0
\$\begingroup\$

I can't figure out why this amplifier is capping out

enter image description here

enter image description here

Aditional details:

I can identify 4 waveforms that resemle the output 1st: enter image description here enter image description here

2nd:

Base of Q42

Waveform same as 1st

3rd:

enter image description here enter image description here

And 4th

enter image description here

Waveform same as 3rd

The rest of them besides Vout were just straight lines.

\$\endgroup\$
11
  • 2
    \$\begingroup\$ Q43 appears to be an incorrect part or incorrectly connected. \$\endgroup\$
    – vir
    Nov 14, 2022 at 23:39
  • 1
    \$\begingroup\$ BTW, "generally" it's considered best to draw your schematics with (+) voltages at the top, and (-) or GND at the bottom. Your very lowest voltage rail (-12V) is at the top and your most positive (+12) is at the bottom. Makes it hard to read. \$\endgroup\$
    – Kyle B
    Nov 14, 2022 at 23:43
  • 1
    \$\begingroup\$ Incidentally, the word for what you're talking about is "clipping", not "capping out". You could also say "saturating"; I've just never heard "capping out" before and was wondering what that meant before seeing the pictures. \$\endgroup\$
    – Hearth
    Nov 15, 2022 at 1:44
  • 1
    \$\begingroup\$ PStechPaul is correct, Q43, the VAS, should be an NPN but the emitter and collector should be swapped. A useful clue is that the Miller cap C1 should be between the base and collector. Just flip Q43 around the other way and see what results you get. \$\endgroup\$
    – GodJihyo
    Nov 15, 2022 at 2:10
  • 1
    \$\begingroup\$ @DinuVA Would you walk me through your thinking with this schematic? (I sure hope it isn't copied from the web -- it's just wrong.) I'd like to listen for a moment to you about what you understand or think you do so that I have an idea of how to better help. GodClara, sorry I mean GodJihyo, has it dead to rights. Your VAS has one foot in the right place and the other foot somewhere it should not ought to be. And that's not the only issue I see. So I need to listen to you talk about how you arrived at this schematic. \$\endgroup\$
    – jonk
    Nov 15, 2022 at 2:52

1 Answer 1

Reset to default
3
\$\begingroup\$

The obvious problem, as noted in the comments, is that the VAS (Voltage Amplifier Stage) transistor is connected incorrectly.

Once you get that straightened out you need to look at the biasing. From what I can tell it looks like your bias currents are too high. The input stage looks to be around 20 mA, I would cut that down to 1 to 2 mA. The VAS stage should be maybe 5 to 10 mA.

One way to find the best bias points is to replace the transistor constant current sources with a constant current source model in the simulator, then step the CCS and see at what value you get the least distortion. Do one at a time, Q40 for the input stage, Q41 for the VAS stage, and Q42 for the class A stage. Using the built-in current source allows you to easily change it for testing. Once you get the values correct you can put the transistor ones back in and adjust them to get the correct currents.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.