Class AB amp help.

[Lightium]

I was following this:

Class AB Power Amplifiers

Amplifiers, explained with the minimum of maths. Amplifier design, Amplifier Classes A to H, NFB, Circuits, Power Amplifiers, Op amps.

www.learnabout-electronics.org

got this:

It's not running as expected.

 

[rjenkinsgb]

That design has some strange combinations of parts & values.

R5 is 47 Ohms, with ideally 5.4V across it idle or average, so just over 100mA standing current in the drive circuit. The average power in the driver transistor would be less than 1W at that, so why a 3055 which is 15A and low gain?

No adjustable bias means it's probably class B rather than AB, but for an oscillator that probably does not matter.

There is no bootstrap on the driver/bias chain so it's not going to be very linear, the positive peaks will distort or clip easily.

And worst thing, R9 will totally mess up the bias and dc feedback on the driver, Q1. It should have a capacitor in there.

Some time back I did a series of sequential mods to improve the generic internet design people keep using; it's supposed to be part of a youtube tutorial but I've not got all the actual amps built to match yet, so I've not completed the video..

Stage 0 is the original, then with bootstrap, adjustable bias & midpoint, higher gain power stage then differential input. All the transistor types are generic for the drawings.

 

[Lightium]

Wow, thanks. I'll digest this and try again.

 

[Lightium]

It seems I am driving Q1 too hard. I am always doing that

 

[Nigel Goodwin]

It seems I am driving Q1 too hard. I am always doing that

Most of your component values seem completely random and way outside of any sensible values - it seems a common issue with people who play with simulators.

 

[Papabravo]

Most of your component values seem completely random and way outside of any sensible values - it seems a common issue with people who play with simulators.

It commonly takes longer for people who don't play with simulators to achieve the understanding they desire.

 

[ronsimpson]

Increased R5 by 10:1 to get more gain.
Increased R9 to reduce gain. (sounds funny)
Increased R1 to bias the output near 6V.
Added R10 because the output was too large. Might have changed R6 with little effect.
Changed Q1, Q2, Q3 because I did not have the right parts.

 

[Lightium]

Here's my bjt's:

 

[Lightium]

The way I am going with it is this:

Still not right though.

 

[ronsimpson]

The way I am going with it is this:

Increased R5 by 10:1 to get more gain.
Increased R9 to reduce gain. (sounds funny)
Increased R1 to bias the output near 6V.
Added R10 because the output was too large. Might have changed R6 with little effect.

 

[Lightium]

Oops, L1 needs to be 900uH not 100uH.

 

[audioguru]

The amplifier was not biased properly.