I breadboarded the audio amplifier in the circuit below. My O'scope indicates that it is oscillating on the positive half of the sine wave. Any ideas? I used the inductor symbol to represent the speaker. If anyone knows where I can find a LTspice model for a LM380 please share. Also what would be a reasonable model for a 2" speaker?

 

Do you have the power supply decoupled? Is the oscillation frequency low enough to be audible?

Post a schematic for those of us who might be using a device that doesn't have LTspice installed.

 

Hello,

Can you post a readable schematic?

Do you have the resistor/capacitor combination on the output of the LM380?



Bertus

 

Do you have the power supply decoupled? Is the oscillation frequency low enough to be audible?

Post a schematic for those of us who might be using a device that doesn't have LTspice installed.

Yes the power supply has a 220 uf cap which I moved to be physically closer to the opamp.

What is a good way to generate a jpeg of my schematic?

 

Yes the power supply has a 220 uf cap which I moved to be physically closer to the opamp.

You need a ceramic capacitor for good high frequency decoupling.

What is a good way to generate a jpeg of my schematic?

Depends. Are you using a schematic editor? Or drawing on paper?

 

Hello,

Can you post a readable schematic?

Do you have the resistor/capacitor combination on the output of the LM380?

View attachment 95534

Bertus

I will have to find a way to generate a jpeg or something that can be displayed easily.
and yes I do have the resistor / capacitor in place. Ironically the oscillation diminished
substantially when I disconnected it. This led to me dress the leads differently. Problem solved!
I never thought lead dress would be critical at 1KHz!

 

Breadboarding can introduce all sorts of coupling and parasitics.

 

I never thought lead dress would be critical at 1KHz!

Your signal may be at 1 KHz but the amplifier has gain at hundreds of kHz or more. This gain at high frequencies with some feedback can cause oscillations from the parasitics dl24 mentioned.

 

Hello,

Make a screenshot of the simulator you use and post it here.

Bertus

 

@jmoffat , Is it the LTSpice sim that is oscillating, or is it the real circuit?

Can you post the spice model you are using?

 

There is no symbol, but a symbol of the resistor can be used.

 

In post #3 by Bertus, there is a 2.7 ohm plus 0.1 uF snubber on the output. Do you have that installed? In 1979, I worked on a commercial product where the original engineer did not put in the snubber, and there were severe problems. The LM380 worked as an amplifier, but was very soft. My theory was it was putting out most of its power at the high frequencies of the ringing (oscillation). Once the 0.1 uF cap was installed, the output became normal.

 

As a rule of thumb, audio power amp chips, old and new, are notorious for breaking into oscillation. They are very sensitive to ground impedance (long wires/thin traces), decoupling, etc. Almost all have reference designs in the datasheet or app notes, and that board layout is not just recommended. Often it is the only thing that will keep the device under control.

ak

 

I made a symbol and created a spice model.

 

eetech, join please symbol element and test scheme.

 

eetech, join please symbol element and test scheme.

Hi Bordodynov

I'm not sure what you mean....please clarify.

 

Attach the files.

ak

 

eetech, join please symbol element and test scheme.

Hi

LTspice files attached.