Audio Amplifier Help

Audioguru

Joined Dec 20, 2007
11,248
I am glad that you ended with a fairly good design (except the 741 opamp is 43 years old and has lousy spec's).
I am also glad that you made your Multisim program work properly.
 

EarlAnderson

Joined Nov 13, 2011
166
well if you get a dramatic power drop whenever you change the frequency to anything above 1kHz, it's probably the op amp. some op amps can't process frequencies over a certian frequency. i may be wrong though. would you mind posting a schematic.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Well like I said, I was using the LM741CN OpAmp in my design...it's probably good until 1 MHz I believe but I'm not sure...
 

Audioguru

Joined Dec 20, 2007
11,248
The datasheet for the lousy old 741 opamp shows that its output has slew rate limiting above only 9kHz. Then its output level drops and a signal turns into a triangle-wave (lots of distortion) because the output cannot slew fast enough.
Modern opamps like the MC34071 go fine to 100kHz.

The max open-loop voltage gain of a 741 opamp at 10kHz is only 100. At 1MHz it has no gain like a distorted piece of wire.
An MC34071 has a max open-loop voltage gain of 500 at 10kHz.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
By the way, that MC34071 you were talking about, when I replaced my LM741CN's with those my amplifier circuit didn't work...like I was getting zero gain out of the opamps...what's wrong? Is there a certain frequency or minimum input voltage required to use them or something?
 

Audioguru

Joined Dec 20, 2007
11,248
By the way, that MC34071 you were talking about, when I replaced my LM741CN's with those my amplifier circuit didn't work...like I was getting zero gain out of the opamps...what's wrong? Is there a certain frequency or minimum input voltage required to use them or something?
What didn't work? Your lousy MultiSIM program? it doesn't have a model for an MC34071 opamp.
Since you don't show your latest schematic then we might never know.

The MC34071 is an excellent inexpensive opamp. It has high gain, wide bandwidth, works from a supply as low as 3V or as high as 44V and its inputs work as low as its negative supply voltage which can be 0V (a single polarity supply). Single, dual and quad MC3407x opamps are available.
But it might oscillate at a high frequency if you build a circuit with it on a breadboard which has high capacitance-coupling between contact strips and wires.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Oh I'm not talking about simulating it in Multisim. The problem is, when I used the MC34071 in Multisim, I was getting a higher gain than the rails. And when I actually built my circuit on the breadboard and I replaced the LM741CN's with the MC34071's, I had no output at any of the opamps or the amplifier as a whole. Why do you think it was so? Like, my circuit on breadboard works with LM741 (I used speakers to test) but it doesn't work when I replaced them with MC's. How come you know?
 

Audioguru

Joined Dec 20, 2007
11,248
The problem is, when I used the MC34071 in Multisim, I was getting a higher gain than the rails.
No you didn't. You had a higher output level than the rails.

And when I actually built my circuit on the breadboard and I replaced the LM741CN's with the MC34071's, I had no output at any of the opamps or the amplifier as a whole. Why do you think it was so?
If you measure the DC voltage at the output of the opamps and at the output of the amplifier then you will know exactly what is wrong.

Like, my circuit on breadboard works with LM741 (I used speakers to test) but it doesn't work when I replaced them with MC's.
It might not work with a 43 years old 741 opamp because many of them do not work with a total supply as low as 10V. But it should work with a newer and better MC34071 opamp because all of them work when their total supply is from 3V to 44V.

It is too bad that you are afraid to post your latest very simple schematic.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Oh my schematic is pretty basic, here it is attached. I reduced the gain on the lousy 741's to a mere 27x fold so I get about 5.6V p-p so there'll be less distortion and stuff.

It works, I built this on my breadboard and I connected my headphone port using an auxiliary cable to my amplifier and the output of my amplifier to a speaker and I can hear the music loud and nice but it's a bit distorted because it's a very basic design.

My problem is that I wanted to increase the voltage gain by using the MC34072P (I could only find this model at the store) and I was getting zero output (there was no sound coming out of the speaker). I was wondering why...
 

Attachments

Audioguru

Joined Dec 20, 2007
11,248
I wanted to increase the voltage gain by using the MC34072P (I could only find this model at the store) and I was getting zero output (there was no sound coming out of the speaker). I was wondering why...
Because you didn't look at the datasheet for the MC34072 to see that its pins are completely different from a 741 opamp because it has TWO opamps inside. An MC34074 has FOUR opamps inside.
Since you connected its pins wrong then it might be destroyed.

Your first opamp can have an AC gain as high as 391 with the newer opamp and it will still have good high frequency response to 20kHz. The gain of the first opamp does not affect the distortion from the amplifier.

Your circuit wrongly biases its input at 0V through the input signal source and its DC gain is the same as its AC gain. If the signal source has a coupling capacitor (so it does not pass 0V to your circuit) or if it has DC then your amplifier will not work. So you should use a 100k resistor to ground to bias its input at 0V and add an input coupling capacitor as I show.

I added a capacitor in series with R1 to prevent the opamp from having DC gain so that it does not amplify its input offset voltage. I increased the values of R1 and R10 so that this capacitor can be a fairly low value non-polarized capacitor and still pass frequencies as low as 49Hz.

Your circuit does not have its output transistors biased in class-AB like all audio amplifiers so it has horrible crossover distortion.
 

Attachments

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Oh shoot! Yeah the pins are completely different! I thought a dual opamp would be larger cause we have dual 741's and they're basically twice as long as a single. Damn, there goes my two MC's :(

And I added your modifications to my simulation but absolutely nothing has changed....is there supposed to be a change in my output and distortion with the added input coupling capacitor and such?
 

Audioguru

Joined Dec 20, 2007
11,248
We don't know if your music source has DC continuity to 0V or if it has an output coupling capacitor which will block biasing to 0V for the input of your first opamp.
We also don't know if your music source has DC on its output which your amplifier will amplify and get saturated.
That is why I added the two coupling capacitors which block DC.

Your unbiased class-B output transistors have high crossover distortion. Below clipping, a class-AB output stage will produce very low distortion.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Oh I didn't put those input coupling caps on my breadboard. I just wanted to see the effect in Multisim and it didn't show any visible results of them helping...but ofcourse, I trust you that they do something useful like block DC ...
 

Audioguru

Joined Dec 20, 2007
11,248
Multisim probably assumes that its signal generator has DC continuity to 0V, then the 100k resistor I added at the input of the first opamp will not make any difference (when there is no input coupling capacitor).

Multisim probably ignores the input offset voltage of an opamp so the capacitor to 0V I added in series with the 10k resistor also will not make any difference.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Oh ok perfect :) I'm going to go grab some new MC opamps either at SAYAL or Active Surplus...damn its far from here :S I wish I knew they were dual opamps because they're dead for sure now I just tested them and they're output is so distorted and "shaky".

In the meantime, AudioGuru, I need you to please recommend me some part #'s for this class AB amplifier I made because I'm not satisfied with stupid LM741 or MJE172/182 transistors...I heard there were better ones ;)

So I need:

-New OpAmp (MC34072P)
-New resistors with 1% tolerance
-New power BJTs (2N3055 any good??)
-Capacitor for load to reduce distortion (ceramic best kind for noise/distortion reduction??)
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
Your amplifier will have a fairly low output power because its power supply voltage totals only 10V. So the peak output current into an 8 ohm speaker is about 375mA or 750mA into 4 ohms.
Medium-power transistors like TIP31 and TIP32 will be fine.

Audio amplifiers do not use a capacitor parallel to the load because it is a dead short at high frequencies. I don't know why Multisim seems to need the capacitor.
 

Thread Starter

zero_coke

Joined Apr 22, 2009
294
Well see thats the issue I have. When I don't have that capacitor at the load, my distortion at output continuously rises for some reason...I have no idea why it's doing that but all the sudden when I put the capacitor it keeps the distortion steady at like 1%....when I don't have it goes from 1% to like 100% steadily.
 

Audioguru

Joined Dec 20, 2007
11,248
I assume you are simulating your amplifier in Multisim. I have never used Multisim. Maybe it gets confused.

Maybe you built the amplifier on a breadboard and the capacitances between all its contacts and wires is causing your circuit to oscillate at a high frequency that sounds bad like distortion. Then a capacitor at the output to ground might stop the oscillation.
 
Top