Many errors in this simple circuit?

Thread Starter

Yabba Rays

Joined Oct 5, 2016

This is supposed to be a basic audio amp. How many errors here? No battery decoupling, DC going through the speaker and what's going on with Q2, Q3? Will that even work - or am I missing something?

Sorry, title should read 'errors' not 'effors' but can't find a way to correct that.

(Mods, please change heading if possible. Thanks)

Janet (Yabba Rays)
Last edited:


Joined Mar 14, 2008
Yes, it has DC current going through the speaker which is generally a no-no.

You are correct, Q2 and Q3 won't work as shown.
You either need a resistor from the collector of Q2 to V+, or Q3 needs to be a PNP.


Joined Jul 11, 2016
are most likely the NPN darlington drawn . . . has it happens -- coz the replacement of upper one with PNP enables a current path through the base of then PNP Q3 whitch eighter saturates the least or emtys the battery fast or reduces the terminal voltage or ruins the Q3 or/and Q2

i considered the collector feedback for the input stage Q1 - because the application runs on a battery - say 2x 1.5V that changes it's terminal votage from 2.9V down to say 2.3V during battery discharge-/service- time
i considered using different polarity transistor for Q1or reversing rails×polarity for Q23 darlington (both emitters go to speaker)

the circuit looks like drawn by instructions at bad quality phone call ???


Joined Jul 11, 2016
on a second thought i'm not so sure about what i just said
► it's the 3× bjt gain of 200³ = 8000000 -- eight million

i attempted something like this - an audio power amp -
and for a reason - i had to negative feedback - in principle -
Q2 and also Q3 e.g. both stages - to get un-distorted
near 1:1 voltage amplified signal ??? while not varying
the supply keeping it steady 3V


Joined Jul 11, 2016
ok, i can't design transistor wide bands (fq. × feed voltage), but since we don't have anything here , here what i attempted
Random_B2A-fp.png Random_B2Ab.png
Random_B2A.png Random_B2Ab.png
(just a blind shot not that i'd had anything speciffic in mind - and i didn't yest compute anything)
optimizing . . .
still hasn't computed anything . . .


Last edited:


Joined Oct 2, 2009
There is not a lot wrong with the circuit, only the drawing of Q3 is incorrect.

BC558 is a PNP transistor. The emitter should be on the +ve side of the battery. The collector is connected to the loudspeaker.

You can replace Q1 and Q2 (BC548) with 2N3904.
Q3 (BC558) may be replaced with 2N3906.
I would choose an 8Ω speaker (or higher), not a 4Ω speaker.

I would decrease R1 from 1M5 to about 560k-1M for less distortion.
Otherwise, the circuit works fine on 3V supply.