chopper circuit -- it SHOULD WORK!

Discussion in 'The Projects Forum' started by amsim, Mar 15, 2011.

  1. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    I've built this and torn it down twice to test the components. What am I missing?!

    On the right side, the leads are

    V+
    Data In
    Data Ground
    Ground

    Are the diodes backwards? Is the snubber causing problems? (They are both NPN transistors, with the power transistor having the standard pinout BCE. Ignore the "BAT81" labels.)
     
    Last edited: Mar 15, 2011
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Where is the schematic? :confused: Parts placement won't help much unless the exact parts are described.
     
  3. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    Sorry. Here ya go...
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Not sure exactly what you're trying to accomplish.

    If JP2 connections are:
    1. V+
    2. Data In
    3. Data Ground
    4. Ground
    then your NPN transistors are upside-down in the schematic (emitters to +V).
    Not sure why you used a 2N6099.
    The 2N6099 would have a very slow turn-off time due to no base return resistor (if it were connected properly).
    No current limiting on the 2N/PN2222 collector or 2N6099 means the 2222 would likely be getting toasty/toasted.

    Where did you get this circuit?

    [eta]
    A bit about schematic conventions...
    1) More positive voltages towards the top, more negative towards the bottom.
    2) Inputs come from the left, outputs flow to the right.
    3) Document voltages used.
    4) Describe logic level signals.
    5) Try to keep wires either vertical or horizontal; avoid diagonals when possible.

    Cadsoft Eagle specific:
    If more than one wire is connected to a terminal, or one wire ends in the middle of another wire, a junction is required.
    Use the Erc function early, and often.
    Be careful that not all transistors have the same pin outs, even though the packages may be similar or the same.
     
    Last edited: Mar 15, 2011
  5. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    I'm being an idiot, sorry.

    Perhaps the redrawn schematic is a bit clearer?

    The power supply is coming in from the left, V++ on top. On the right is what goes out to power the rest of the circuitry.

    2n6099 is what I had on hand. Somehow, my TIP102's, stored in an anti-static bag, got fried. Would a 200 ohm resistor be about right between the 2n2222 and 2n6099? What about a 10k base return resistor on the 2n6099?
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, you're using a more-or-less Darlington configuration as a voltage follower, so your output on JP2 pin 4 will never be higher than your logic input signal less ~1.2v; lower if the load is significant.

    So, if your logic signal goes to 5v, your output will be <3.8v. Doesn't really matter what V++ is on JP3 pin 2 (except for power dissipation in the 2N6099, perhaps).

    Why don't you describe what your inputs are (like the voltage on JP3) and what your logic signal will be (levels, frequency, etc) and what the output on JP2 pin 1 should be?

    Without that kind of info, we're taking potshots in the dark.
     
  7. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    I am trying to use around 35kHz PWM on a 24V supply. The logic supply is 5V at no more than 40mA. The output will probably be about 3-6% duty cycle (an effective current of about 200mA?).

    This is a small part of a multi-discipline endeavor, and I am, by no means, an expert. This is a circuit I more-or-less hobbled together. I would be open to a better circuit if a link could be provided.
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, that means you'd have to switch the transistor(s) on and off mighty quickly. I'm afraid that a Darlington configuration will be pretty slow, even WITH base return resistors.

    Your load - what is it? Purely resistive?
    Just as a rough ballpark, 200mA/3%=66.6A, 200mA/6% = 33.3A. That would fry many transistors. If the load is inductive (like a motor, solenoid, etc) then the transistor would get zapped unless there was a freewheel diode or snubber circuit; so I'm suspecting your load is inductive as you've mentioned snubber.

    I see. Well, without getting all of the requirements defined, it's going to be difficult-to-impossible to determine what might meet your needs.

    Getting little bits and pieces of information at a time could make this thread go on forever, and I don't have that much time.
     
  9. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    I suspect that I'm going to have to go in a different direction with this project. Have you ever had the feeling that you have gone off the path and into the mud? I've felt like that for about a week, now.

    Thank you, I genuinely appeciate your time and input, SgtWookie!
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I'm not trying to be hard on you. I simply am not certain of what you are actually trying to accomplish.

    You mention a 24v power supply, 35kH PWM at 3% to 6% ON, and I only suspect that you have an inductive load, but not the nature of the load. Do you have any specifications for it?

    Old-style chopper drivers (like the L297/L298 pair) used Darlington followers on the high side due to the relatively fast turn-on/turn-off times, but they dissipated one heck of a lot of power. About the best you could get out of them was 2A if they were REALLY well heat-sinked; you'd get about a 4V-5v drop across each half-bridge.

    Using Darlingtons as saturated switches really isn't much better, as Vce(sat) is quite high, and response time is slow.

    More modern type bridges use MOSFETs and a high-side gate driver. Synchronous buck configurations are pretty slick; they minimize power dissipation in the driver circuit. A power MOSFET is used across the load as an "ideal diode" to keep the current flowing.
     
  11. amsim

    Thread Starter Member

    May 24, 2009
    14
    0
    Hey, I just noticed that you are from Orlando, FL! I grew up there! I lived off of Goldenrod Road, between HWY 50 and University Blvd. It looks a whole lot different on Google satellite view these days, though. My intention was to drive a stepper motor with an H-bridge, but I had already figured out quadrature encoding, and a DC motor is a lot easier to control anyways. Maybe in another project I'll give steppers another go.
     
Loading...