I want to produce a good quality PWM signal with transistors(BJTs, FETS, etc..). I do not want to use an IC at all(learning to use transistors properly). I'm currently using a BJT Astable Mutivibrator enter image description here

@Source pin is a mistake I meant drain

As you can see the signal is fine until I attach the mosfet.

I've tried the configuration using 1N4148s to improve rise time, however due to the current limitations of the switching diode the signal isn't strong enough, I've tried various amplification techniques I found online, but the results were so horrible they weren't worth posting. The DC-DC converter works like this, however it is not the subject of the question.

How can I stabilize this signal for practical general purpose use or is this an exercise in futility?

EDIT To Newbies, DO NOT BUILD THIS! If you want to build a boost converter from jellybean parts, use an NE555 and an OP-AMP or (better) comparator to regulate the voltage via the CV pin, even then there's a nasty ripple voltage. However the answer to the question is correct emitter follower(pushpull amp), etc was the best way to go. Oh and please don't use an autotransformer, it does lower the stress on your MOSFET, however the voltage isn't controllable without a switching regulator on the output end.(Which should probably be implemented in any case. )

  • \$\begingroup\$ What frequency is it oscillating at? \$\endgroup\$ – Andy aka Jul 21 '16 at 12:07
  • \$\begingroup\$ 105KHz, the only difference between the readings is that I pulled the gate pin from the breadboard in the second one. \$\endgroup\$ – iuppiter Jul 21 '16 at 12:28
  • \$\begingroup\$ Which nodes in your schematic are"Source Pin" and "Output"? I suspect you mean the drain and gate of the MOSFET, respectively. \$\endgroup\$ – Dave Tweed Jul 21 '16 at 13:04
  • \$\begingroup\$ What happens to the voltage on the supply line when the MOSFET turns on? \$\endgroup\$ – JimmyB Jul 21 '16 at 13:09
  • \$\begingroup\$ @Dave Tweed: You are correct the output was from the drain pin, that was a mistake. \$\endgroup\$ – iuppiter Jul 21 '16 at 13:21

Add a buffer using complementary emitter-followers to drive the MOSFET gate as follows:


simulate this circuit – Schematic created using CircuitLab

This will greatly reduce the loading the MOSFET gate is causing on the collector output. At your ~100kHz frequency the MOSFET gate will be almost like a short.

|improve this answer|||||
  • \$\begingroup\$ For some reason it knocks the oscillator out in the schematic above, but the lower power oscillator with the 1N4148 diodes does work with it, still getting a low rise output, but much better than the original. Frequency is stable. Should I be expecting a perfect square wave or is this supposed to happen? \$\endgroup\$ – iuppiter Jul 21 '16 at 13:29
  • \$\begingroup\$ You should never expect a perfect square wave since it's impossible, but this kind of multivibrator loads the output with the drive for the opposite side, so it perhaps won't look as pretty as you might like. It can be cleaned up with a buffer stage that has gain and perhaps hysteresis (unlike the one I showed above- based on your 'signal is fine before...' comment). \$\endgroup\$ – Spehro Pefhany Jul 21 '16 at 13:49
  • \$\begingroup\$ Wouldn't that require an Op-Amp? \$\endgroup\$ – iuppiter Jul 21 '16 at 21:33
  • \$\begingroup\$ No, a Schmitt trigger can be made with a couple of transistors, maybe one more if you need it to swing very close to both rails. \$\endgroup\$ – Spehro Pefhany Jul 21 '16 at 22:19
  • 1
    \$\begingroup\$ Thank very much you for that very simple yet effective answer, I've found that it is quite common in driving FETs. I'm a student and you know the more you learn, the more you realize you don't know. At the time I didn't even have a spice simulator. \$\endgroup\$ – iuppiter Sep 25 '16 at 12:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.