millenium bypass won't light up

[amonte]

I'm having trouble getting the M2 to light up.  The bypass function works properly.

I'm working off of this version, but I am questioning whether it's correct - most of the other schematics I've seen have at least two diodes and another resistor. 

[pappasmurfsharem]

Mark has that down as a millennium 2 but I believe the millennium 2 uses another transistor. I had similar problems with the millennium 1

[slacker]

The layout is correct, it matches the schematic here http://www.geofex.com/Article_Folders/Millenium/millen.htm it is missing the low leakage diode but it will work without this.
Does it work if you connect the control line straight to ground? If it does make sure your pedal has a resistor to ground or volume pot as the last thing in the circuit.

[amonte]

I do have a volume control last.  I will try to hook the control to ground and see what happens.

One thing I'm not following, looking at this board...



unless I am reading it wrong, it looks line it goes 9V power supply -> limiting resistor -> anode, cathode -> drain

the board goes above goes 9V -> anode, cathode -> limiting resistor -> drain. 

Or am I reading this totally wrong...

[duck_arse]

the electrons are a simple minded bunch. they don't know or care if they encounter the led or its current limiting resistor first, as they are all in series. either way works.

[R.G.]

The Millenium circuits are excruciatingly sensitive detectors of the difference between resistances of a couple of megohms or less and several millimeters of air. You can test your Millenium circuit builds completely outside a pedal by powering it completely outside the pedal and touching the control wire to a 1M to ground.

[amonte]

That turned out to be great advice because I did have the board wired correctly.  The problem is the circuit switching doesn't seem to be affecting whether the light goes on or not, yet the switching does work (in terms of bypassing the pedal and turning it on properly).  The last thing I have is a 100K pot - that is what goes to the switch.  Should I add a large resistor to ground off the volume pot? 

[R.G.]

That turned out to be great advice because I did have the board wired correctly.  The problem is the circuit switching doesn't seem to be affecting whether the light goes on or not, yet the switching does work (in terms of bypassing the pedal and turning it on properly). 

This is one of the hidden things in the Millenium - it does not affect the signal switching at all. It only indicates its control line has been connected to a resistor to ground, which is a byproduct of the true bypass switching setup.

The last thing I have is a 100K pot - that is what goes to the switch.  Should I add a large resistor to ground off the volume pot? 

A 100K pot that has signal on one side and ground on the other will be find, no additional resistor needed.

I recommend that you test your Millenium board as follows:

(1) Check the actual manufacturer's datasheet for the proper pinout of the MOSFET. Now check it again. The pinouts on some MOSFETs are confusing. Be sure what you think is source, drain, and gate actually is.
(2) Remove your Millenium board from  your pedal. Temporarily power and test it outside the pedal, leaving the control wire open.
(3) Temporarily parallel the diode with a 1M resistor. Verify that with the 1M in place and the control line open, the LED lights up. If it does not, there is a problem with the connections or integrity of the MOSFET, power, resistor, and/or LED.
(4) If it does light, touch the control wire to ground. The LED should go out. If it does, temporarily solder a 100K resistor to the end of the control wire. Touch the end of this resistor to ground, being careful not to "short" things with your fingers. It's that sensitive. You should be able to still get the on/off action from the LED.
(5) If all is good so far, remove the 1M paralleling the diode. Does the LED still light when the control wire + resistor is open? If not, this particular 1N914 diode does not leak enough to turn on the MOSFET on this board. This could be because of conduction through excess flux left over from soldering - it's that sensitive. If there's residual flux, clean it with an old toothbrush and rubbing alcohol. If it still doesn't work. parallel up to two more of the 1N914s with the one that it there. Some 1N914s are so good today they don't leak enough.

Biasing the MOSFET on with the leakage of the diode is the central idea in the Millenium. Modern diodes leak pico amperes (that's 10 to the -12th!), but the gold-doped high speed signal diodes leak 100 times more, in the nano-amperes, and this is enough to turn on a MOSFET. But some signal diodes are both fast and low leakage, so sometimes you have to look for the right diode or use more of them. It's a by-product of modern semiconductor processing being so very, very good.

[amonte]

I decided to remake my M2 terminal strip from scratch, being careful about keeping everything nice and neat this time.  Hooked to a 9V outside the pedal, it worked fine.  Then I put it back into the pedal and mounted the LED and it worked fine when I used the 1M resistor to test the LED.  Added the control wire and it was working fine with the switch.   Everything looked good so I closed up the pedal (which is in a thin, tight sloped enclosure.

Well, when I plugged in to power I saw the yellow LED go from yellow to orange to red, blink twice and go out.  Since then, it has not lit up again.  My guess is that I accidentally grounded something and burned out the LED, but is that even possible?  Or perhaps damaged the mosfet?  Any ideas what might cause this weird behavior?

[duck_arse]

that sounds like your current limit is wired between the battery and the dc socket, with the led wired between the dc socket and the board. it'll work okey w/ battery, will dark the led w/ external. check that wiring, again.

[amonte]

I'm actually not getting the LED to light with a battery or with a power adapter now.  Could it be that something shorted and took out the BS170 or the LED (if that's even possible)?

Here's how I have it wired - I actually have a 9V battery snap connected to the DC socket and then a second battery snap wired to the pedal.  It's setup so that I get +9V from the red lead on the "board" battery snap whether it is connected to the DC line (via the first battery snap) or connected to a 9V battery.  I have confirmed this with my meter, it shows 9V (or 8.58 and 9.39 for the battery and DC power adapter respectively).  The 9V connection to the M2 board comes from the same point as the 9V point for the rest of the board. 

[R.G.]

I'm actually not getting the LED to light with a battery or with a power adapter now.  Could it be that something shorted and took out the BS170 or the LED (if that's even possible)?

It's not just possible, it's likely. I think you have a Darkness Emitting Diode (DED).

[amonte]

haha, OK I will try to swap in a new LED and see what happens.

[bluebunny]

I think you have a Darkness Emitting Diode (DED).

I once made a Darkness Emitting Triode.  Shortly before that (and very briefly) it was a "super-bright" Light Emitting Triode.       Ooops...

Back to your regular programming...