1
\$\begingroup\$

I've designed this Voltage Controlled Amplitude circuit and I've simulated it in Multisim.

When simulating I get a waveform like this:

Multisim screenshot

The green waveform is the modulator signal and the red is the output of the circuit. This seems to work OK.

But building it on a breadboard, I get this on my oscilloscope:

Oscilloscope screenshot

The green signal is a square wave that is the input signal. This is modulated by a low frequency sine wave. The yellow signal is the output signal, that is obviously not the expected square waveform.

I can't find the error on the breadboard. Any advice on how to fix this?

\$\endgroup\$
1
\$\begingroup\$

I can't find the error on the breadboard... Any advice on how to fix this?

Put your scope directly on the output of the LM13700 and confirm that the output is sat at a DC level close to 0 volts (half way between -12 volts and +12 volts). See what the signal looks like at this point. If it's OK then debug why it isn't coming out of U3 correctly.

If it's not OK then replace (temporarily) U2's output with a 10 k pot connected across supplies and feed the wiper to the 22k resistor. If the LM13700 output looks OK then suspect how you have implemented U2's circuit on breadboard.

If it still isn't ok from the LM13700 then suspect problems around that part. If all else fails then use this circuit: -

enter image description here

And check that you have all the pins connected to where they should be. I can't see if you have correctly connected pin 5 to pin 7 for instance. Neither can I see where you have connected pin 2.

It's called fault finding and it's all about reducing a big job to a smaller job.

\$\endgroup\$
  • \$\begingroup\$ I put the scope at the output of the LM13700 but the problem was still there. The problem is not at U3A I'm sure. I tried to connect a +-12V sine wave to the 22k resistor and the output is the same, so U3 is not the problem. I will try to take a closer look to it as soon as possible. Thank you \$\endgroup\$ – JulianC Jan 19 '17 at 20:53
1
\$\begingroup\$

I found the problem! Negative supply leg in the LM13700 had bad contact with the -12V supply. I'm sorry this was a very noob errorenter image description here Now it's working nicely! Thank you for your help.

\$\endgroup\$

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.