1

I have been designing an A/B voltage amplifier using the TIP31C/32C power transistors (schematic shown below.)

I have a significant loss in voltage gain which I haven't able to negate.

Is this inherent to a class A/B output stage or is the gain of my transistors just not capable of maintaining the voltage supplied? Could someone possibly point me in the right direction of some resources that may help me improve this?

Circuit schematic:

enter image description here

Below is a graph of my transient response, with red being the amplifier ouput, green blue and blue are the vin and base voltages.

enter image description here

JRE
  • 67,678
  • 8
  • 104
  • 179
Samuel Newport
  • 13
  • 2
  • It is not class-AB, instead it is class-B with severe crossover distortion at low levels. The input is missing a series input coupling capacitor causing the input transistor to be saturated. If the input has a coupling capacitor then the biasing of the input transistor causes it to be almost cutoff. I agree that the values of R8 and R13 are way too high and cause output voltage loss. The ratio of R12 and R11 also reduce voltage gain. – Audioguru Mar 03 '22 at 16:28
  • Sorry, but I don't see the [severe] crossover distortion you mentioned. – SteveSh Mar 03 '22 at 22:21
  • @Audioguru The outputs in the simulation (shown above) dont indicate any distortion Thank you for highlighting the R11/12 ratio but i misunderstand how you can tell there is distortion when the output is matching the input – Samuel Newport Mar 05 '22 at 17:51
  • It is interesting that the simulation doesn't show crossover distortion. You should need to have at least four diode drops between the bases of Q5 and Q6 to avoid crossover distortion. It might be interesting to run the simulation at much lower amplitude, like 2 Volts on the input and see if it becomes more apparent. – rpm2718 Mar 05 '22 at 18:55
  • @rpm2718 When reducing the supply rails to +- 15V there is indeed a slight crossover distortion. Thank you ive added a couple more diodes. In the lab the circuit works perfectly aside from the large noise factor... Im assuming this is thermal and from the parasitic capacitances accross the breadboard – Samuel Newport Mar 06 '22 at 11:20

3 Answers3

2

I believe R8 and R13, at 10 Ohms each are your issue. Best case if Q2 is fully conducting with your 8 Ohm load, the output voltage will be:

V = -30 * (8/(8 + 10)) = -30 * 8/18 = -13V

You're getting -10V which is to be expected because Q2 will never be 0 Ohms.

Same situation on the high side. So as-designed you will NEVER see more than +/- 13V at the output.

jwh20
  • 7,812
  • 1
  • 16
  • 27
1

Just do simple math on the peak positive voltage and you'll quickly see where the problem is: -

enter image description here

  • The +28.5 volts is due to using a Darlington configuration for Q6 and Q1

  • The +12.7 volts is due to the potential divider of R8 and R15

Andy aka
  • 434,556
  • 28
  • 351
  • 777
0

The amplifier is not providing the gain you expect because the inverting voltage gain stage that consists of Q3, R11, and R12 is not designed correctly.

Due to the lack of an input coupling capacitor and appropriate dc biasing network at the base of Q3, Q3 saturates, putting the collector voltage just a bit above the emitter voltage. So instead of providing inverting gain, the collector voltage of Q3 pretty much just follows changes in the Q3 base signal.

Next, your Darlington output stage ends up dividing down the voltage by a factor of 2.25 due to the voltage divider formed by R15 and R8. So the net effect is that your amplifier has gain of roughly 1/2.25. This is reflected in your graphs, where the output seems to be approximately this 2.25 factor less than your input.

Could someone possibly point me in the right direction of some resources that may help me improve this?

It appears that you started off trying to do something like this: enter image description here

(image source)

except with a Darlington output stage. Then you added feedback instead of an input coupling capacitor and dc bias network. Normally, if you want to make a feedback amplifier, your input stage would be differential, with the feedback signal brought into one of the differential inputs.

As for resources, a widely-used reference on the subject is the book Audio Power Amplifiers, by Bob Cordell.

rpm2718
  • 2,632
  • 1
  • 6
  • 14