My Power Amplifier Circuit

Thread Starter

KTF.1994

Joined Nov 1, 2019
2
hello.
this is my first post here. i am working on an audio amplifier for an electric guitar.
this is a simple power amplifier i did not include the per-amp stage here.
any suggestions, improvements and corrections are appreciated.
pic.png
 

crutschow

Joined Mar 14, 2008
34,281
Looks basically okay to me.

You might want to add a pot with R7 or R8 to allow adjustment of the output stage bias current.
Try a 500 ohm resistor in series with a 1k ohm pot.

If you want the maximum output closer to the supply rails, you might consider adding a bootstrap bias for the output stage.
 

AlbertHall

Joined Jun 4, 2014
12,345
As shown it has a gain of just one which means if you want a ten volt output, you will need a ten volt input. You could increase R9 to give it a little gain.
 

OBW0549

Joined Mar 2, 2015
3,566
this is my first post here. i am working on an audio amplifier for an electric guitar.
this is a simple power amplifier i did not include the per-amp stage here.
any suggestions, improvements and corrections are appreciated.
You may end up with a "long feedback loop" stability problem with this circuit, caused by the wide gain-bandwidth product of the TL082 together with the frequency characteristics of your output transistors Q1 - Q4. This could cause a high-frequency peak in the amplifier's frequency response or even oscillation.

One possible cure for this would be to connect a small capacitor (100 pF to 1000 pF) between the TL082's output and its inverting input to force a low-frequency rolloff at just above the audio range.

Hard to say whether you'll actually encounter this, but it's worth looking out for...
 

AnalogKid

Joined Aug 1, 2013
10,986
As shown it has a gain of just one which means if you want a ten volt output, you will need a ten volt input. You could increase R9 to give it a little gain.
Yes, but - since this is an inverting configuration, there is no low frequency pole to reduce the forward gain to unity at DC. The result is that any offset errors are amplified and appear as DC through the speaker voice coil. Without knowing what the V+, V-, and Vin values are we can't estimate the gain needed for a "full power" output at max input, but the error could be enough to matter.

Separate from that, there is a school of thought that prefers inverting opamp circuits for audio because the almost zero voltage swing at the input reduces harmonic distortion in the opamp input stage. But - another school of thought strongly prefers that the audio signal phase be preserved through the signal chain. That is, there should be either zero inverting stages throughout the system or an even number of them, so the net phase shift through the system is zero degrees (as opposed to 180 degrees). In theory, with a system like this, a positive pressure on the microphone diaphragm during the original recording propagates through to the speaker cone moving outward, creating positive pressure at the ear drum.

Disclaimer - I'm not a golden-eared loon, but I used to work with some.

ak
 

Papabravo

Joined Feb 24, 2006
21,159
Yes, but - since this is an inverting configuration, there is no low frequency pole to reduce the forward gain to unity at DC. The result is that any offset errors are amplified and appear as DC through the speaker voice coil. Without knowing what the V+, V-, and Vin values are we can't estimate the gain needed for a "full power" output at max input, but the error could be enough to matter.

Separate from that, there is a school of thought that prefers inverting opamp circuits for audio because the almost zero voltage swing at the input reduces harmonic distortion in the opamp input stage. But - another school of thought strongly prefers that the audio signal phase be preserved through the signal chain. That is, there should be either zero inverting stages throughout the system or an even number of them, so the net phase shift through the system is zero degrees (as opposed to 180 degrees). In theory, with a system like this, a positive pressure on the microphone diaphragm during the original recording propagates through to the speaker cone moving outward, creating positive pressure at the ear drum.

Disclaimer - I'm not a golden-eared loon, but I used to work with some.

ak
I worked with those guys too!
 

Thread Starter

KTF.1994

Joined Nov 1, 2019
2
thank you all. i start making modifications.
another question. what do you guys can tell me about the R1 and R2 resistors? all i know about it that they provide a discharge
path for the base-emitter capacitor of the Q2 and Q3 to decrease off time.
 

MrChips

Joined Oct 2, 2009
30,708
It is imperative to have good power supply decoupling. However I don't see any valid reason for overdoing it.
I would eliminate C6, C7, C8, C9.
Keep C4 and C5 close to IC1. Change C2, C3 to 10μF. Then feed IC1 power rails via 100Ω resistors off V+ and V-.

Add 100μF caps to V+ and V- power rails.

I am not sure why C1 is there.
 

crutschow

Joined Mar 14, 2008
34,281
In theory, with a system like this, a positive pressure on the microphone diaphragm during the original recording propagates through to the speaker cone moving outward, creating positive pressure at the ear drum.
That makes the dubious assumption that the phase is conserved through the whole electronics chain from microphone, through the recording medium, the reproductive mechanism, all the amplifiers, and finally to the speaker.
I'd say the chance of that is about 50-50. :)
 

AnalogKid

Joined Aug 1, 2013
10,986
That makes the dubious assumption that the phase is conserved through the whole electronics chain from microphone, through the recording medium, the reproductive mechanism, all the amplifiers, and finally to the speaker.
I'd say the chance of that is about 50-50.
Actually, it is well over 90% for professional recording studios, cutting/pressing labs, and both pro and consumer stereo gear. The standard is that each piece of equipment is non-inverting from in to out, so no matter how many processing stages / equipment strings / etc., the net phase change is always zero. If the phase of the mic preamp is correct, everything else is fine.

ak
 

ronsimpson

Joined Oct 7, 2019
2,988
the phase is conserved
In my opinion:
The speed of sound is about 343 meters per second 1,125 ft/s. C6 note is about 1khz. So an instrument playing a 1khz tone will do 360 phase shift every foot on the way to the microphone. Most likely every instrument, every singer, every thing making noise has a different distance/phase to the microphone. The base and treble controls can bump the phase a little. A three way speaker has a different distance from each driver to the person. Distance to left ear verses right causes phase shift. There are so many places for the phase to get reversed at one frequency but not another. How do you know phase?

When the left and right channel are not the same (in low frequency phase there is a real problem, but most people can't hear it.
non-inverting
Probably best. The amplifier may be used with a different amplifier in the same room. (low frequency cancelation)
 

DickCappels

Joined Aug 21, 2008
10,153
C1 is necessary in some high frequency video amplifiers otherwise there can be a dip in the frequency response within the passband. It makes intuitive sense to use C1 but with the high gain of the opamp in the loop is it probably not useful in this circuit

Since the output stage has a gain of 1 and the offset of a TL082 is small, the quiescent current through the speakers will also be small and not a worry.
 
Top