# Push-pull amplifier oscillation

I try to build simple unitary push-pull amplifier which amplifies also DC. Here is a result of my work:

Unfortunately high frequency oscillation effect occured due to strong negative feedback:

My idea to resolve the problem is to limit bandwidth of my circuit by using low pass filter. I found "simplified non-inverting amplifier filter circuit" on this site: http://www.electronics-tutorials.ws/filter/filter_5.html

Here is the schematic:

Is this good approach to resolve my problem? Let me know: 1. Where in literature I can find information about especially this filter circuit? 2. Will this filter circuit work if I embrace in negative feedback loop also push-pull stage? 3. How to calculate the filter (the cutoff frequency and ensure unitary gain in passband).

Why I could't find much information about non-inverting amplifier filter in the Internet?

• What is the purpose of the feedback? – Glenn W9IQ Jul 9 '17 at 10:41
• Have you added local DC supply-bypass-capacitors to your breadboard? Does load resistor R1 return its current to the DC supply ground, or to the op-amp's "-" terminal or to the function generator ground? Wiring conditions matter at mega-hertz. – glen_geek Jul 9 '17 at 12:02
• @GlennW9IQ To improve linearity. – Madras Jul 9 '17 at 15:57
• @glen_geek I haven't added any decoupling capacitor. Load is connected directly to DC supply ground. To DC supply ground is connected generator and scope ground. – Madras Jul 9 '17 at 16:02
• Your meter of wiring to the power-supply may be the problem. Add 2 large bypass capacitors (1,000uF at 25 volts) from -17v to GND of the load resistor, and from +17v to GND of the load resistor. – analogsystemsrf Jul 9 '17 at 22:38

The first thing I note is that you have not used decoupling capacitors on the power rails of your op-amp and what can happen is this; the power rail drops cyclically with the output signal because of the load current and this causes the TL082 power rail to wobble up and down. That wobble inevitably finds it way to add (constructively) to the signal that you want to amplify. This can cause severe ringing and spurious oscillations and that is what I believe is the cause.

The ability of an op-amp to avoid this is called PSRR (power supply rejection ratio).

Here's why I don't think it is the extra phase shift caused by the poor high frequency characteristic of the output transistors. Below is the TL082 open loop gain and phase response and I've superimposed red and blue lines to show what I believe happens whan the output transistors modify the loop gain: -

At about 200 kHz the phase (red) starts to be modified and this will tail off toward 180 degrees at a lower frequency than for the op-amp on its own. But the gain (blue) will also tail off and, as you should be able to see there should still be significant phase margin to avoid oscillation. Another clue is that oscillation appears to be bigger at the peaks of the waveform.

This points to PSRR problems. Solution - use decouplers directly on the power rails to the op-amp and, prior to those decouplers, insert series 10 ohm resistors in the feeds to the op-amp power rails to form a decent rejection filter at high frequencies.

The PSRR figure of about 80 dB in the data sheet (the call it $k_{SVR}$) hides the fact that this is probably only at 50 or 60 Hz, Modern op-amps will usually show a graph and, at upwards of 100 kHz the PSRR figure will be very poor for the TL082.

• Ok who is the idiot down voter. How rude! – Andy aka Jul 9 '17 at 15:09
• I do not know how has voted your post down (it wasn't me :)) ). I think you presented your theory very clearly and it makes sense what you talk about PSRR. But why you think push-pull darlington stage affects characteristics from about 200 kHz and not from for example 50 kHz ? So the conclusion is we need decoupling capacitors and also more modern opamp (which one would be good, I haven't know TL082 is not modern before) ? – Madras Jul 9 '17 at 16:44
• @Madras the TL082 is fine for this application. I looked up the data sheet for the Darlington and noted that either rise or fall time was quoted at 4 us so I estimated, in my head that it would start to produce noticeable attenuation about 200 kHz but it could be a bit lower than this. If it was lower, it would still be the same story and that story is adding the transistors is not, in my opinion, causing phase margin instabilities. – Andy aka Jul 9 '17 at 19:27
• @Andy aka .Chill I upvoted because I also think that the output follower stage will not be enough to stuff it up .Maybe when people downvote they should give username and reason .This would knock spurious downvotes on the head. – Autistic Oct 19 '17 at 2:55

See Nyquist's stability criterion: at the frequency where the loop gain reaches 0dB, the entire loop phase shift must be lower than 180°. And a substantial margin of 45° is very desirable, so in practice it should not exceed 135°.

Your problem is that the output stage is slower than the opamp, so it introduces too much phase shift into the feedback loop. Also this particular opamp has a wimpy output stage, unable to source much current or drive the difficult load that a class-AB output stage presents. Also, its slew rate is low, so expect problems at crossover.

A quick fix would be to:

• Add a cap between the output of the opamp and the "-" input to provide local feedback at HF
• Isolate the opamp from the output stage by putting one resistor on the output (after the cap) and another resistor on the "-" input.

Like in this schematic (check C3):

(note this output stage is different from yours, it is actually a simple current feedback amplifier with gain).

If you process DC signals, it will work. Now, for lower distortion on AC signals, I'd use a real audio amplifier output stage, with discrete transistors instead of darlingtons, and an opamp with a bit more output current ability than TL072.

If it fits the specs, it is much simpler to use an integrated power op amp like LM1875 or LM3886, these are foolproof, cheap, work very well, and you get output short circuit protection too!

JLH Edit

This one isn't bad for its age and simplicity, however...

You say it is "current feedback", well yeah, but the benefits of current feedback (ie speed) are only there when the output slews down! and Tr3 can conduct hard, pushing base current into Tr1. However if you attempt to slew in the other direction then Tr4 runs out of bias current and turns off. Tr3 turns off too.

And then you'll have to wait about... forever for the stored charge in Tr1's base to flow out through R7 which only pulls out a ludicrously low 300 µA or so.

Expect a generous amount of rail sticking on clipping, cross-conduction when coming out of clipping, and this can only run in class A since the amp is unable to turn off Tr1 fast enough to go through the crossover without humongous amounts of distortion.

Also, how is the bias set?

Guys let's be honest. It was good for its time, but now it is obsolete junk (unless you only do sawtooth with the fast edges going down...) ; LM3886 would beat this in pretty much any measurable way (and probably on sound too).

Run a simulation if you wanna have fun. I simmed a MOSFET variant a while ago, was quite fast but I had to use a 30mA pulldown on the gates... and it still had cross-conduction on clipping. Would've had to add drivers, then it's just as complex as another amp. Didn't even bother to build it. There are much better schematics.

• I would point to that schematic as a bad design for audio amp: 1) while local feedback is good, using C3 to do it is bad; 2) the topology for the ops is bad -> it is very difficult to stablize. pretty much the only mosfets you can use here are the laterals; 3) another problem the fets present is gate capacitance, especially when driven via collectors. this amp would be very difficult to stablize. – dannyf Jul 9 '17 at 12:24
• over the last 30 or so years, audio has moved from focusing on high (open loop) gain + deep negative feedback to focusing on good open loop performance (linearity + band width) + low negative feedback. the one above is better suited for dc applications. – dannyf Jul 9 '17 at 12:26
• Yes the schematic is crap, but it has the capacitor C3 as I wanted to show it (and I was too lazy to draw a brand new schematic). LM3886 would offer much better chances of success for a beginner. – peufeu Jul 9 '17 at 12:31
• @dannyf Interesting is what you talk about gains and nagative feedbacks. Why audio has moved from high open-loop gain + deep negative feedback to good open loop performance + low negative feedback? I guess my schematic is the first one :)) Could you give me sample schematic with good open loop perfromance + low negative feedback? You mean the second one is better for DC application? – Madras Jul 9 '17 at 16:14
• It's a legend originating from Otala's work back in the day. High feedback was thought to create distortions like PIM, TIM, SID, etc. This has been proven wrong since then. These all come from improper input stage design, making it nonlinear on fast signals and are easily eliminated with proper input stage: use degeneration resistors on your input pair to make transconductance linear! (and enough bias current on the input so it doesn't run out of linear excursion on normal signals). – peufeu Jul 9 '17 at 16:55

Unfortunately high frequency oscillation effect occured due to strong negative feedback:

the opamp you used may not be unity gain stable. the output transistors could be too slow. ...

1. How to calculate the filter (the cutoff frequency and ensure unitary gain in passband).

think of the capacitor as a variable resistor whose resistance is frequency dependent. Set your gain so that the phase margin is sufficient at unity gain.

• Why TL082 may be not unity gain stable? How to know which opamp is best for unity gain application? – Madras Jul 9 '17 at 16:17