Need help with basic LM386 Audio Amp circuit

[Doctor Jones]

Hello there folks, made an account here just for this question. I should clarify that while I made plenty of analog circuits during my undergrad, but this is my first audio amp that I've ever built. My problem is that I cannot for the life of me get this thing to amplify a given input signal. Below is a copy of my circuit diagram:

I have a 555 Timer generating a tone, passing it into an LM386 audio amp, and outputting that result to a speaker, all driven from a single 9v battery. However, when I plug my speaker directly to the 555 Tone Generator output (IE directly after C2), it's as loud as the output from the LM386 amp.

I have tried using at least 8 different variations of the basic LM386 amplifier circuit **broken link removed** (gain = 20), but no matter what I do, I cannot generate a sound louder than what is being generated by the 555 Tone Generator.

As for things I've tried:
- Completely replace all components
- Verify components are all working
- separate power sources per application (knew this wouldn't work but tried it anyways)
- Try different layouts of amp design

And yet, all my efforts have failed. As far as my reading and research has shown, this should work, and yet the evidence is contrary. Any advice, besides dropping the LM386 for a superior opamp? If my understanding of audio electronics is goofed, please correct me, because this is driving me nuts.

 

[betwixt]

That's probably because the 555 can produce more current into the loudspeaker than the LM386. The difference is the LM386 is a linear(ish) device so it can amplify small signals to full volume, the 555 produces a 9V waveform which is already at the maximum the supply line can possibly provide through the amplifier.

Incidentally, your volume control is wired wrongly, the signal should be fed to the top of the pot, the bottom should be ground and the input to the LM386 taken from the wiper.

Brian.

 

[dick_freebird]

Given that they share the same supply rails and the 555 is a
bang-bang (rail-rail - 2Vbe-Vce(sat)) output, why would you
expect the 386 to provide any higher amplitude?

C4 and an 8-ohm speaker gives a corner frequency of 625Hz.
What's your tone of interest? From the datasheet curves I
make it out that 3.3K, 1uF makes sub-kHz frequency but too
crude to say whether you're on the right side of it. And have
you selected a -nonpolar- blocking cap or at least ensured
that the polarity as-installed does not go reverse?

Now here's another question. Your signal source is a square
wave. The audio amp is meant for linear amplification but
with only two states, why not put something like a power
MOSFET driver (w/ higher supply) behind the 555 and get
amps of output drive (subject to thermal management)?

LM386 datasheet says 325mW power, but THD goes through
the roof at about 200mW per the chart. That would be (into
8 ohms) about 158mA RMS. Which is not that different than
the 200mA rated current of the 555 if you look at the peak
instead. So again, the 386 can't make any contribution really.

There exist higher-Z speakers (like I've seen 32ohm) which
may offer higher SPL per watt input, and be a better companion
to the little ol' 386. There are also higher power audio amp
ICs (LM380, TI.com has 63 Class AB "speaker amplifier" products
to pore over.

 

[Audioguru]

Your speaker is not connected to ground (0V) on your schematic and an important capacitor is missing across the 9V battery.

The datasheet for the 555 shows that its output voltage swing at the output of your C2 is about 3.5V peak into your pot.
The datasheet for the LM386 says its maximum allowed input is 0.4V peak so you are feeding an input to the LM386 that is 8.75 times too high which is probably destroying it. Another amplifier that survives an input that is the 3.5V peak from your 555 IC and with a gain of 20 then its output would be 3.5V x 20= 70V peak and this new amplifier would need a power supply that is almost 146V(!) and the output power with a squarewave will be 616 Watts(!).

20 times the voltage gain produces 100 times more output power.

The datasheet for the LM386 shows an output of 3V peak into an 8 ohms speaker when its supply voltage is 9V. Then the output power is only 0.56 Watts with a sinewave or a 1.1W buzz with the squarewave buzz. A little 9V battery will quickly die.

Your speaker's impedance is not shown but if it is 8 ohms then the output of the 555 IC with the low value for C2 into the 8 ohms speaker produces a low level buzz.

 

[Doctor Jones]

That's probably because the 555 can produce more current into the loudspeaker than the LM386. The difference is the LM386 is a linear(ish) device so it can amplify small signals to full volume, the 555 produces a 9V waveform which is already at the maximum the supply line can possibly provide through the amplifier.

Incidentally, your volume control is wired wrongly, the signal should be fed to the top of the pot, the bottom should be ground and the input to the LM386 taken from the wiper.

Brian.

That would definately explain why the outputs are audibly identical. As for the diagram, that one's getting scrapped for a redesign, but I'll remember it for next time, cheers.

Given that they share the same supply rails and the 555 is a
bang-bang (rail-rail - 2Vbe-Vce(sat)) output, why would you
expect the 386 to provide any higher amplitude?

In all honesty I was sceptical when I was first designing the circuit based on other designs I'd seen around the net, but my ignorance got the better of me here.

Now here's another question. Your signal source is a square
wave. The audio amp is meant for linear amplification but
with only two states, why not put something like a power
MOSFET driver (w/ higher supply) behind the 555 and get
amps of output drive (subject to thermal management)?

LM386 datasheet says 325mW power, but THD goes through
the roof at about 200mW per the chart. That would be (into
8 ohms) about 158mA RMS. Which is not that different than
the 200mA rated current of the 555 if you look at the peak
instead. So again, the 386 can't make any contribution really.

There exist higher-Z speakers (like I've seen 32ohm) which
may offer higher SPL per watt input, and be a better companion
to the little ol' 386. There are also higher power audio amp
ICs (LM380, TI.com has 63 Class AB "speaker amplifier" products
to pore over.

Given this information, yes I'll have to look into higher power audio amp ICs. My end goal is to still be able to run this circuit off of a 9v battery (design considerations for the application), so min/maxing the capabilities of the parts is something I'll have to put more effort into researching.

Your speaker is not connected to ground (0V) on your schematic and an important capacitor is missing across the 9V battery.

I've thus determined I'm really bad at drawing circuits

The datasheet for the 555 shows that its output voltage swing at the output of your C2 is about 3.5V peak into your pot.
The datasheet for the LM386 says its maximum allowed input is 0.4V peak so you are feeding an input to the LM386 that is 8.75 times too high which is probably destroying it. Another amplifier that survives an input that is the 3.5V peak from your 555 IC and with a gain of 20 then its output would be 3.5V x 20= 70V peak and this new amplifier would need a power supply that is almost 146V(!) and the output power with a squarewave will be 616 Watts(!).

20 times the voltage gain produces 100 times more output power.

The datasheet for the LM386 shows an output of 3V peak into an 8 ohms speaker when its supply voltage is 9V. Then the output power is only 0.56 Watts with a sinewave or a 1.1W buzz with the squarewave buzz. A little 9V battery will quickly die.

Your speaker's impedance is not shown but if it is 8 ohms then the output of the 555 IC with the low value for C2 into the 8 ohms speaker produces a low level buzz.

This just reinforces that the 386 is going to need to be swapped out, or my signal generation is going to need a rework. Probably both. Regardless, I appreciate the responses, I'll mull over this and do considerably more research before diving back into the project.

 

[Audioguru]

LM386 datasheet says 325mW power when clipping like crazy (10% distortion) into 8 ohms when powered from only 6V. The datasheet shows 700mW with 10% distortion. With low distortion a graph shows an output that is 350mW with a 9V supply.

The output power is less when the speaker impedance is higher because then the speaker current is less.

--- Updated Jan 28, 2022 ---

A little 9V battery is too weak to make more sound for a short duration than a cheap clock radio.
You need a higher battery voltage with a higher battery current rating plus an amplifier that will produce the power you want into a speaker that survives that amount of power.