0

I would like to have some advice on the power amplifier in the diagram below before building it. Will be used as guitar amplifier.

The current rises significantly with as input frequency rises. I do not understant why totally.

I also do not understand how the quiescent current for Q5 should be chosed.

Thank you, Have a nice day,

enter image description here

  • Where did the design come from? – jonk Jun 22 '18 at 15:54
  • It is a mix of many circuit I have seen in books and online and what I understand from the Douglas Self audio amplifier book. –  Jun 22 '18 at 16:36
  • Why are different values used for R1, R2, R25, and R26? What power do you expect (one or two watts? or what?) Why no provision for setting up a quiescent class-AB current? What is C1 doing for you? Are you intending to combine biasing of Q5 with NFB function by R10? Q1 and Q6 might use base resistors (dampen oscillation tendencies.) Since this is your own amalgam, I think you owe us some discussion about your thinking, rather than just throwing this out on the table and asking for comments. Your mental process is essential to saving us time and effort in helping you. So talk. Please. – jonk Jun 22 '18 at 16:38
  • The output stage current you show is unrealistic, you would be dissipating about 118W in the output stage (4 devices). Choose an output stage idle current of about 50 mA. – Jack Creasey Jun 22 '18 at 16:38
  • What C1 do? Why R1 and R2, R25, R26 are not in the emitter side? – G36 Jun 22 '18 at 16:39
  • @Jonk - I would like to have between 5 and 10 Watts of power. - Different values are my mistake. They should all be 0.75R. Sorry. - In my limited comprehension of the whole thing, C1 and R10 where used for NFB. –  Jun 22 '18 at 16:45
  • @SimonTurcotte See: [1W class AB #1](https://electronics.stackexchange.com/questions/374108/how-to-keep-current-from-draining-out-the-bottom-of-this-push-pull-amplifier/374135#374135), [1W class AB #2](https://electronics.stackexchange.com/questions/330169/design-of-a-discrete-bjt-current-buffer-for-low-current-op-amp/330205#330205), and [1W class A](https://electronics.stackexchange.com/questions/368614/why-cant-class-a-amp-drive-8-ohm-speaker-with-just-one-bjt/368660#368660), to start. – jonk Jun 22 '18 at 17:20
  • @SimonTurcotte Then see: [5W class AB](https://electronics.stackexchange.com/questions/363367/function-of-specific-capacitor-in-audio-amplifiers-negative-feedback/363376#363376), [10-15W class AB](https://electronics.stackexchange.com/questions/348493/highly-distorted-output-of-class-b-power-amplifier/348566#348566), and [50W Self amplifier](https://electronics.stackexchange.com/questions/325384/bjt-2-stage-amp-output-problem/325502#325502). – jonk Jun 22 '18 at 17:21
  • This design is NOT rational and should not be used... –  Jun 22 '18 at 18:49
  • It appears the user Simon Turcotte has been deleted (by their own request, perhaps.) Either way, their name has been replaced and they are gone. – jonk Jun 23 '18 at 19:53

3 Answers3

1

The current rises significantly with as input frequency rises. I do not understant why totally.

You appear to have a 100 uF from the output to the positive supply rail and this doesn't look like a good idea for anything: -

enter image description here

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

The only DC input to R5 is the R10/R9 voltage divider output. This divide ratio was apparently chosen to result in a particular voltage across R14 when the output rail is at its midpoint. Q5 then acts like a current sink that works against R5 to set the bias point of Q1 and Q6. Those in turn control the high and low side drivers.

The DC voltage of the output rail is therefore set by the R10/R9 divider. You should make sure that this DC output voltage is close to half the supply voltage. If not, adjust either R10 or R9 a little bit to make it so.

The quiescent current of the final output stage is poorly controlled. The final high and low side drivers (Q8,Q9 high side, Q2,Q10 low side) could even be mostly off with no input signal. With signal, they are turned on more, and could dissipate significant power then.

This is a dubious design. Run away. You can find better out there. The way the output stage current is controlled seems to be relying on the fact that it will be low with no signal. That relies on minimum gain of some of the transistors, and will also result in crossover distortion. The complete lack of global negative feedback for the signal is also troubling. Then what the heck is C1 supposed to accomplish?

Run away.

Olin Lathrop
  • 310,974
  • 36
  • 428
  • 915
0

This is a poor design for low impedance 8 Ohm load drive but adequate yet overkill as a 50x gain amp into a 1k Ohm load.

The fundamental reason is the output stages are current sources driven by 240 ohm base resistors and that means high impedance into low impedance reduces the gain according to speaker non-flat impedance.

There is also no negative feedback so distortion is high.

Any audio power IC will do a much better job on bridge mode without the massive DC blocking cap.

Tony Stewart EE75
  • 1
  • 3
  • 54
  • 182
  • The massive cap is something I have at home that I want to include in the amplifier... Where should I start to improve the design to make it adequate for an 8ohms load? –  Jun 22 '18 at 19:22
  • Why make when there are so many car radios in surplus , pick and pull aka recycling shops with high power Delco-like radios with 22W per side – Tony Stewart EE75 Jun 22 '18 at 19:54