Millenium LED ON for ever!

[Shakal]

Hey, good news U did it man! Wrong datasheets sucks!!  But me? No... I didn't fix it...

I guess I will have to change my FET.

[Shakal]

IT'S DRIVING ME CRAZY!!

The FET isn't the problem... I think my pedal is cursed by some anti-DIY ghost! 

Well... Today I started a search for the problem and I found that there are 2 pinouts for the same FET. 

So I tried the knew one, DSG, and... Again always OFF. Then I held the little Millenium board with my hands and the LED turned on with a very little light. I realized that when I touched the Source and Drain the led turned on, and when I pressed it harder it became brigther... But the most weird thing is when I pressed the stompswitch the led turned OFF!! Whatahell! To simulate my hand's resistance, I put a resistor there and tried the pedal. Put the coord, played a little with the pedal in Bypass(with the LED off), then I turned the pedal on (So did the LED) and found that the Level was on 0, I turned the pot and the pedal worked just fine... BUT! When I turned the pedal OFF the LED did stayed on, so I turned the pot to 0 again and the LED turned OFF!

In short: I placed a 2.2K resistor between Source and Drain, but the Millenium works just when the Level Pot is on 0.

I made a sketch on paint:

Obs: The Output is at the middle of the Level Pot (It's connected to Control, when it's on Bypass)

What is going on!?

[Shakal]

I made a mistake (again)! I changed the Source pin to Drain pin... 

I fixed it and now it stays off all the time...

Do someone know how to fix it? Pleaaassee??

Otherwise I think I gonna make another switch to ON/OFF the LED. 

[Shakal]

ALELUIA! IT WORKS!

I was about to give up... but fortunally I decided that I had to try again.

I mounted the circuit in a protoboard, but happened the same thing. Then I changed all the componets, instead of BC245C I tried J101 and MF102, changed the Diodes from 1N4148 to 1N4007, tried another resistors, short circuited all the pints of the FET, I tried all the possible pinouts, I reversed the diodes, and EVEN THE RESISTOR! I tried all that was possible, then somehow I got it to work! THANK GOD!

The final schem is the same as the picture above, without the Red resistor, with this parts:
D1 & D2 = 1N4007
D3 = 1N4148
R1 = 1Kohm
Q1 = MF102
LED = Green Ultra bright LED with 2.5V

Now it works, no matter how the volume pot is. But I got something strange, I think it's typical for this version of Millenium Bypass. The thing is when the cord is IN and the pedal is on Bypass, the LED keeps a super little bitsy light... actually it's not even a light, but when you see the LED from top you can see that it's a green LED, it keeps a little green mark inside it. But when I pull out the cord the LED fades completly.

Well, thanks for everything! I hope that everyone that see this thread, searching for millenium problems, can solve the problem.

See ya!

[Timebutt]

Great news!
Congratulations on getting it to work, I know how frustrating it is when you can't get it to work

[Shakal]

Man, it's so frustrating... I couldn't even think to start another project, that LED has kept hauting me!

Let's move on! 

[R.G.]

The annoyance of the very-dimly lit LED with the Millenium 1 is what motivated the Millenium 2.

[Shakal]

Ya... I read about that before, but I just wanted to see how was the little light. The next one will be a Millenium 2, I think.

I saw that  Millenium C... What are those CMOS inverters? What is the part number of them? I heard that this Millenium C is cheaper than the Millenium 2, is it true?

Thanks!

[Timebutt]

Well, if you know that the Millenium 2 circuit costs less than 0.45€ ... can you possible want it even cheaper?

[Shakal]

Man... I'm not European, not even north American...

This kind of thing cost a little more in here... Of course it's not expensive but, why pay 5 cents when you can pay 3?

The Mosfet costs here, R$3.00 , something like 1.00€.

Does someone knows the part name of a CMOS Inverter?

[R.G.]

CD4049
CD4069
74C14
74C04

Many others can be used as well. Inverting gates can be used by tying the unused inputs active.

[Shakal]

Thanks so much!

I'll compare the prices of both (Millenium 2 and Millenium C) and see which one is the BE$T.

I didn't get one thing... On the Millenium C schem there are 2 diodes connected between the Control Resistor and 9V, why there are 2 of them connected in the same place? There are a "..." between the too diodes, this means that you can put 1 or 2 or more diodes? Will it work with only one?

See ya!

[Shakal]

Millenium prices:

2N7000 = R$3.00
CD4049 = R$0.45 
2 diodes, 2 resistors and a LED = R$1.00 :

Millenium 2 = 4.00
Millenium C = 1.65

We have here a 60% off! I've made my choice, Millenium C indeed! Now I've to see if it really works.

See ya!

[joelap]

hehe... same problem here now.  Using a BS170 Millenium 2 setup.  I know the BS170 is DGS pinout, since I used it before in an octobooster no problems.  It is being powered in an effect that has a 15V voltage source.  My BS170 is rated for 60V so its fine.  Also, I upped the 4k7 resistor to 10k.  LED is a high brightness blue (clear diffused) from smallbear.  Basically, the first time I used it, it worked fine but had a lightly dim blue glow in bypass (nothing unbearable).  That was using 2 1N914 diodes.  After playing for a while, the off position became increasingly bright.  So I clipped the lead on the diode from Source to Gate, and same thing resulted.  I then took the collector and base of a 2N3904 NPN transistor in place of the Source to Gate 1N914 diode.  I tried this in both orientations, and the LED is very bright and almost flutters.  The circuit works fine though, bypasses great, but the millenium is staying on.  Odd, considering it was working fine and then progressively weakened.  Has anyone figured out a solution, or just puckered with it until it finally rectified itself?  I'm assuming my soldering is fine, considering it did at one point work almost perfectly.  Also performed continuity checks and nothing that shouldnt be touching is.  The circuit in question is a vintage 70s MXR Flanger... and before anyone jumps at me for drilling holes in a vintage piece, it was already butchered before I got it... I just tried to remedy the damage a bit! 

I appreciate any help.

EDIT: I've now tried replacing the source to gate diode with a 1N4007 diode, LED is still very very bright.  I've also placed a .047 cap coming from the switch to the millenium control too, and no luck there.  Re-doublechecked my soldering and pinouts again, and everything is good.  Frustrated just about sums it up!   

[miqbal]

hehe... same problem here now. 

If you integrate the Mill2 circuit with the main PCB, better try to make a separate one first. Make sure it work well. ALL socketed for the BS170, 1N914, and the low leakage diode (2N3904, etc). In this way, you can try out all your stuffs (diode, BS170). My experience: I have a working Mill2 after my 4th BS170!! Different diode, different leakage too. Try it without the low leakage diode first.
It was frustating at that time.

[joelap]

Those are good suggestions.  I wonder why this circuit is so picky?

[R.G.]

It's picky because it relies on sensing currents that are so small they're hard to measure with real, no-fooling laboratory instruments.

That is in fact why it's good - it senses large resistances to ground as "short circuits" and can tell the LED when to turn on and off.

[joelap]

Interestingly, the LED gets brighter over time.  I'm trying to figure out how that could happen.  An increase in drain current will cause the intensity of the LED to increase... so what could be causing the drain to draw more current then?  I'm trying to figure out the currents... Maybe the problem is because I'm powering the supply with 15Volts instead of 9?

[tranceracer]

Try cleaning the flux from the soldered side of the board and sanding between the traces and the pins of the transistor and the other components. 

I had a similar problem troubleshooting my latching relay driver using the 4049 IC where my relay would stay latched when I pressed the button.  The LED would work fine.

I found for some reason there was "ghost" current keeping my relay latched even though I had good solder joints and no solder bridges.  I throughly cleaned the flux from the PC board with a brush and acetone and the relay was still in latched position.  After rechecking the ckt on the board numerous times and over an hour of frustration I was about to give up and trash the board so I decided to sand between the pads on the IC.  Now it works! 

-tR