Help please with strange problem or phenomena with a miniboster with some tweaks

[solderman]

Hi
I have a very strange problem or phenomena I cant understand. I have built the schematics below. As you can see its basically a miniboster with some tweaks. It works and sounds grate. But when I cause ruble (the kind that is produced when you touch the signal tip of your cable to the amp or from the guitar) it all of a sudden stops working. At first I thought that it was a lose cable or a short of some kind. But after examine the PCB I discovered that when I touched the solder side it jumped back to life. After a long time examining and testing I discovered that the thing operates like a switch. When exposed to hum from a non earth signal it switches off and when I momentarily shorten the Gate of Q1 to GND I jumps to life right away. I have tried different settings to the trim pot with the same result.


Can anyone pleas tell me what I should do to fix this behavior. I want to be able to change guitar without screwing off the lid of the box and touching the solder side every forth time??


Can anyone tell me why this is happening so I can eliminate the cause in the rest of my builds?



Can it have something to do with the size of the PCB that some components get to close? 
Q1 is the one next to the electrolyt 



Readings
    Working            Not Working
Q1
D=4,70                  5,46
S=0,45                  0
G=0                       1,85 and decending to 0 (and jumping to life)

Q2
D=8,93                  8,93
S=4,70                  5,46   
G=4,25                  4,25


Thanks for any suggestions'.

//Solderman

[alanlan]

Main problem is that Q1 gate is floating - you need a pull down here.

[petemoore]

  I called it a pulldown too, seems like it'll pull the gate voltage down...
  not sure but I think 'pulldown' has come to mean 'anti-pop' resistor, pulling the DC voltage down on capacitors in SP to switch.
  But yupp, Q1 gate looks like it could use a 'bias resistor'.

[solderman]

Main problem is that Q1 gate is floating - you need a pull down here.

Thanks
If i understand you correct, a R=1M? from Q1 Gate to GND would fix this??
Ill try this.

//Solderman

[R.G.]

To expand on that, when you put large signals into the input, the gate of the first transistor conducts on alternate half cycles when the polarity lets it. This deposits charge on the input capacitor that can then not leak away because the gate is a low leakage diode. Putting a resistor to ground on the gate lets this charge leak away. It's the same thing as grid blocking in tube amps, except that a JFET setup can't eventually leak down without a resistor.

MOSFETs don't do this because their gate is insulated and does not act like a rectifier.

[solderman]

Thanks all
I soldered one R=1M from the gate of Q1 to GND. Now when the hum brakes loose i can here it go down but it gets back up again by it self in a Blink.
Thanks to R.G I remotely understand why as well

Again thank a bunch, you saved my most valuable Build, Mr MiniMan. He just needed a pacemaker 



//Solderman

[alanlan]

Main problem is that Q1 gate is floating - you need a pull down here.

Thanks
If i understand you correct, a R=1M? from Q1 Gate to GND would fix this??
Ill try this.

//Solderman

Yes, but obviously it will half the input impedance to 500K.

[alanlan]

BTW I really like your build and the circuit board is a nice bit of space engineering.

[solderman]

BTW I really like your build and the circuit board is a nice bit of space engineering.

Thanks, Glad you like it
You can go here if yo are intrested

http://solderman.fatabur.se/MiniMan/build/MiniMan_build.htm

BTW I can live with 500K i think 

//Solderman