Signal with power disconnected — puzzles me …

[morten_beta]

So I build this one:

http://www.muzique.com/images/buff8.gif

But with an LM741 for the IC, and it seems to be working. However, when I disconnect the power — signal goes away for a few seconds, but then a signal returns, How can that be?

I haven't had a chance to actually test the buffer with a lot of cable and true bypasses yet, but I'm assuming that the unpowered signal is different to the powered one — still I'm puzzled as to why I get sound with power of. (I have an LED installed so I can tell that the power actually goes away).

Cheers for any input.
Morten

[Hatredman]

A video of this one would be interesting.

[morten_beta]

Not the worlds best audio — but I think you'll get the idea.

http://1777.dk/buffy.mov

[GibsonGM]

Perhaps there is bleed-thru between the - and + inputs? (??)

Can't open the .mov file, unfortunately.   But there's always a reason for things like this! Maybe your signal is large enough to open the gates on the opamp? But I'd expect it to need to be over .5V, anyway.  Sounds really weird.

[Thecomedian]

With gain at full, my fuzz faces tend to work with enough force against the strings.

It might have to do with the internal differences of the chip

[morten_beta]

Sounds interesting — the thing is though, I don't have to use any force. Just playing nice and quite will still put sound trough. And I still find it puzzling that it goes of for a few seconds before sound come back on — I mean, the caps are probably draining, but then what happens ...

It's more one of those things as to why it is happening — because I assume the thing is working.

I am going to dig out some looong cables and test it out.

[Hatredman]

Ghost in the shell

[R.G.]

There is always some leakage of voltage through a complicated clot of semiconductor junctions and resistors - an integrated circuit, in other words.

I suspect that this is sensitive to loading; it may be the high input impedance of an amp connected to it can pick up a signal, where if you put a 10K to ground after the unpowered circuit the signal might drop to nil. Just a suspicion.

I've run into similar situations before. In some cases there is a path through the circuit that is turned off when the power reverse biases some junction and causes it to not pass signal through that path, forcing it to go through the intended path for power on.

I think this is a curiousity, not a "let's run to the patent office!" kind of moment.

[morten_beta]

@R.G. a 10K almost took out the signal, there seemed to be a little bleed left with full humbucker (I have a spin-o-split) in single coil-mode signal was gone.

I wasn't worried about the patent office  — but rather that I screwed something up and had to go back in, since it's point to point soldered, and I just didn't want to have to go there.

Your comment puts me at ease though.

[Hatredman]

It would be interesting to check voltages when the weird stuff happens. You may discover where the ghost power is coming from.

[Thecomedian]

Yeah, I forget whether it was a 9v battery guitar or a completely passive pickup setup, but battery powered fuzz faces of mine would "work" with medium or greater strumming on the strings (with the battery disconnected). Think it required the gain full open.

Could just be another version of this.