Strange LED behaviour with Millennium bypass

Started by d3im0s, June 06, 2015, 08:02:26 AM

Previous topic - Next topic

d3im0s

Hi, we have built a couple of stompboxes and in both we see a strange behaviour with the LED. We're using the Millennium1 bypass and a DPDT switch wired as usual. The problem is that when the effect is off and we raise the volume pot the LED becomes brighter while it should be off. We tried to compensate with a trimmer inseries with the LED resistor and finding the value that keeps the LED off all the times, the only side effect is that when the effect is on the LED intensity is reduced by a lot...

Any suggestion? This happens with a 50k volume pot at mid-course and with a 100k pot at 1/3... The only thing I could think is that the volume pot should direct at ground the leakage current of the Millennium diode so that it doesn't open the current flow on the FET but if the pot resistance is high enough the FET opens anyways. Still I have no idea how to fix this...

Cozybuilder

#1
I don't know what you have wired in there, but if you want reliable Millennium Bypass, wire it like this:



Note: its important to keep the R and LED on the Drain side of the N-ch MOSFET, otherwise the function is unreliable. Don't sub for the 1N4148 either (*1N914 is OK)- the gold doping is what gives enough leakage to make this work.

Its also important that the circuit out be tied to ground- this can be through a pot.
Some people drink from the fountain of knowledge, others just gargle.

karbomusic

#2
QuoteWe tried to compensate with a trimmer inseries with the LED resistor and finding the value that keeps the LED off all the times

Try a bigger resistor parallel instead of series? I have a very similar circuit to the MB and I just strap a high value resistor across the output terminal/GND such as 3.3M.

d3im0s

Quote from: karbomusic on June 06, 2015, 08:50:19 AM
QuoteWe tried to compensate with a trimmer inseries with the LED resistor and finding the value that keeps the LED off all the times

Try a bigger resistor parallel instead of series? I have a very similar circuit to the MB and I just strap a high value resistor across the output terminal/GND such as 3.3M.
My bad, the trimmer is in parallel with the led...

The wiring scheme is like this

The FET is a BS170, the diode a 1N4148

Cozybuilder

The circuit you posted shows a JFET, not a MOSFET. Try the one I drew up- you have the correct parts for it.
Some people drink from the fountain of knowledge, others just gargle.

duck_arse

I also had a bad millenium, using a mosfet, with the led brights when bypassed following the volume setting. I don't know what the problem was, but when I replaced the surface mount fet with thru-hole, it worked.
don't make me draw another line.

d3im0s

Quote from: Cozybuilder on June 06, 2015, 09:14:38 AM
The circuit you posted shows a JFET, not a MOSFET. Try the one I drew up- you have the correct parts for it.

You were right, that was the cause. Actually moving everything to the drain side would have meant redoing the pcb... so I took a shitty pedal I had lying around and scavenged it for a couple of JFETs. Works like a charm now!

Cozybuilder

Some people drink from the fountain of knowledge, others just gargle.