/projects/DIY_Devices/homemade_power_pulse_generator.htm

    As far as a Tesla coil, they are AC not DC. Although some are DC powered the actual coil is AC. The DC is inverted to power it.
     
  7. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    The controller still works great. I added a snubber to it before I started trying to drive a coil.

    The problem is I get no {inductive collapse / back emf} or even a spark from the coil when I use this driver. If I use a 3055 and relay it works very well.
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The trouble is due to D1 and the intrinsic body diodes in the MOSFETs. They provide a current path for the coils' reverse-EMF, so that the primary current decays very slowly, so the field collapses very slowly. This works great for PWMing a motor or a lamp, but lousy for an ignition coil where you WANT the current to stop/reverse/oscillate quickly, and the field to rapidly collapse, so that energy in the primary is transferred to the secondary.

    Trying to mod that PWM circuit wouldn't be a good idea, because you'd have to replace the MOSFETs with high voltage ones, add a power resistor, remove D1, and a couple of high-voltage caps - and if you made a mistake, you could burn up the whole thing.

    The really old points/condenser ignitions they made back in the 50's up to the 70's had an 8 Ohm current limiting resistor between the battery and the coil (+) terminal, and a condenser (capacitor, I think 0.22uF) from the (+) coil terminal to ground. The (-) terminal lead was connected to the points and condenser (another 0.22uF capacitor).

    When the points close, current flows through the 8 Ohm resistor through the coil and the points. As the current builds in the coil primary, the magnetic field expands. When the points finally open, you wind up with what is basically a charged inductor with capacitors connected to a fixed voltage reference (ground) on either end of it. The current keeps flowing which rapidly stores electrons in the cap across the points, and pulls them out of the cap on the (+) terminal of the coil primary. This continues for a bit, then the voltage potential across the coil in the reverse direction becomes so great that the voltage pressure causes current flow to start in the other direction; the magnetic field collapses and expands again but polarized oppositely, reversing the charges on the caps again, causing the flow of current to reverse again, etc. This occurs a number of times until the parasitic resistance in the circuit and the dissipating of energy in the secondary due to sparking across the gap causes the current flow to stop.

    I hope I didn't get too elementary or detailed here; just trying to bring it to a level where most people can understand what's going on.
     
  9. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    Well that answers it quite well.

    Thank You,

    I guess that means I need to build my own.

    Do you have any threads to point me to?
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, I used to have some schematics kicking around using relays as points, but the results from those were mixed.

    I think what you need is a circuit that energizes the coil at a ~67% duty cycle, and has a frequency that varies from how often the engine would trigger it at idle to maximum RPM.

    In the old days, a V8 engine would use 30° dwell (points closed) for a V8. Since there were 8 cylinders and 360° per rotation, 360°/8 = 45°; 30° dwell / 45° = 66.666...%, which is how I arrived at the 67% duty cycle. If the dwell was too high, the points would burn because they opened too slowly. If the dwell was too low, the points would "bounce" at high RPMs causing erratic ignition.

    I don't have a good circuit for you offhand.
     
  11. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    I have been trying to get it to self trigger but I have not been able to get the dwell time long enough to produce the 10kV I am after. Really bright 1cm sparks.

    I have been using a 2n2222(6 inch wire to base) to trigger a 3055 this will automatically start to oscillate in the tens of kHzs which is too fast for the coil. I cant think of a way to slow it down substantially. If I add resistance or capacitance to the 2222 base it will slow down a little but it also seems to lower the dwell as well which lowers the power. ????

    I guess this is what I need.
    http://www.rmcybernetics.com/shop/c...ators/power-pulse-modulator-ocxi/prod_40.html
    But I will not spend $100US on that.

    Lets learn to make our own.
     
  12. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    @SgtWookie, the rotation for a V8 engine is actually 720° They are four stroke engines, two turns of the crank for one cycle, and cylinders are 90° apart.
     
    Last edited: Jul 10, 2011
  13. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
    Last edited: Jul 11, 2011
  14. JasterMake

    Thread Starter New Member

    Feb 19, 2009
    7
    0
    Where's the link?

    Thanks,

    :D
     
  15. shortbus

    AAC Fanatic!

    Sep 30, 2009
    4,013
    1,531
Loading...