lm386 voltage divider?

Started by mmadscientist, April 09, 2006, 05:09:02 PM

Previous topic - Next topic

mmadscientist

Hello all....  I've been working on a very simple lm386 based distortion unit.  All was well untill I bought some new 9V batteries today.  When plugged in and playing, the battery starts getting hot and one of the clipping diodes also gets hot.  Also the signal begins to break up.  W/ the old battery in (which is running at about 6V) everything is cool.  The diode gets a little warm, but nothing serious and no battery heating.  The question is: Should I set a voltage divider to bring the V+ down to 6V from 9?  Has this happened to anyone else?  The circuit is similar to the nut-drive except simpler.  A electro and resistor connecting 1 and 8, another cap between 5 and 4, and two clipping diode from 5 to ground.   

Peter Snowberg

Your clipping diodes should never get warm.

I'm guessing you don't have a coupling capacitor between the 386 output and some clipping diodes like a Distortion+?

Nothing should be getting hot or warm at all. What schematic are you using?
Eschew paradigm obfuscation

mmadscientist

Thanks for the reply.  I have a cap from5 to 4, which is from the output to grd.  After that comes the diodes.  Is that what you mean?  The schematic this originates from is the nut-drive, found in the layout section.  See if this works:
http://aronnelson.com/gallery/album02/nut_drive

jrem

um, the LM386 spec sheet I have states it's an audio amplifier . . . ?   and that pin 5 is the output, and pin4 is ground, so you're cap and diodes are from output to ground? 

I'm probably missing something here . . .

jrem

#4
well looking at that layout I would check to make sure you're B+ isn't on pin 6, and all of your battery current is sinking through those diodes.  That would heat a few things up quickly.

<edit>  I mean pin 5 . . .    duh.

Peter Snowberg

Unfortunately those are not all good circuits in that folder.

Look at a datasheet for the 386 (just Google LM386 datasheet) and you will see the cap in series with pin 5. That is required to stop DC from flowing through the output. Add a cap of 10uF or more in series with pin 5 (+ side pointing toward pin 5) and that will make things work. I would also add a 10uF cap in place of the wire shorting out pins 1 and 8, and I would eliminate the "Gain" pot from the input as it's fairly useless. See the data sheet for proper 386 usage.
Eschew paradigm obfuscation

mmadscientist

Thanks for the reply.  Pin 6 is Vs and where the V+ is supposed to go.  The cap and diode arrangemnent is the same as the nut-drive except there is one set, not two.  It is actually the same schem besides a few changes:  I moved the 100uF electro from pin 7 to pins 1 and 8 along w/ a 220k resistor.  As of right now there are no pots.  The 150pF cap is now a 100pF and there are 2 diodes, not 4.  

mmadscientist

Thanks!!!!!  I'm going to check that out....  I posted that reply before seeing yours, pardon the jumble.  I wondered why that 150pF cap was going from out to ground, but I'm still learning and especially when it comes to chips just following schems, not necessarily understanding. 

mmadscientist

That did it!  Now the volume is in the red, but I'm sure I can work that out in the FAQ.  Thanks again!

Peter Snowberg

Eschew paradigm obfuscation