design of a pulse width modulation page 2 cont.

Now replacing the two 1.8K ohm resistors, for a single pot. of 5K ohms, and using a current limiting resistor at each base of 100 ohms,
I now have a manual control signal into the base of these PNP transistors to work the circuit.

Built the cuircuit and I get a almost a full range from 0% - 90% duty cycle.

However the frequency drops slightly as it goes from the 50% duty towards both ends,
starts out lower freq. as it approaches 50% duty, the frequency increases, then past 50% duty, the freq. decreases again, back to original.

However at the desired freq. of operation, this has little effect on the output, so this is working very good for this configuration.


Now it's time to work this for a single input manual control, the way it worked before, was the pot was used to simultaneously change the resistance at both transistors at the same time, to give a PWM signal.

But now that a long tail pair is incorporated (diff. amp config.), it is for the sole purpose of converting a double input into a single input, so that a feedback signal can be fed into it to raise the output as needed when the motor is stalling.

The way this was done was by, experimenting by trimming the resistor value at the base of the diff amp, (non inverted output), which is connected to the side of the astable that runs the output, by trimming this resistor value to a point where the LED (astable output is shut off, but on the threshold to turn on, this turned out to be around 2.2K ohms.

After that was established, I then trimmed a resistor value from the positive supply to the base of the (inverting input of the diff. amp), until the astable began to oscilate, then I continued experimenting to find the right value of resistance that would turn the output off, and turn the output to almost 100% duty cycle.

Once that was found it was a matter of incorporating it into the circuit build to come up with the almost finished PWM unit, all is needed now is some feedback circuitry, to be designed and built into this.

The only quirk with this design is there is a great frequency change when the output is driven harder, it drops from 4Khz to 400 hZ.
However the load being drivien is able to handle this so no problems with driving a load.


Here is a video of it working with all the circuitry added.

Next I'll work on the feedback circuitry.

Blog entry information

Last update

More entries in General

More entries from hobbyist

Share this entry