Is this even stable?

[fryingpan]

I presume OP wants to stay discrete & bipolar..

If so, and taking into account his contenteness at amplification stage, a stand-alone (not directly coupled) BJT buffer could fulfill his requirements..

e.g. Emitter biased at about 5.5V with Emitter resistor of 3k3 could easily drive Q2 output of about 2.8V_RMS (4V amplitude) into a 10k load..

Yes. That's certainly an easier (and more economical) way. I was trying not to

[antonis]

That's certainly an easier (and more economical) way. I was trying not to

Then think of my last post "edit..

edit:
@fryinpan: I'd configure Q4 as current source (2 series diodes from Base to GND, delete R11, make R3 10k or so and R12 value set according to current taste..)

[fryingpan]

That's certainly an easier (and more economical) way. I was trying not to

Then think of my last post "edit..

edit:
@fryinpan: I'd configure Q4 as current source (2 series diodes from Base to GND, delete R11, make R3 10k or so and R12 value set according to current taste..)

Isn't this basically a different way to do what I did?

(And technically isn't it a current sink? Same difference, of course...).

[antonis]

Isn't this basically a different way to do what I did?

It's just a matter of taste..

(strictly speaking, one of the diodes (in close proximity to Q4) balances for V_BE thermal variations..)

[R.G.]

It's probably stable because the CFP pair doesn't have enough poles in its response to oscillate on its own, and adding the emitter follower adds another pole or two, but because it's gain-bandwidth is all used on bandwidth, not gain, the follower pole is way the heck up in frequency.

This also accounts for the peak in response at very high frequencies in the sim. I would expect the peak to be out near the GBW of the 2N3904, which OnSemi says is a minimum of 300MHz. Our old rue of thumb was that the phase change from a pole starts taking effect one decade before the actual nominal pole location, so it could be starting up at 30MHz, drizzling phase shift above that.

This is only rule of thumb speculation of course, but it seems to fit the facts. It might be interesting to remove the emitter follower and run a stability gain/phase sim on the first two stages to see what they do without the follower.

[Rob Strand]

It's probably stable because the CFP pair doesn't have enough poles in its response to oscillate on its own, and adding the emitter follower adds another pole or two, but because it's gain-bandwidth is all used on bandwidth, not gain, the follower pole is way the heck up in frequency.

This also accounts for the peak in response at very high frequencies in the sim. I would expect the peak to be out near the GBW of the 2N3904, which OnSemi says is a minimum of 300MHz. Our old rue of thumb was that the phase change from a pole starts taking effect one decade before the actual nominal pole location, so it could be starting up at 30MHz, drizzling phase shift above that.

This is only rule of thumb speculation of course, but it seems to fit the facts. It might be interesting to remove the emitter follower and run a stability gain/phase sim on the first two stages to see what they do without the follower

There's plenty of things that eat-up the phase margin of the basic circuit a lot earlier than that.
- A big one is lead inductance of the power supply.
  Bypass caps don't always help as well as you would expect.
- Capacitive loading.
- The impedance feeding the input line.
  You don't want the circuit oscillating at some position of the volume pot.
  Even simple things like being stable with the input open and shorted.
  For the open case I don't mean capacitive coupled positive feedback, which makes things worse.
  I mean the act of opening the input changes the loop gain and phase-margin.

That's why it's a good policy to add a compensation cap.   It will ensure the amplifier
is stable under all *practical* conditions.   For a preamp on a PCB you will get away with less.

You might build a unit and it's stable on the bench but it's only stable for those conditions.
When you change the set-up it's not uncommon for it to oscillate.

[amptramp]

You may not be able to see oscillation if it is at RF frequencies.  It may show up as loss of headroom as the radio frequency signal riding on top of the audio reaches the headroom limits and causes the audio to appear to be compressed at voltages it should be able to pass.  I use a 10 MHz scope for audio signal tracing and even then, there may be some RF signal riding along at higher frequencies that I could miss.

[R.G.]

You might build a unit and it's stable on the bench but it's only stable for those conditions.
When you change the set-up it's not uncommon for it to oscillate.

All too right. There's only so much that can be done in just the bare circuit design without also installing protective measures for the implementation outside the bare circuit itself.
I was mostly answering the question "how is this even stable?" It's stable because the frequency response of the output follower is way out there.

I recognize in this circuit the same basic structure/architecture as early power amplifiers. The input transistor serves as a differential transconductance amp, with the base being the noninverting input and the emitter being the inverting input. It's collector modulates a current input into the base of the PNP voltage gain stage, which provides all of the voltage gain for the circuit. The follower provides a high impedance input (as modified by the feedback networks in this case) to keep the voltage gain up, and the follower buffers the voltage gain stage. What's different about this as compared to real power amps is that the output follower in real amps do not have nearly the wide bandwidth that a single high frequency NPN does, so they add in a much lower pole.

A dominant pole compensation cap from collector to base on the PNP would cut the gain out at the highest pole very easily on this particular circuit, and is an almost exact analog of the compensation cap on most power amps; I completely agree with your note

It's generally a good idea to have the option to add a cb capacitor on Q2
to help stability. 

If I were doing this, that's where I'd put it.
And yes, filtering the input is a good idea. Wide bandwidth circuits fed with anything from DC to daylight get into trouble easily.
@amptramp:

You may not be able to see oscillation if it is at RF frequencies.  [...] I use a 10 MHz scope for audio signal tracing and even then, there may be some RF signal riding along at higher frequencies that I could miss.

Too right!!