LM386 Low Power Audio Amplifier IC question

[Brad Davy]

I recently built a low power audio amplifier based on the LM386 IC to boost
the output of my Palm Zire 71 MP3 player. I followed the standard circuit
in the datasheet for a gain of 20, but when I tried it out, there was a
tremendous amount of noise in the circuit. Adding a .1 microfarad
capacitor between each of the audio inputs and ground eliminated this
noise, and the amplifier works beautifully. Can someone tell me why adding
the small capacitors between the input and ground eliminated the noise?

Thanks

 

[Frank Pickens]

I recently built a low power audio amplifier based on the LM386 IC to boost
the output of my Palm Zire 71 MP3 player. I followed the standard circuit
in the datasheet for a gain of 20, but when I tried it out, there was a
tremendous amount of noise in the circuit. Adding a .1 microfarad
capacitor between each of the audio inputs and ground eliminated this
noise, and the amplifier works beautifully. Can someone tell me why adding
the small capacitors between the input and ground eliminated the noise?

Thanks

Did the noise sound like a 60 cycle hum?

 

[mikem]

I ran into something like this when building a stereo amp to boost the
output of a Sony CD player so that it could drive my aircraft
headphones (hiZ). I discovered that the "common" side of the headphone
jack on the Sony was tied internally to the tap between the two 1.5V
batteries, instead of being referenced to the ground side of the
external power DC power jack.

If I tried to run the Sony on external DC power, there was a
tremendous amount of "hash" present as a common-mode signal between
the audio jack and the external ground. (Seemed to have to do with the
chopper that controls the motor that spins the CD).

The solution was to build an audio amp which has a true "differential"
input (common-mode-rejection ratio of ~90db).

MikeM

 

[Brad Davy]

Brad Davy wrote:

Did the noise sound like a 60 cycle hum?

No, it was definitely not 60 cycle hum. It was more like a high pitched
squeal. My theory is squeal. My theory is it was coming from the common
ground from the Zire 71 (processor operates 71 (processor operates at
144mHz). I don't have an oscilloscope to track it down, so I was just it
down, so I was just playing. Had the capacitors, so I tried it, and it
worked. Note that the worked. Note that the amplifier is powered by a 9v
battery, and everything is tied to a common gr is tied to a common ground,
including the audio output from the palm.

 

[Walter Harley]

No, it was definitely not 60 cycle hum. It was more like a high pitched
squeal. My theory is squeal. My theory is it was coming from the common
ground from the Zire 71 (processor operates 71 (processor operates at
144mHz). I don't have an oscilloscope to track it down, so I was just it
down, so I was just playing. Had the capacitors, so I tried it, and it
worked. Note that the worked. Note that the amplifier is powered by a 9v
battery, and everything is tied to a common gr is tied to a common ground,
including the audio output from the palm.

The LM386 is notorious for being unstable and very sensitive to supply
bypassing. Make sure the supply pin is bypassed to ground with a
high-quality aluminum electrolytic (or a lower-quality one paralleled with a
0.1uF ceramic), close to the chip.

By the way, your post contains some very odd repetitions. Read it closely -
perhaps I should say "Read it cl read it closely."

 

[Dave VanHorn]

The LM386 is notorious for being unstable and very sensitive to supply
bypassing. Make sure the supply pin is bypassed to ground with a
high-quality aluminum electrolytic (or a lower-quality one paralleled with a
0.1uF ceramic), close to the chip.

The original app notes have a series RC on the output pin, to ground, which
is a very good idea.

 

[Michael]

No, it was definitely not 60 cycle hum. It was more like a high pitched
squeal. My theory is squeal. My theory is it was coming from the common
ground from the Zire 71 (processor operates 71 (processor operates at
144mHz). I don't have an oscilloscope to track it down, so I was just it
down, so I was just playing. Had the capacitors, so I tried it, and it
worked. Note that the worked. Note that the amplifier is powered by a 9v
battery, and everything is tied to a common gr is tied to a common ground,
including the audio output from the palm.

Maybe due to grounding, try grounding the speaker directly to the battery.