Q about NPN/PNP amplifier circuit

[mdcmdcmdc]

Hi folks,

I was digging around in an old issue of wireless world and found this circuit fragment that looked interesting:



I just breadboard'd it and while it's doing some actually not-bad tone shaping to a guitar signal it's not boosting the output volume at all - it seems to be right around unity.
I'm far out of my depth when it comes to knowing what to expect by looking at a fragment like this, but I'm curious if I should be seeing some amplification of the input signal? It seems like a useful buffer either way (the accompanying article says input impedance of ~1.5M, output around 30R).

Thanks!

[PRR]

Tr2 C connects direct to Tr1 E.

Unity gain super buffer.

[Rob Strand]

It also has bootstrapping on the base of TR1.

The common circuit with gain and no bootstrapping is this one.   The cap and resistor on TR2's emitter is optional, although you can't just rip them off you need to adjust 5.2k (on you schem) by making it smaller.

https://www.electronics-notes.com/articles/analogue_circuits/transistor/amplifier-npn-pnp-two-transistor-circuit.php

[mdcmdcmdc]

That's the one Roger Mayer uses a lot, if I'm not mistaken?

[amz-fx]

Here is a related circuit that has gain. I have breadboarded it and verified that it works though some resistor values will need to be tweaked for best results.



You could change this to use an NPN on the input instead of the 2N4416, but the bipolar transistor would need a resistor from the base to V+. Again, values would need to be adjusted, especially the base-to-ground R. This would make it similar to the one that Rob posted.

regards, Jack

[11-90-an]

sziklai pair, perhaps?

i learned about these from Jonny Reckless's use of them in his PMC... i noticed his implementation here have some gain...

(smart people please do correct me if i'm wrong... )

[mdcmdcmdc]

Thanks for these - very helpful!
I'll play around with the fragment you posted, Jack - looks interesting, I'm curious to see how it sounds.

I'm familiar with Sziklai pairs (in theory at least), but the bootstrapping in the first circuit I posted isn't something I'd seen before so I wasn't really sure what its purpose/effect is. The article was about increasing impedance by using two-BJT configurations, and there was a short section of "practical applications" at the end with little to no explanation, so I was curious (especially since it was a 9V circuit which I haven't seen in radio magazines too often).

In any case, if anyone happens to need a 9V super buffer, this seems like a pretty decent one.

[Bunkey]

https://www.electronics-notes.com/articles/analogue_circuits/transistor/amplifier-npn-pnp-two-transistor-circuit.php

"The two transistor amplifier offers a reasonably high impedance while providing a low output impedance. It is an ideal transistor amplifier circuit for applications where a higher level of gain is required than that which would be provided by a single transistor stage."

Not to bring the level of this thread down but;
If the amount of gain signal output in a common emmitter amplifier is limited by the supply voltage, does something like this allow for a stable circuit that gets closer to the full supply swing without the reduction in gain that you would get from C-B resistor feedback (for example) on a single BJT?
Am I paraphrasing this correctly?

The other benefit being that it would negate the need for an emitter follower preceding a gain stage to buffer the input of a boost pedal for example? Or does the addition of R4 between E of Tr1 and C of Tr2 on that linked amplifier schem no longer make it a buffer??

[antonis]

If the amount of gain signal output in a common emmitter amplifier is limited by the supply voltage,

We don't deal with a common emitter amplifier here..!!!

[mdcmdcmdc]

Here is a related circuit that has gain. I have breadboarded it and verified that it works though some resistor values will need to be tweaked for best results.



You could change this to use an NPN on the input instead of the 2N4416, but the bipolar transistor would need a resistor from the base to V+. Again, values would need to be adjusted, especially the base-to-ground R. This would make it similar to the one that Rob posted.

regards, Jack

This looks to be a similar circuit adapted for 2xBJT:



Though if the calculation is the same, it looks like 10L/100R is going to yield 100x Vin? Seems like a place to start.
EDIT: just noticed that the 2xBJT version omits the resistors on the C and E of the PNP transistor...

Getting back to the bootstrapped buffer at the top of the thread, just curious (and clueless) if there's anything in that circuit fragment that affects frequency response intentionally, or if any changes to the sound of what's coming out of it are by-products of the circuit doing its thing and stabilizing the voltage?

[Bunkey]

If the amount of gain signal output in a common emmitter amplifier is limited by the supply voltage,

We don't deal with a common emitter amplifier here..!!!

I appreciate that; I was asking what the difference/benefit to the output signal level would be if comparing the two topologies but I realise my own oversight now in that I misinterpreted "a higher level of gain" as output, when in fact the output would be limited in the same way. Don't mind me, I'm just stringing along at the back.

[Rob Strand]

Not to bring the level of this thread down but;
If the amount of gain signal output in a common emmitter amplifier is limited by the supply voltage, does something like this allow for a stable circuit that gets closer to the full supply swing without the reduction in gain that you would get from C-B resistor feedback (for example) on a single BJT?
Am I paraphrasing this correctly?

The circuit is basically two common emitter (CE) amplifiers in cascade, then then feedback is added.    The fact you have two amplifying stages means you have a lot more gain to play with.      The output stage can swing nearly rail to rail provided the load isn't too low.   While the output impedance is low it hides the fact the output can't swing to full negative rail if the load impedance is too low.  That's set by the 10k resistor.   Simple one transistor buffers also suffer from this effect.   

A CE amplifier with bypassed or no emitter resistors will produce a gain of about 170.  The CB resistor is applying feedback so naturally the gain must reduce from this base-line 170.  The side effect of adding the CB resistor is it lowers the input impedance which can then load the circuit driving it.

The other benefit being that it would negate the need for an emitter follower preceding a gain stage to buffer the input of a boost pedal for example? Or does the addition of R4 between E of Tr1 and C of Tr2 on that linked amplifier schem no longer make it a buffer??

If you make R1 and R2 high enough then yes the input impedance can be made quite high.   For the general circuit in the link, TR1 is connected as a CE amp not a buffer.   The feedback helps raise the input impedance of TR1 which would otherwise be quite low.   
However, the catch is the more gain you ask from the circuit the lower R5 needs to be and there is a point where there's not enough gain in the circuit to make the input impedance look low.    So if you are careful with the design and you didn't ask too much gain you can get a high input impedance from the circuit.     Another catch is when you rely on feedback to raise the input impedance if the amplifier clips, as it would in a distortion pedal,  you can loose the feedback and input impedance will drop dramatically to it's natural low value.   A circuit with an input buffer is a more immune to this effect.