1
\$\begingroup\$

enter image description here For the standard emitter follower you typically need input coupling capacitors and a DC bias of roughly half the supply voltage however what if the input signal itself swings around half the supply voltage?

e.g if the previous stage to an emitter follower is a standard op amp non-inverting amp with a split supply (+12V,-12V) with the input referenced to centre point but the succeeding emitter follower is referenced to -12V (i.e +24V, 0V), so The op amp output is referenced to ''+12v'' relative to the emitter follower? Could this replace the input coupling cap and resistor biasing or would it not work?

Load would be a pair of headphones (50 to 300 ohm)

example of circuit is attached below

enter image description here

Differential emitter follower? differential emitter follower

Inverting amp?

enter image description here

\$\endgroup\$
19
  • \$\begingroup\$ Draw the full idea of the circuit to stop ambiguity and sentences that make no sense. \$\endgroup\$ – Andy aka Apr 26 '20 at 11:03
  • \$\begingroup\$ I did.what sentences make no sense? \$\endgroup\$ – Jay Apr 26 '20 at 11:08
  • \$\begingroup\$ but the preceding emitter follower - what preceding emitter follower? Crop your diagram to fit the page to make it legible, \$\endgroup\$ – Andy aka Apr 26 '20 at 11:12
  • \$\begingroup\$ thanks for correction, it should be succeeding. also fixed size of image... ms paint is not the best editing software \$\endgroup\$ – Jay Apr 26 '20 at 11:15
  • \$\begingroup\$ The board has a built-in schematic editor, which is better to use than paint for schematics. \$\endgroup\$ – Neil_UK Apr 26 '20 at 11:20
0
\$\begingroup\$

Could this replace the input coupling cap and resistor biasing or would it not work?

A more convenient idea would be this: -

enter image description here

It's called a DC coupled amplifier and uses feedback directly from the emitter to set the gain at the output to be: -

$$1 + \dfrac{R2}{R1}$$

Of course, if you really want the return current for your headphones to go via -12 volts (potentially noisier) then you will need an output capacitor.

\$\endgroup\$
7
  • \$\begingroup\$ The issue is I cant really modify the op amp configuration as its part of the output buffer on a DAC PCB. I was thinking there may be another option making use of the differential or ''balanced'' output of this DAC. A bridged amplifier that would recieve this differential signal from the DAC does not reference its output to ground, the headphone/speaker is basically 'floating' between the outputs of each half the bridged amp. I added updated version of schematic to show what I mean. \$\endgroup\$ – Jay Apr 26 '20 at 13:51
  • \$\begingroup\$ This will not work because the NPN is in the loop and adds open loop gain making the circuit unstable. The NPN will have to be a PNP in that case and resistors will have to be added/adjusted. \$\endgroup\$ – le_top Apr 26 '20 at 15:23
  • \$\begingroup\$ @le_top I don't think so - it's an emitter follower and has a voltage gain less than unity and is quite likely to have a frequency response many times greater than the op-amp hence, instability won't happen. If it were a MOSFET source follower, you'd have a more realistic argument. \$\endgroup\$ – Andy aka Apr 26 '20 at 15:33
  • \$\begingroup\$ Ok, my look at the circuit was a bit too fast. Sorry. \$\endgroup\$ – le_top Apr 26 '20 at 16:42
  • \$\begingroup\$ @Andyaka I added a version of your circuit to the OP that using inverting amp, do you see any issues with that? \$\endgroup\$ – Jay Apr 30 '20 at 0:25
1
\$\begingroup\$

If you mean ...

schematic

simulate this circuit – Schematic created using CircuitLab

then with suitable op-amp biassing so that its output swings somewhere between the rails, the emitter follower will 'follow' the output at about 0.7 V lower.

\$\endgroup\$
1
  • \$\begingroup\$ yes, thats it just if the op amp is configured as a non-inverting amp/buffer. So the question is would referencing the op amp input to ground take care of biasing? \$\endgroup\$ – Jay Apr 26 '20 at 11:45

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.