Beginner questions regarding amplification of different audio signal experiments.

Thread Starter

bignobody

Joined Jan 21, 2020
97
Hello! First time poster. We had a short lived "Electronics Club" at work that served as a pretty good introduction to electronics, but I am definitely still at the beginner level. When we joined the club we were asked what we hoped to get out of it. For me it was "I want to make noise and flash lights". That's still true and what I'm currently pursuing.

After building a couple of 556 Stepped Tone Generators, I've started experimenting with some other voltage controlled oscillators (CD40106 hex Schmitt trigger, mostly so far) and quickly noticed that the voltage signal they produce is much lower, and thus much quieter when connected to a speaker.

So, I need to build an audio amplifier. I found this article elsewhere on the Internet: http://www.circuitbasics.com/build-a-great-sounding-audio-amplifier-with-bass-boost-from-the-lm386/

I feel like I can build this at my current skill level, but given that I plan to use it to amplify various experiments I am uncertain about the best way to handle various Input Voltages.

The datasheet for the LM386 gives a MIN/MAX Input Voltage of -0.4 to 0.4 V, so I need to be sure the signal doesn't go outside of this range.

Should I be trying to do this on the amplifier side? (Maybe with a Zener diode or something like that?) Or should I just be ensuring any experimental circuits I plan connecting to it don't go out of this range by adding a Voltage Divider before the signal out? Or something else? I did try searching these forums but wasn't really successful at finding any relevant answers.

Thanks for reading! Any feedback is appreciated.
 

dl324

Joined Mar 30, 2015
16,846
Should I be trying to do this on the amplifier side? (Maybe with a Zener diode or something like that?) Or should I just be ensuring any experimental circuits I plan connecting to it don't go out of this range by adding a Voltage Divider before the signal out? Or something else? I did try searching these forums but wasn't really successful at finding any relevant answers.
If the voltage range of the input signal is known, you can use a voltage divider on the input of the amplifier. If it's variable, you can use schottky diodes to clamp the input to the supply rails.
 

Thread Starter

bignobody

Joined Jan 21, 2020
97
If the voltage range of the input signal is known, you can use a voltage divider on the input of the amplifier.
Thanks for your feedback, Dennis! The intent is to be able to amplify whatever audio signal circuits I happen to be experimenting with both now and in the near future, so it's best to assume the signal voltage will vary.

If it's variable, you can use schottky diodes to clamp the input to the supply rails.
Maybe I'm confusing my terms, but I wouldn't want to clamp to the supply rail (9V), rather I'd want to clamp to +/- 0.4V.

My next thought would be to add a variable voltage divider on the input, that way I could adjust the range as needed... But looking at the amplifier schematic again, that's effectively how the volume potentiometer is set up so maybe I'm over-thinking this whole issue...
 

dl324

Joined Mar 30, 2015
16,846
Maybe I'm confusing my terms, but I wouldn't want to clamp to the supply rail (9V), rather I'd want to clamp to +/- 0.4V.
My bad. You can put anti-parallel schottky diodes connected between the input and ground.

1580320024396.png
R1 is to limit the current from the signal source when the signal is being clamped.

Note that the forward voltage for schottky diodes is less than 0.4V (closer to 0.3V).
 

Thread Starter

bignobody

Joined Jan 21, 2020
97
Awesome, that's clear. I'm pretty sure I've got a couple of schottky diodes waiting in my component collection, I'll just need to check what their values are to make sure they'll do the job.

Thanks again!
 

Audioguru again

Joined Oct 21, 2019
6,674
The gain of an LM386 is usually 20 times and with a 9V supply its output is a max of 3V peak. Then an input that is more than only 0.15V peak will cause its output to have severe distortion. Turn down the volume control so that there is no severe distortion.
Do not feed it a very high level from another amplifier's output.

The LM386 produces a loudness about the same as a cheap clock radio. Is that enough?
A volume control is supposed to be logarithmic, not linear so that it has the same smooth huge range from loud to very soft as our hearing.
The tutorial you posted copied the datasheet of the LM386 with an error. C6 is shorted. C6 is supposed to be connected between the amplifier's inputs to prevent it from amplifying a local AM radio station's very strong radio frequency that is next door.
 

Thread Starter

bignobody

Joined Jan 21, 2020
97
The tutorial you posted copied the datasheet of the LM386 with an error. C6 is shorted. C6 is supposed to be connected between the amplifier's inputs to prevent it from amplifying a local AM radio station's very strong radio frequency that is next door.
First of all, thank you especially for this. I will be sure to make this correction in my build.

The gain of an LM386 is usually 20 times and with a 9V supply its output is a max of 3V peak. Then an input that is more than only 0.15V peak will cause its output to have severe distortion. Turn down the volume control so that there is no severe distortion.
Do not feed it a very high level from another amplifier's output.
If I understand correctly, for my experiments that produce low voltage signals (say, 0.1 - 0.5 V) I can just keep the amplifier volume low, but in cases where the source signal is more powerful (like 5-7 V from a 556 oscillator) I should do something to lower the signal before I send it across the connection to the amplifier.

The LM386 produces a loudness about the same as a cheap clock radio. Is that enough?
For now, I'm only beginning this journey :)
 

BobTPH

Joined Jun 5, 2013
8,813
I would put two inputs on your amp, one that passes the signal directly to the volume pot, amd the other tha passes it through a resistor 9X that of the volume control to achieve a 10 to one reduction.

Bob
 

Thread Starter

bignobody

Joined Jan 21, 2020
97
I would put two inputs on your amp, one that passes the signal directly to the volume pot, amd the other tha passes it through a resistor 9X that of the volume control to achieve a 10 to one reduction.
Bob
I really like this idea. Thanks very much, BobTPH!
 

AnalogKid

Joined Aug 1, 2013
10,987
The schematic you have chosen has an input volume control that can attenuate any signal input down to zero volts. If you start each experiment with the volume control at zero and then increase it slowly, you should be able to handle different input levels without needing clipping diodes.

ak
 

Thread Starter

bignobody

Joined Jan 21, 2020
97
Thanks for your input, ak! That's very clear. I guess I was overthinking this, but I did end up with a lot of valuable feedback!
 
Top