# Low hz audio amplifier problem

#### jaydnul

Joined Apr 2, 2015
175
I am making a low frequency amplifier and have it complete up until the stage I pictured:

The problem is something to do with the capacitor and speaker. I removed the cap and speaker and attached a pot from the emitter straight to -9V. At first it looked fine, emitter signal looked the same as the base signal, but as I was rolling the potentiometer resistance down I noticed at about 2k the signal got all weird and distorted. Here is a picture of the signal going into the base of the transistor:

And here's a pic of the signal coming out of the emitter when the cap and speaker are connected just like in the first picture:

The speaker is 4 ohms and the sine wave is about 5Vpp. Also the power supply and the BJT are rated at 2A.

Any ideas of where I'm going wrong? Thanks and sorry for the big pictures .

#### Kermit2

Joined Feb 5, 2010
4,162
A question for you.

Does a capacitor block DC current flow?

#### jaydnul

Joined Apr 2, 2015
175
Ya it blocks DC, why?

#### #12

Joined Nov 30, 2010
18,224
At what frequency is Xc equal to 4 ohms?
Hint: Almost 40 KHz.
Is that what you call, "Low frequency"?

Wait. Is that a sloppy uf instead of the nf it looks like?
That would be more like 20 Hz.

So the transistor takes the capacitor voltage up to the peak of the sine wave and where does that current go during the lesser voltages to drive the speaker?

#### jaydnul

Joined Apr 2, 2015
175
Sorry, ya that's a 2000uF (2mF) cap. I tried putting a 1k resistor in parallel with the cap/speaker, but got the same results.

Last edited:

#### #12

Joined Nov 30, 2010
18,224
So the transistor takes the capacitor voltage up to the peak of the sine wave and where does that current go during the lesser voltages to drive the speaker? There is no escape route for the current! The voltage on the capacitor can't go below the peak voltage any faster than the parallel resistor bleeds it off.

This is why push-pull amplifiers were invented.

#### jaydnul

Joined Apr 2, 2015
175
I see. the transistor is going to block the current because of the reverse biased BE diode, got it.

I would use a push-pull but I'm trying to avoid pulling any current through ground because it is a virtual ground made with a resistor divider into an op amp; the op amp cant handle that much current. I'll keep trying stuff, thanks for the help!!

#### Bordodynov

Joined May 20, 2015
3,180
Your system is not complete. The current positive half provides NPN transistors. What makes the current negative half? You still needed PNP Transistor.

#### tophericks

Joined Jan 29, 2016
5
I tried putting a 1k resistor in parallel with the cap

#### Bordodynov

Joined May 20, 2015
3,180
See:

#### dannyf

Joined Sep 13, 2015
2,197
Put a low value resistor from the emitter to ground.

Your amp is very inefficient and may not function in its intended application.

#### Bordodynov

Joined May 20, 2015
3,180
There is practically no difference.
Once a long time used amplifiers Class B. If you need a high-quality amplifier, it certainly is not as my amp. Source amplifier worthless. You can put a resistor instead of choke.

Last edited:

#### #12

Joined Nov 30, 2010
18,224
Speaking only about one transistor amplifiers:

If you want to do this as a class A amplifier (one transistor), the method is to use a resistor from the emitter to ground then the capacitor goes sideways from the emitter to the speaker and the speaker is in parallel with the first resistor. The resistor uses up most of the current and that is why this is called, "inefficient".

If you have a dual supply voltage and reference the (+) input of the op-amp to zero volts by using an input capacitor and a resistor to zeroV, the output resistor and capacitor are not necessary because the speaker provides the current path to -9V and the output wave is centered on zero. With the entire power section connected from +9 to -9 the only load on the zero voltage point is the input resistor on the op-amp, and that might be in the nanoamp range. The problem here is the speaker drawing over 4 amps of DC at zero output. Again, horribly inefficient.

By now you should be realizing that the only way to make this work in class A is to use a transformer to lower the DC current and send the AC to the speaker or use a higher resistance speaker, like 16 ohms or 32 ohms, and it's still going to waste power like a class A amplifier. If you want anything like efficiency, you are going to have to go with a push-pull arrangement.

#### jaydnul

Joined Apr 2, 2015
175
This might be semantics but why did you keep the capacitor in this circuit? Now that it's a push/pull, there wont be any DC bias, correct?

#### Kermit2

Joined Feb 5, 2010
4,162
Why do you want a cap there anyway.

It has some distortion from the zerocrossing portion of the input being muted, but it appears to work fine for a four ohm load and even better with 16 ohm at 20 hertz with no cap at all

#### Kermit2

Joined Feb 5, 2010
4,162
of course those transistors better have a good heat sink. They are each dissipating a half watt of heat.

1 volt input and about 1 or 2 mA of current are needed. But ditch the 9 volt transistor batteries, they will NOT deliver the current required to drive a 4 ohm speaker and if they do they will only last about a single minute.