0
\$\begingroup\$

I'm working on a class AB power amplifier for a school project. The desired specifications are a load impedance of 4-16 Ohms, 1% Total Harmonic Distortion (THD), and 10W amplification.

The input signal is expected to be around 10Vrms, so the first step is to attenuate it appropriately.

enter image description here

Next, the attenuated signal is directed to the non-inverting input (IN+) of the TL081CP operational amplifier.

enter image description here

I've implemented a negative feedback loop with a gain factor of 10.

enter image description here

To power the circuit, I'm utilizing a dual rail voltage source supplying +12V and -12V. The +12V rail is routed through a BD140 PNP transistor, while the -12V rail passes through a BD139 NPN transistor. In both cases, I've included a 1k ohm resistor for biasing.

enter image description here

However, I'm encountering a problem. Achieving a gain factor of 10 seems to be elusive. The signal starts to clip, and a negative offset occurs. I'm currently at a loss as to why this is happening.

measuring over load resistor

For further reference, I've included a screenshot of the entire schematic. If you'd like to explore the simulation, you can follow this link

enter image description here

I'd greatly appreciate any feedback or suggestions on how to enhance this schematic. Thank you!

\$\endgroup\$
3
  • 2
    \$\begingroup\$ I’m fairly sure the gain in non inverting configuration is 1 + rf/rg, so your gain is actually 11 and not 10 \$\endgroup\$
    – Bryan
    Oct 11, 2023 at 21:46
  • 1
    \$\begingroup\$ The input signal is expected to be around 10V Vpp or Vrms? That's quite a difference! In audio, you'll probably want to stick with rms everywhere when it comes to levels, and mention pp where needed, but be clear about it. \$\endgroup\$ Oct 11, 2023 at 21:57
  • 2
    \$\begingroup\$ A tip with falstad - you can use "create link" and put the resulting (long) URL in your posting as a link. Then it won't disappear. \$\endgroup\$ Oct 11, 2023 at 22:06

3 Answers 3

1
\$\begingroup\$

It's being limited by the Falstad 741 model output swing and current drive. This is causing somewhat asymmetric clipping which is influencing the average output voltage.

Two things seem to help the clipping:

  • increase the voltage to 15V
  • increase the 741 model current limit to 40mA

This seems to get rid of the clipping. Try it here.

You can also try using Sziklai configuration to boost the current to the transistor bases. This seems happy enough on +/-12V without tweaking the 741 model. Try that here.

\$\endgroup\$
1
\$\begingroup\$

Since \$P=V^2/R\$, thus \$V=\sqrt{P R}\$. To get 10W on the highest load impedance, we need to put \$V=\sqrt{10\cdot16}\approx 13{\,\rm V_{rms}}\$ across it. So that means 18V peaks.

So, in fact, you don't need to attenuate the signal in some cases. You'll need to amplify it a little bit from input to output for the highest-impedance load. You will want to attenuate it for lower impedance loads, and in general just to maintain volume control as would be expected on any audio amp.

The amplifier doesn't really need an op-amp for the distortion level you are aiming for. Any of the "standard" class-AB audio amp designs will work better than that. An op-amp could be used as a DC servo to maintain a fixed bias.

To get some headroom for 13Vrms on the output, with a ground-referenced load, a ±24V supply would be adequate. Then you can have a class-A preamplifier to get a bit of voltage gain, then a driver, and finally the output stage.

\$\endgroup\$
1
\$\begingroup\$

Your amplifier circuit is a very old design that produces a lot of class-A heating. Instead you should design a normal class-AB amplifier circuit.

Here us an old circuit that is (was) similar to yours: 4.5W amplifier

\$\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.