Total beginner Audio Amplifier (LM386)

[Rene]

I am trying to build an audio amplifier using the LM386, since I have no
idea what am I doing, I wired the chip as good as I could and it worked but
its very noise. I am certain that I am doing something wrong, especially
because I have no idea where all the grounds in the diagram should go.



Could you please look at the picture on the link below and tell me were I am
screwing up? Please look at the battery and speaker jack connections.



http://home.comcast.net/~rene413/LM386Con.GIF



Thank you.

 

[Joerg]

Hello Rene,

Some grounds appear to be going only to the input jack, others to the battery.
Connect all together at shortest possible path and see what happens. Best would
be to have a ground plane on your experimental board.

Regards,

Joerg.

 

[Tim Wescott]

Hello Rene,

Some grounds appear to be going only to the input jack, others to the battery.
Connect all together at shortest possible path and see what happens. Best would
be to have a ground plane on your experimental board.

Regards,

Joerg.

Yes, generally _all_ the ground symbols go to the same place, unless
they're different shaped or otherwise annotated (and sometimes they go
to the same place anyway).

 

[Jan Panteltje]

I am trying to build an audio amplifier using the LM386, since I have no
idea what am I doing, I wired the chip as good as I could and it worked but
its very noise. I am certain that I am doing something wrong, especially
because I have no idea where all the grounds in the diagram should go.



Could you please look at the picture on the link below and tell me were I am
screwing up? Please look at the battery and speaker jack connections.



http://home.comcast.net/~rene413/LM386Con.GIF

You absolutely need a big electrolitic cap, in aprallel with say a 100 nF
poly of ceramic on the supply line.
1000uF would be nice...
Also watch you input lines, things may be oscillating.
I have seen the input pot used in a different way, look up application notes.
Oh, I have this old National Audio/Radio handbook, nice book!
Here they have the LM 386, and that input you use has a low pass with 1 k
and 2200 pF! before the external lead and pot.
Also they use a small ferrite bead with 2 turns in parallel with a 47 Ohm
resistor in series with the pen 5, and then 5nF to ground...
So oscillation ! Do you have a sope?
JP

 

[Rene]

But if all the grounds go to the same place (say the battery ground) then I
would only be able to connect one of the wires coming out of my speaker jack
(The one that gois in the 10k pot), where would I connect the other one??
Please check out the new link:

http://home.comcast.net/~rene413/LM386Con2.GIF

 

[Tim Wescott]

But if all the grounds go to the same place (say the battery ground) then I
would only be able to connect one of the wires coming out of my speaker jack
(The one that gois in the 10k pot), where would I connect the other one??
Please check out the new link:

http://home.comcast.net/~rene413/LM386Con2.GIF

The wire with the '?' goes to the negative terminal of the battery, too.
When you do this your circuit is referenced to the computer ground.

BTW: you want to use the ground lead of the computer speaker jack. This
is the long part (not the tip) if it's a 1/8 inch audio jack, or the
outer part if it's an RCA-type coaxial.

 

[John Jardine]

Hello Rene,

Some grounds appear to be going only to the input jack, others to the battery.
Connect all together at shortest possible path and see what happens. Best would
be to have a ground plane on your experimental board.

Regards,

Joerg.

You need a hulking great electrolytic capacitor across the battery
terminals. Something like 470uF would be OK.
regards
john

 

[Seth Koster]

I am trying to build an audio amplifier using the LM386, since I have no
idea what am I doing, I wired the chip as good as I could and it worked but
its very noise. I am certain that I am doing something wrong, especially
because I have no idea where all the grounds in the diagram should go.



Could you please look at the picture on the link below and tell me were I am
screwing up? Please look at the battery and speaker jack connections.

The lm386 has, I believe, a 10%THD (total harmonic distortion),
this means it will be noisy no matter what, but the most obvious
problem to me is that the grounds are hooked up wrong. Take the
negative lead from your input jack (negative from the computer cord)
and hook that up to ALL grounds, then hook up ALL grounds to the
negative lead from your battery.

 

[Eamon Skelton]

Could you please look at the picture on the link below and tell me were I
am screwing up? Please look at the battery and speaker jack connections.
http://home.comcast.net/~rene413/LM386Con.GIF

The ground side of the computer audio jack and
the - (inverting) input of the LM386 should be
connected to the same ground as the rest of the
circuit.

I would also suggest that you put a 1 microF
capacitor in series with the input (from PC jack
to 10K pot), a 10microF capacitor from pin 7 to
ground and a power supply de-coupling capacitor
of several hundred microF from pin 6 to ground.

The LM386 data sheet is at:
http://www.national.com/ds/LM/LM386.pdf

I have made a few changes to your schematic.
Send an e-mail to nospam AT oceanfree.net
if you want me to send you a copy.
Replace " AT " with "@"

HTH.

E.S.

 

[N. Thornton]

I am trying to build an audio amplifier using the LM386, since I have no
idea what am I doing, I wired the chip as good as I could and it worked but
its very noise. I am certain that I am doing something wrong, especially
because I have no idea where all the grounds in the diagram should go.

http://home.comcast.net/~rene413/LM386Con.GIF

Theres 2 things missing there. Your ground connections are connected
in 2 sets, with no connection between them - they all need connecting
together. And I'd put a 1000uF capacitor across the 2 power lines.

Regards, NT

 

[Terry Pinnell]

In addition to (or reinforcing) other replies:

1. I had great difficulty eliminating noise in a rather similar
circuit using the LM380. Major contributor was indeed the
configuration of ground wiring. But sorting that was more trial/error
than following any fixed rule.

2. One complication you may have was that input and output sockets
were earthed via chassis. And in my case I had two *sections* of
chassis which were electrically connected only when screwed together,
unless I added a connection wire.

3. Large electrolytic across supply, and 100n across pins 2 to 6.

4. Extra ripple suppression with 10uF from pin 7 to ground.

5. Screened cable if input socket more than couple of inches from pot,
and from pot middle to 386. (And then experiment to find optimum: both
or one end of screen grounded? To where?)

 

[Animesh Maurya]

Since discussion is going all about LM386, it's a nice place for me to
start.

Is Slew Rate of an Op-Amp a good parameter, to be taken into
consideration while designing an audio amplifier.

If yes, why don't the datasheet of LM386 carry its Slew rate.

Or maybe LM386 is solely designed for audio power amplification, so
designers don't have to worry lot about its SR.

Thanks a lot

Best regards,
Animesh Maurya

 

[Tom Bruhns]

Oh, my. Well, ALL the grounds must be connected together, and your
diagram shows some connected to the 'puter and some to the battery,
etc. Best, lacking any other info, is to connect them all in a "star"
configuration: pick one point that you call ground, and run each
connection through as short a wire as is reasonable to that point.

Cheers,
Tom