Need Help With New Distortion

Started by Box_Stuffer, July 01, 2023, 08:22:52 PM

Previous topic - Next topic

Box_Stuffer

So I am working on this LM386 based distortion that I have devised from a combination of several other schematics that I have seen, but didn't really like.

The Problem :
It sounds really great when I first plug it in, but after a couple of minutes it starts sputtering and gets raspier sounding and the level seems to undulate. It sounds better when the LED gain loop it cranked - it is most noticeable when it is turned down. If I unplug it and wait a while it seems to "reset".

I have looked at every LM386 schematic that I can find and I have noticed that pretty much all of the circuits that use the non-inverting input have the power directly to pin 6 and all of the circuits that use the inverting input have some sort of modifier in the power path - like resistors or capacitors or both. I have seen both with a 4148 diode in the power section. I have none of the above. I just put power straight in, but circuit is inverted. I tried a 10n cap from Pin 7 to ground, but that didn't help. I tried it one way that I saw where I ran the power through a 470ohm resistor and 220uf cap. It still fizzled out.

What can I do to stabilize this thing? It sounds really great while it works.







Matthew Sanford

Try a 10r from pin 5 to your cap to ground, a 100n cap from pin 7 to ground placed close to the chip, and a 100u cap from pin 6 to ground, like the Ruby amp, might help
"The only knowledge is knowing you know nothing" - that Sew Crates guy

Controlled Chaos Fx

duck_arse



still the cap is backwards.

show us the thing you have built, please.
" I will say no more "

Box_Stuffer

Quote from: duck_arse on July 02, 2023, 10:26:18 AM


still the cap is backwards.

show us the thing you have built, please.

I thought it looked weird too, but that is how the Indian guys in the youtube video that I watched had done it. The video was all in Hindi with no subtitles so I had to watch very closely and I'm sure that is how they attached it.


antonis

Quote from: Box_Stuffer on July 01, 2023, 08:22:52 PM
I tried it one way that I saw where I ran the power through a 470ohm resistor and 220uf cap.

That calls for a voltage drop of 1.88V (for a qiescent current of 4mA)..
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

ElectricDruid

Quote from: Box_Stuffer on July 03, 2023, 11:22:09 AM
I thought it looked weird too, but that is how the Indian guys in the youtube video that I watched had done it. The video was all in Hindi with no subtitles so I had to watch very closely and I'm sure that is how they attached it.

Maybe learn Hindi?! For all you know, they could be saying "We stuck this in here backwards to screw up those idiots who just try to copy our circuits!!".

More probably, it was an honest mistake - but it's still a mistake. Negative end of the cap should pretty clearly be to ground on a positive supply. It isn't the first time or likely to be the last time that something published has errors in it.

duck_arse




आपका 100 uF कैपेसिटर उलट है।
" I will say no more "

Box_Stuffer

When I turn the cap around the "right" way the problem actually gets worse.

I think I may try to re-work it based on a radio amplifier circuit I found. It is the only example from the data sheet that used the inverting input.



Rob Strand

#8
Quote from: Box_Stuffer on July 07, 2023, 08:19:35 PM
When I turn the cap around the "right" way the problem actually gets worse.

I think I may try to re-work it based on a radio amplifier circuit I found. It is the only example from the data sheet that used the inverting input.
You need to connect the 100uF cap the right way then deal with it.  The 100uF cap around the wrong way is probably loading the battery and hiding some of the problem, not fixing the problem at all.

See,


Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

Box_Stuffer

I rigged up something based on this circuit from the datasheet. It already seems better. I changed around the input and output capacitors and diodes from before to suit the new arrangement. I used Infra-red LEDs in the gain loop (replacing the 10uf cap) and 1N34A germanium hard clipping diodes instead of the speaker.



Box_Stuffer

The Blue Rajah LIVES!!!

I tweaked the radio amplifier circuit quite a bit and I finally got it all together. It sounds really good so far. I ended up using different diodes. There are germanium hard clipping diodes that give a smooth break up all the time and then a gain loop with 1n4148 diodes. As you crank it, it adds some silicon crunch. It's not super extreme - definitely in the hard rock / blues realm.