Simple class A amplifier how it works

[tomizett]

I'd also recommend one of R.A. Penfolds books on audio power amplifiers as a good introduction - Babani press had some good beginners books, although they may be hard to find outside the UK. And yes - get a 'scope! Doesn't matter how basic. "It's like a second pair of hands" as my grandfather would have asid.

 

[schmitt trigger]

Ditto getting a scope.
Even a plain analog one will do for audio.

 

[billybob]

Ditto getting a scope.
Even a plain analog one will do for audio.

Is there one you would recommend? Im afraid of accidentally frying it, Ive gone through about 5 multimeters in the past 2 years

 

[Diver300]

Is there one you would recommend? Im afraid of accidentally frying it, Ive gone through about 5 multimeters in the past 2 years

Oscilloscopes are more difficult to fry than multimeters, as oscilloscopes don't measure current or resistance.

My money would go to Picoscope if I was buying now.

 

[unclejed613]

ok... here's a first draft of what i was talking about... it uses a pretty standard input stage with diff amp and voltage amplifier.... the output stage is a bit unorthodox, it's a single ended emitter follower stage with a constant current source for the emitter load... the current source allows the output stage to operate more linearly than using a resistor load and reduces the distortion a LOT... even as a first draft with no tweaking it has 0.1% THD where normal single-ended class A amplifiers have around 1 to 3% THD, mostly even harmonic distortion. while this currently consists of a dual PNP output stage with a PNP diff pair, it can be flipped upside down easily to use the OP's 2SC5200 output devices with an NPN diff pair... the C5200 devices have less beta droop, so even without much tweaking, the distortion introduced by the output stage would be reduced a lot... i'm not a big fan of pure class A amps, but done right they can perform pretty good compared to the usual "burning amp purist" designs

PS... there is a bit of an "easter egg" in the schematic.... something that probably would let out the "magic smoke" in a real build of this design... anybody who can tell me what it is gets a "Wile E. Coyote Sooooper Genius" award....

 

[MacIntoshCZ]

Is there one you would recommend? Im afraid of accidentally frying it, Ive gone through about 5 multimeters in the past 2 years

ds1054z
mso5074 - can be unlocked to 800Mhz
but for newbie search for second hand digital one. ds1052e could be cheap

 

[rjenkinsgb]

something that probably would let out the "magic smoke" in a real build of this design...

Well, one thing could be that is has a standing current in the output stage of around 25A or so, with continuous power dissipation of around 500W. It would need rather beefier output transistors.

If anyone really wanted to try it, I may be able to come up with a suitable heatsink. I'd only use it in winter when you need the extra room heating, though!

 

[billybob]

Well, one thing could be that is has a standing current in the output stage of around 25A or so, with continuous power dissipation of around 500W. It would need rather beefier output transistors.

If anyone really wanted to try it, I may be able to come up with a suitable heatsink. I'd only use it in winter when you need the extra room heating, though!

View attachment 129650

Thats a pretty buff looking heatsink!

 

[audioguru]

I am miles away but I feel the heat and smell the smoke from the class-A amplifier.

 

[Nigel Goodwin]

I am miles away but I feel the heat and smell the smoke from the class-A amplifier.

And can hear the distortion!

 

[unclejed613]

here's the mostly PNP version and mostly NPN version side by side....

just flipped it upside down and used complementary transistors (except for one which the basic install of LTSpice omitted for some reason or other)....

 

[unclejed613]

Well, one thing could be that is has a standing current in the output stage of around 25A or

i kind of expected the output stage current to be rather high myself but it isn't... it's actually around 1.5A.... Q7 is limiting the current, and because of the global feedback keeping Q7's emitter at 1/2Vcc at idle. as i said it's a first draft.. lots of room for tweaking component values... i was actually worried about the voltage amplifier stage being driven beyond the limits of the 2N3904/3906 devices, but even those seem to be well within safety limits. i'll have to try a build of this it seems like an interesting project...... i was thinking about what i said to the OP, and decided to find out whether it was doable.... why i haven't seen anyone else try it is anybody's guess...

 

[rjenkinsgb]

i kind of expected the output stage current to be rather high myself but it isn't... it's actually around 1.5A

The constant current source cannot be working at all, then; if it were, it only regulate once the voltage across the emitter resistor exceeded both the base-emitter and LED voltage drops, probably 2.5V or so, giving around 25A.

The sim is likely limiting it due to the base current via R11 only being few mA.

 

[unclejed613]

i'll work with it a bit more when i get home from work... i worked on the output stage by itself and ballparked everything to get it working within it's linear range, then did the same thing with the input/VAS stage... then i merged the two and tapped the feedback from the output point... but there is plenty of room for optimization.... i think the addition of feedback affected the output stage operation somewhat as well...

 

[Tony Stewart]

Simple scaleable class AB

https://www.ti.com/lit/ds/symlink/opa454.pdf

 

[Nigel Goodwin]

Simple scaleable class AB View attachment 129669

https://www.ti.com/lit/ds/symlink/opa454.pdf

And that's REALLY nasty - quality, what quality

 

[audioguru]

The opamp with an 80V supply driving T0-220 output transistors to maybe 2A peak will melt from trying to produce 2.6W into the transistors. The transistors also might melt trying to produce 77W into a normal low impedance speaker.
The severe crossover distortion will sound awful.

 

[unclejed613]

And that's REALLY nasty - quality, what quality

if you can SEE the crossover notch distortion then you know it's going to sound awful....


so i tweaked R11/R19 by changing it to 50 ohms, and changing D4/D8 to 1N4004.... the idle current is now 3.5A (70W dissipation in the combination of output device and CCS... still not making full use of the CCS, but about 4-6db improvement in distortion

 

[Tony Stewart]

I'm sure you all can back up your opinions. I know I can https://tinyurl.com/yxmqb3gm Show me the excess power dissipation and crossover distortion with 100k=Aol and 33=Acl

 

[unclejed613]

Show me the..... crossover distortion with 100k=Aol and 33=Acl

run the sim again this time with the input signal at 100mV instead of 1000mV and tell me what the zero crossing on the output looks like