Analysing Power Amplifier circuits-General discussion

I never say the point of the circles.

Maybe if it's a Darlington pair, the circle shows that it's one component and not two? I always use two arrows to symbolise a Darlington so the ring isn't needed.

The only time I think a circle is required is if it isn't obvious, for example it's a matched pair of transistors on the same die.

there are darlingtons with internal resistors and those should be drawn with the internal resistors and circles. unfortunately some receiver manufacturers have a habit of drawing those devices as a regular transistor. some manufacturers also use a second, sometimes dotted circle to show the device is on a heatsink, which can be useful information in a schematic. schematic symbols are intended to be somewhat descriptive of function, and that makes for some interesting variations. resistors are drawn as zigzags because a resistance at one time was usually a wirewound device like a heating element. a zener diode is drawn with a depiction of the device's characteristic curve on the anode stripe. an inductor symbol is a depiction of a coil of wire, and a cap symbol is a depiction of two insulated plates.

unclejed613 said:

there are darlingtons with internal resistors and those should be drawn with the internal resistors and circles. unfortunately some receiver manufacturers have a habit of drawing those devices as a regular transistor. some manufacturers also use a second, sometimes dotted circle to show the device is on a heatsink, which can be useful information in a schematic.

Click to expand...

That makes sense for transistors on heatsinks or those that contain additional components but circling every single transistor is pointless in my opinion.

schematic symbols are intended to be somewhat descriptive of function, and that makes for some interesting variations. resistors are drawn as zigzags

Click to expand...

On the other hand, the square box best represents carbon composition and thin film resistors.

What is effect of Q7-8 in the attached diagram?


is the voltage bias for the VBE MULTIPLIER = VBB = 2 x ( VBE Q10 + VBE Q12)?

fantabulous68 said:

What is effect of Q7-8 in the attached diagram?


is the voltage bias for the VBE MULTIPLIER = VBB = 2 x ( VBE Q10 + VBE Q12)?

Click to expand...

As we've told you repeatedly, that 'circuit' is completely nonsense - it wouldn't work, so they have no effect.

Q7-Q8 is a darlington pair connected to the collector of Q3. In the case that Q3 conducts anything, Q7-Q8 slams the output drivers against the Vcc rail producing a square wave.

As we've told you repeatedly, that 'circuit' is completely nonsense - it wouldn't work, so they have no effect.

Click to expand...

The circuit consists of stages that are used in the design of a Power amplifier.
input, driver and output stage.

And those stages comprise of sub-circuits

eg. output stage: composite transistors (Darlington & Sziklai types)
driver: VBE multiplier and current source(current mirror, bootstrap circuit)

So if I attach a diagram of a sziklai darlington configuration and ask you what it does....your response is going to be
As we've told you repeatedly, that 'circuit' is completely nonsense - it wouldn't work, so they have no effect.
Because its not in a "functional" circuit.

But there is so much that can be said about that diagram.

I think you shouldn't reply at all to my thread if you dont want to help.

Your output circuit is missing important resistors that quickly turn off the transistors.
It is also missing emitter resistors for the output transistors.

Audio amplifiers do not use 3 transistors in series on each side. They use a darlington pair or a Sziklai pair.

That circuit uses both a Sziklai and a Darlington.

Q1 ans Q2 from the Sziklai which is in Darlington with Q3. I've never seen this before, it mush have a very high input impedance and a high voltage loss too.

How do you pronounce "Sziklai" by the way? I've only seen it written on paper and never heard anyone say it. I still can't spell it, I have to Google every time for the correct spelling.

fantabulous68 said:

The circuit consists of stages that are used in the design of a Power amplifier.
input, driver and output stage.

And those stages comprise of sub-circuits

eg. output stage: composite transistors (Darlington & Sziklai types)
driver: VBE multiplier and current source(current mirror, bootstrap circuit)

Click to expand...

I'm fully aware of the stages of an amplifier, have designed, built, and repaired many over the years.

But you can't just randomly throw the pieces together, and the 'circuit' in question is completely wrong in every respect, and doesn't bear any resemblance to an amplifier. The first circuit you posted wasn't quite as bad - at least that was a recognisable design, and just needed correcting. This last one is so far wrong, it wouldn't be correctable - throw it away and start again.

So if I attach a diagram of a sziklai darlington configuration and ask you what it does....your response is going to be
As we've told you repeatedly, that 'circuit' is completely nonsense - it wouldn't work, so they have no effect.
Because its not in a "functional" circuit.

Click to expand...

I've no problem with that part of a circuit, but as already mentioned by someone else - it's a non-practical example out of a text book - check the Amplab link I posted earlier that shows the rest of the parts, both of the output stage, and the rest of the amplifier. A Sziklai is just a different kind of darlington.

Here's the actual link to the Sziklai version, note there's 8 pages in the project, showing three different output stage versions:

Sziklai - Complementary Feedback Power Amp Output

But there is so much that can be said about that diagram.

I think you shouldn't reply at all to my thread if you dont want to help.

Click to expand...

Forums are an historical record - if we let you post complete rubbish and designs that are wrong in every respect, we're going to get people trying to build them. There are hundreds (if not thousands) of working designs on the net, why not pick one of those?, or at least look at them, and compare them to what you keep posting.

Your output circuit is missing important resistors that quickly turn off the transistors.
It is also missing emitter resistors for the output transistors.

Click to expand...

Yeah you are right, to make that diagram practical would include the principle addition of base emitter resistors which would assist in turning the devices off & also bias stabilising resistors in series with the emitters of the two power transistors Q3 & Q6 to reduce thermal runaway

But you can't just randomly throw the pieces together, and the 'circuit' in question is completely wrong in every respect, and doesn't bear any resemblance to an amplifier.

Click to expand...

Repeat: I did not design the circuits I post

Forums are an historical record - if we let you post complete rubbish and designs that are wrong in every respect, we're going to get people trying to build them. There are hundreds (if not thousands) of working designs on the net, why not pick one of those?, or at least look at them, and compare them to what you keep posting.

Click to expand...

I did pick the circuit of the net. This thread is for general discussion as i stated when I started the thread. If i wanted to build the amplifier, i would post my thread in the design project forum. I do not ask you how the entire amplifier/cct works. I target my questions to specific sub circuits.

fantabulous68 said:

Repeat: I did not design the circuits I post

Click to expand...

That's even worse - you're copying someone elses complete rubbish!.

I did pick the circuit of the net. This thread is for general discussion as i stated when I started the thread. If i wanted to build the amplifier, i would post my thread in the design project forum. I do not ask you how the entire amplifier/cct works. I target my questions to specific sub circuits.

Click to expand...

You can't ask questions on a non-working sub-section of a non-working design - like I said (and everyone else has) the 'design' is complete nonsense.

If you got it off the net, then post the link - previously you posted you got it off a South African Uni site.

That's even worse - you're copying someone elses complete rubbish!.

Click to expand...

I am not copying, i said that i did not design the circuit and i put the link up already.

You can't ask questions on a non-working sub-section of a non-working design - like I said (and everyone else has) the 'design' is complete nonsense.

Click to expand...

Whats the VBB required by the VBE multiplier ? What components make up the VBE multiplier? All these questions have answers which can be obtained from the diagrams posted.

fantabulous68 said:

Whats the VBB required by the VBE multiplier ? What components make up the VBE multiplier?

Click to expand...

What does VBB mean? The voltage from a Blubber Belly?
R5, R6 and Q9 form the Vbe multiplier.

you also invalidated the current source by putting a capacitor across it's output...

using a darlington as a VAS stage is NOT a good idea. first of all you are still missing a compensation cap in the VAS. with a darlington as the VAS, your compensation cap will have to be much larger than normal. all you have right now if you were attempt to build this would be a radio transmitter for a few seconds, then a smoke generator.

What does VBB mean? The voltage from a Blubber Belly?

Click to expand...

its the bias voltage, its standard terminology used when dealing with a VBE Multiplier.

You showed me what the VBB was equal to on that diagram posted on another forum site.
VBB=Vbe1+Vbe2=total vbe for that diagram

all you have right now if you were attempt to build this would be a radio transmitter for a few seconds, then a smoke generator.

Click to expand...

lol, i believe you. I was standing in a lab and then i saw this guys circuit smoke & he wasn't even aware .


Any comments on the reference voltage in the diagram?

previously you posted you got it off a South African Uni site.

Click to expand...

This explains it. Whoever designed the circuit had spent more time smoking weed than studying electronics. Or perhaps they designed it while smoking weed.

fantabulous68 said:

lol, i believe you. I was standing in a lab and then i saw this guys circuit smoke & he wasn't even aware .


Any comments on the reference voltage in the diagram?

Click to expand...

actually the reference voltage for the current source was better when you had the two transistor current source. if i were you i would scrap this whole design, and start off with a simple Lin topology amp and improve and optimize that one this single ended thing with the whacky current mirror VAS is not going to work very well if at all....

here's a simple amp with loads of room for improvement and tweaking. i "built" it on LTSpice in about 10 minutes, and it ran on the first try with no tweaking of component values. there is a lot of room for improvement on this design, and they're all performance tweaks, not changes to just get it working. but as-is, the amp has 0.1% distortion at full output into 8 ohms.





(edit) actually i did have to tweak one component, the compensation cap C2. when i added an 8 ohm load, there was a very small (less than 100mV) 1Mhz oscillation at the output. it was easily cured by changing C2 to 50pf