Can you upload the actual .asc file instead of just an image of the schematic? That will save everyone the trouble of redrawing the schematic themselves.
The connection at the bottom of R12, the 220 ohm resistor, connecting the drivers to the feedback node might not supposed to be there? Edit: at second glance i see it's symmetric, never mind. Haven't worked with a szlaski pair like this before, or whatever it's called.
I suggest you use a (slightly) slower op-amp. Also keep in mind the op-amp you selected is not suitable for your supply voltage.
If the circuit behavior is that dependent on the op amp used then the circuit is probably marginally stable at best.
When a circuit has stability issues like this it's not a good idea to just try things to see if they make the problem go away. Since this circuit has global feedback you really need to analyze the loop gain response. LTspice has a Tian probe example that shows how to do this. Look under the "examples\Educational" folder for a file named "LoopGain2.asc" and insert the Tian probe and its associated directives into your schematic as shown. Then run a simulation. What you are looking for is the phase shift at the frequency where the loop gain falls to unity (0 dB). A phase shift of 180 degrees (or more) turns negative feedback into positive feedback and the amplifier becomes unstable.
Again, post your .asc file and maybe someone here can offer some help.
When a circuit has stability issues like this it's not a good idea to just try things to see if they make the problem go away. Since this circuit has global feedback you really need to analyze the loop gain response. LTspice has a Tian probe example that shows how to do this. Look under the "examples\Educational" folder for a file named "LoopGain2.asc" and insert the Tian probe and its associated directives into your schematic as shown. Then run a simulation. What you are looking for is the phase shift at the frequency where the loop gain falls to unity (0 dB). A phase shift of 180 degrees (or more) turns negative feedback into positive feedback and the amplifier becomes unstable.
Again, post your .asc file and maybe someone here can offer some help.
The 100 pF capacitor C4 that you added to the original circuit probably provides enough lead compensation to stabilize the amplifier. Did you try this modified circuit with the original op amp?
Last edited:
Would the universal opamp be an ideal opamp by any chance? If so, the reason it doesn't oscillate is that the universal opamp has infinite bandwidth.
I suggest that you introduce yourself to the concepts of phase margin, gain margin, and stability analysis. Sergio Franco, "Design with Operational Amplifiers and Analog Integrated Circuits" is a good place to start.
Tom
Usually, R1 is 10 Ohm.
C4 is too large.
Combining with 22K, C4 kicks in about 70khz. You loose performance. I would say remove it.