blowerbox: need help (for the sake of my sanity)

[cylens]

hey there,

I'm a pretty stubborn guy, and I would love to have some advice on a build that I can't troubleshoot.

The circuit is the Idiotbox Blowerbox (for the record, it's not the first one i've built)

Here's the schematic (close to the vero by mirosol, and hosted by johnk):

http://johnkvintageguitars.homestead.com/Effects/Fuzz-ODs/Idiotbox/BlowerBox_Schematic.png

The problem is that it actually works... but "after some time": when I power the circuit, there's nothing at the output (and after probing, I have some signal at pin 3 of the LM308, but nothing at pin 6). after some time i get a heavily gated signal (starts at pin 6, too),and then, it starts sounding like a regular blowerbox...

I've checked for misplaced components, solder bridges, etc. a zillion times (though I'm well aware that it still can be there and I can't see it...) I've swapped the IC, I've changed all the caps in the power section and in the feedback loop of the IC... I've disconnected the 10uF cap after pin 6... I'm at a loss... Thing is, it already would have been much faster/easier to start from scratch, but I really "need" to understand what's going on... I think it's pretty difficult to troubleshoot, I prefer, when there"s no signal or when the sound is obviously "bad"...

any advice would be great! to me, there's a cold solder joint somewhere or a solder bridge, but I really can't find it...

[Silver Blues]

Do you have a 1M or larger resistor from J201 gate to ground in your build, like on Miro's vero (heh)? If not this may be the cause and I'd try first inserting that resistor and then if necessary replacing the FET.

[cylens]

hey, thanks for your reply, yes, I built it from miro's layout, so the 1M resistor is in place. changing the J201 didn't help, and to my understanding, the problem is "around" the LM308...

[cylens]

I KNEW IT, I F***ING KNEW IT

just posting a message here after > 10 days of finding nothing is enough to find the solution and make a fool of myself :D

 the 1M resistor biasing the LM308 was just missing... I have yet to understand why it was "somehow" working

[Silver Blues]

[Frank_NH]

 "

cylens wrote

the 1M resistor biasing the LM308 was just missing... I have yet to understand why it was "somehow" working

It would still "work" as long as you had some signal above 0V.  Without the bias to 4.5V you would get half your signal chopped after it passes through the input cap, but you'd still have something.

Now for extra credit - what would happen if you biased the op amp input at 5.0V?  4.0V?  Hmmm.  

(edit - I just thought...what if you substituted a 100K trimpot for one of the 100K resistors in the voltage divider that supplies the bias voltage.  That way you could experiment with some slight "misbiasing" - might sound good.)