PWM problems

Discussion in 'The Projects Forum' started by franzschluter, Jun 1, 2009.

  1. franzschluter

    Thread Starter Active Member

    Jun 1, 2009
    Hello this is my first post...
    And I am an amateur in electronics...
    I do have some understanding with transistors and ICs..
    I would really appreciate if you could point out some mistakes (If there are any) in my circuit and guide me to the correct path.
    I wish to learn and improve my skills in electronics.
    And finally I am aware of the voltages and currents I am talking about.. It can kill me if I go wrong and touch it. This is why I am building it module per module and testing each out with small voltages first before applying the bigger ones.

    Anyway.. I am building the same circuit principle as yours for over 2 months as of today.. Though I have tried using a 74LS123 but unsuccessful so far. I have included a partial schematic for making the PWM...

    I used LT1057 as pre-amplifier and then AD797 as final stage... Then perhaps optocouple it (If I can find a fast one that can go 100Khz) to a simple mosfet driver circuit using IR2112. If I can't find a fast optocoupler, I'll try conecting directly to the driver.. If it burns I have many spares and the fire extinguisher ready :)...

    Experience during my test was... I got squarewave output but the DC and Frequency would not vary.. I am using a very very old ancient oscilloscope during the 80's and I am doubting the output a lot... In any case I will be purchasing a 2nd hand oscilloscope in the next few weeks to recheck my output.

    If it still doesn't work I'll test it out first this time on the breadboard... (Tired of etching 6 boards of the same kind so far..)

    The schematics do have some mistakes such as the BJT's must be connected to the highside and not lowside.. Since it's NPN and it doesn't seem to want to switch correctly on the low side..

    There are a few Pots for varying Freq,Pulse Width and Amplitude. I wanted to absolutely vary amplitude in this case because some power mosfet switch at 10V and some at 15V... And since I am poor in computing expected voltages I have included a POT for amplitude to first hand test the gate driver voltage before connecting to the power mosfet.

    Then finally I plan to drive a powerful mosfet such as this one..
    IXFN 150N15

    My intentions of my pulse and parameters is..
    Controlling Frequency and Duty Cycle..
    I hope to go up as much as 100kHz.. and down as low as 500Hz... @ 100V and maximum current of 80 Amps (Depending on my shunt watt resistor)

    My transformer is a 3-phase 380V-75V 100Amp using a 3phase rectifier system.. 700V diodes to protect my circuit from over voltage.. There are also 700V diodes on my electrode and workpiece in series so that reverse voltages will be blocked or as a safety precaution for my expensive mosfet that the drain never sees funny phenomenons such as ringing effects....

    My servo drive circuit I have not included because it already works for my previous RC-generator prototype. This time I'll be applying a pulse instead of charging and discharging capacitors...
  2. beenthere

    Retired Moderator

    Apr 20, 2004
    Hi franzschluter, and welcome.

    It is less confusing if you start a thread with your question rather than placing it into an existing thread.
  3. franzschluter

    Thread Starter Active Member

    Jun 1, 2009
    Ok thanks.

    My question to this circuit..

    Can a BJT be used to switch a capacitor? I have used a 74LS123 and would like a very wide range of frequencies.. Usually it is very dependent on the values of Rt and Ct. In order for me to accomplish a wider range of frequencies I must use multiple Ct's. So can this be done using BJTs? Or should this be a mechanical switch?

    I wanted to try out a BJT switch since it would be less copper track and no wires.. But now I need an experts opinion.. Perhaps the Collector and Emitter is also a capacitive load? And this would affect my capacitor on Ct?
  4. hobbyist

    AAC Fanatic!

    Aug 10, 2008
    Jist in my own opinion instread of BJT. I would use a mosfet, only because a BJT, is a current operated device to the input, just seems like there is not enough isolation to act as a switch for capacitors to be switched in. while a mosfet enhancement type, would need a voltage input..

    I did an exzperiment one time I wantred to switch in the emitter bypass capacitor to an CE amp, to have the gain of the stage happen with a electronic signal. The BJT worked somewhat, but I had better results with a mosfet.

    Again just my opinion.
  5. franzschluter

    Thread Starter Active Member

    Jun 1, 2009
    If I use MOSFET there is no problem but just a few modification. If I may ask more detail.. Is there a certain capacitance between Gate and Source of MOSFET enhancement type when it is switched on or saturated? Did you use N-channel or P-channel? I would like to try that kind of setup .
  6. hobbyist

    AAC Fanatic!

    Aug 10, 2008
    They are N-channel enhancement
    2N7000 series and the IRF 610,

    As far as capacitance I'm not sure what the value is, you need to download the data sheets off the innernet, for that info.

    May I suggest to breadboard a small sample circuit using this to see how well it works for your application before comitting to a circuit board.

    For my application it worked, but your application may need a different setup altogether.