1
\$\begingroup\$

I'm working on a push-pull stage audio amplifier. I'd like your opinion on whether I'm on the right track with this, both theoretically and practically.

In this circuit I'll be using the following components:

Q1&Q2 TIP31C & TIP32C with LM358N op-amp (this was the only available at my university.)

My circuit:

enter image description here

enter image description here

\$\endgroup\$
24
  • 3
    \$\begingroup\$ I have to point out that you are asking opinions and have no specific question about the circuit. We also don't know what specifications you require from the amplifier, so it will be impossible to say if you are on the right track or not. You might want to explain why each part is there before someone asks why R1, C2 or C6 are in the circuit. \$\endgroup\$
    – Justme
    May 30, 2022 at 17:59
  • \$\begingroup\$ Sabretooth, you can do all that with a single rail. Do you already have a dual rail supply you want used? Or would a single supply be cheaper and easier? \$\endgroup\$
    – jonk
    May 30, 2022 at 18:05
  • \$\begingroup\$ The steady-state power dissipation on Q2 & Q3 is near 3 W, which is a "little" too high. If R7 & R8 are 5 k, it should be 200 mW. \$\endgroup\$
    – Antonio51
    May 30, 2022 at 18:10
  • \$\begingroup\$ NB: the voltage on 8 Ohm load without distortion (with R7=R8=1 kOhm) would be 10 V peak to peak... EE&O. \$\endgroup\$
    – Antonio51
    May 30, 2022 at 18:24
  • \$\begingroup\$ Changing Q2-Q3 to Darlington BD675-BD676 --> voltage on 8 Ohm load without distortion (with R7=R8=1 kOhm) would be 20 V peak to peak \$\endgroup\$
    – Antonio51
    May 30, 2022 at 18:56

2 Answers 2

6
\$\begingroup\$

This modified circuit will have "better" performance due to negative feedback via R2 connected directly to the output and, dc coupling the op-amp to the power-transistor stage: -

enter image description here

The above modifications are pretty much what most class AB audio amplifiers use.

I'd like your opinion on whether I'm on the right track with this

In my opinion, you are not really going places with your design. I mean it will work but, it will have higher distortion than mine (due to the way I've implemented negative feedback) and, if you improved the op-amp type (higher speed and slew rate) mine would be even better whereas your design wouldn't improve with a better op-amp.

\$\endgroup\$
1
  • \$\begingroup\$ apologize for the late reply, yes, I'll be starting from scratch \$\endgroup\$ Jun 1, 2022 at 5:45
1
\$\begingroup\$

D1 and D2 on your circuit set up the idle current. This is a poor match to Vbe. Put a rubber diode which is sometimes called a VBE multiplier in place of the 1N4001 pair. Place the rubber diode connected transistor in good thermal contact with the output transistors, like on a common heatsink.

\$\endgroup\$
5
  • 1
    \$\begingroup\$ I'd like to mention that "rubber diode" simply means a Vbe multiplier, which is far more common term for it. \$\endgroup\$
    – Justme
    May 30, 2022 at 20:11
  • 1
    \$\begingroup\$ @Autistic we don't have that available at the Uni \$\endgroup\$ Jun 1, 2022 at 5:47
  • \$\begingroup\$ @sabretooth2438 .You can make your own rubberdiode with say a TIP31 and try 2 say 1 K resistors . \$\endgroup\$
    – Autistic
    Jun 1, 2022 at 9:42
  • \$\begingroup\$ Can I make it with any other transistor? Since we have limited amount for each student \$\endgroup\$ Jun 1, 2022 at 10:27
  • \$\begingroup\$ @sabertooth2438 yes you can use a different transistor.Do measure your idle current and adjust if needed. \$\endgroup\$
    – Autistic
    Jun 2, 2022 at 7:42

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.