Author Options:

POT ARCING , HELP ! Answered

Potentiometer ARCING inside :(
Anybody know what's going on  ?
I'm using a 100k to control a 50k ohm load .
i smell lots of smoke .
It's a LM386 with pots at the output section to a 8W power amp .



7 years ago

Something's a little off here...

First of all, most POTs are rated for only about 1/4 of a watt--if you tried to use a POT as an attenuator on the output, it would probably burn out, 'cause some 386's can supply up to one watt....IFthe load was a speaker.

But it isn't.  You're using the 386 as a preamp for a separate power amp. It's very likely you are correct, that the input impedance for that power amp is much higher (50K?) than a speaker. So the current draw should be very much smaller.

And it sounds like the POT is correctly connected as a voltage divider (one side to output of the 386, other side to ground, and the center wiper to the power amp).

You have a coupling capacitor between the 386 output and the POT/power amp? If not, you're passing DC through, and that could be the problem.

Otherwise, check your wiring...

NOTE: that HUGE 470 uF coupling cap in the "Portable Guitar Amp" instructable? You only need a large capacitor with a low-impedance load like a speaker (4 or 8 ohm), not with a 50K load. Plug the numbers into this RC calculator--compare the cutoff frequency for a 8 ohm load vs. a 50K load...

Anyway , i now think it's a bad idea to have a LM386 preamp .
Thanks anyway , i guess its the cap .


7 years ago

I don't blame you for dropping the LM386 as a preamp. 386's usually have a pretty harsh and abrupt "decay" when used as such; the more distorted, the more abrupt the cutoff. There are a few 386 stompboxes out there, but it takes some additional circuitry to prevent the unpleasant artifacts.

For a simple replacement, maybe look at the Fetzer valve ?

Good luck!

Oh yeah--the REPLY buttons are missing from our two comments below, so I couldn't reply to your msg. Another day, another bug...

You didn't tell me it was harsh , I LOVE IT .
Well , the input impedance is TOO LOW to be useful with my strat .

Why not add a simple buffer, like a JFET before the LM386, to increase the input impedance?

Any input Z over ~250K should be sufficient, and the JFET buffer on AMZ is about 1M ohms...

And now those two comments are "sticky" and above everything else? Weird.

Its a bug in the comment featuring system.

By the way , does anyone know the parts of the Piledriver pedal from modampkits ?

Sounds like there's too much current going through it - unless your pot is damaged and it's arcing from a weak connection...

Alternatively it could be too high voltage...

I didn't know the voltage, however is there a big current passing through it?

If not then it's likely a bad connection inside the pot itself - if you have a spare one of the same spec wire it in to check...

what the ... if it is then it doesn't sound good anymore :\
i searched my whole town and only they sell the A100k pot ...

I meant that single pot might have a manufacturing defect...

Otherwise posting the circuit diagram might be an idea just in case it's off in some way...

oh okay .
That's how it is , i followed that 'ible and the output and pot goes into a 8W power amp ,
When looking from up ( the rotary thing ) , 1 is to ground , input 3 , output 2 ( into power amp )

Can you link to the instructable? Also the comments of the instructable might be worth reading - other people may have had the issue themselves if it's a circuitry issue.

Unless the pins are mixed up it still sounds more like the pot's damaged...

It's this one :

That circuit doesn't call for any resistors so where did you add it in?

In the output lines themselves?

Right, when you change the value, say from maximum to minimum is there any appreciable difference in output?

It may also be easier to control it with a voltage divider - one inline resistor and the pot in parallel with the resistor.

My other thought is that you could increase the efficiency of the idea by controlling the input level as opposed to the output.

Also in that circuit if you can find a variable capacitor with the right values you could have a distortion knob too...

The distortion ? i have a 5 switch gain circuit as shown above .. not to worry about that , i left that off .
i tested it , there is a appreciable difference , or should i bypass the pot ?

Better plan - looked at the datasheet for the IC.

You can control the gain on the IC via pins 1 to 8 by adding a resistor inline with the capacitor that turns it up to maximum

The other way is a capacitively coupled resistor from 1 - ground

I see that the distortion controls are done via 1-8 as well, do the caps affect the volume much, essentially any capacitor should give the same volume, once bypassed the gain is set to maximum unless you add a resistor in series...

The internal resistor for minimum gain is 1.35 Kohms