Hi everybody,

to have true bypass + LED, i have often used a DPDT switch following this scheme with PNP transistors and I have never had problems

https://www.muzique.com/schem/dpdt2.gif

but in this booster

https://ibb.co/jvV5sF7J

it does not work and the LED always stays on even when the effect is off and even if I put a 100k resistor between out and ground and if I increase the resistance the brightness fades in both the on and off states. I had thought of controlling the transistor with the gain pot but then where do I put the effect output ? Same problems with the Millenium bypass 2 and the Joe Davisson scheme using a PNP and NPN, or it may happen that the LED works in reverse, ON when it should be OFF and vice versa.

Any suggestions ?
Thanks in advance

Try a lower value resistor from OUT to GND..
It should dominate booster's gain so it's value should be a compromise..

I tried it but it don't changed

The problems with bipolar and darlington bipolar transistor LED lighting circuits in this kind of arrangement is what motivated me to come up with the Millenium Bypass. The Millenium II using a MOSFET switching element and the Millenium C work even more reliably.

See http://www.geofex.com/Article_Folders/Millenium/The%20Next%20Millenium.pdf and other geofex articles on bypassing.

As i wrote, unfortunately i also tried the Millennium bypass 2 with 2N7000 mosfet but i had the same problem.

Interesting. The Mill 2 has been very, very reliable and predictable.

All of the Millenium circuits, and I think the others as well, are detectors of whether the wire to the indicator control pin pulls the control wire to ground with a low enough resistance.

Since you have used this circuit before and it worked well for you, there is some question of whether the LED circuit is not working correctly, or the effect circuit doesn't work with it.

The way this circuit works is that the base of the PNP has to be pulled down towards ground with enough current/voltage to starve the LED from forward voltage. Test your LED circuit: set the circuit to ON (that is, not bypassed). This should leave the base of the PNP transistor connected only to the switch pin. The LED >>should<< be on.

Now take a 1K resistor and temporarily touch it to the switch pin that connects to the PNP base, and to ground. If the LED does not go out, something is wrong with the PNP, the resistor, and/or the LED. If it goes out, do the test again, but with a 10K, a 100K, and a 1M resistor. At some value of resistor, the transistor will no longer pull down the base of the transistor enough, and the LED will not go out.

Your schematics don't show the value of the volume pot, and this is the resistor that is connected to the transistor base when the switch is moved to the "off" position. It may be that your volume pot is just too high a resistance.

This, plus other considerations is what led to the Millenium and the Mill2. The Mill 2 will happily work with multi-megohm resistors. 

Quote from: R.G. on May 01, 2025, 04:46:03 PMYour schematics don't show the value of the volume pot

Because there isn't a volume pot, perhaps.??

Quote from: R.G. on May 01, 2025, 04:46:03 PMInteresting. The Mill 2 has been very, very reliable and predictable.

Maybe I didn't specify well, I'll do it now.

At the beginning I specified the booster schematic in which I would like to use the DPDT (I've always used it with a 3PDT but now I wanted to try the DPDT). For clarification, the schematic I posted refers to the Analogguru booster that doesn't have a ground resistor at the output but only a 10uF C3 cap. However, later I tried to put a 100k resistor between the output and ground but without success.

Now, if you look at the schematic, in this booster the P1 gain pot would also be there, it's 10k.

Well, I wanted to connect the P1 pot to the DPDT control pin but, as I had already written, I noticed that if I had done I didn't know where to connect the C3 output.

I hope I've clarified.

Quote from: antonis on May 01, 2025, 05:09:32 PM

Quote from: R.G. on May 01, 2025, 04:46:03 PMYour schematics don't show the value of the volume pot

Because there isn't a volume pot, perhaps.??

If you look at the schematic, there is no output pot but on the source of the jfet there is the gain pot.

The whole batch of LED turner-offers require a resistor to ground to shut off the LED. If there is no resistor to ground, they won't turn the LED off.

This is why I suggested trying a resistor to see if the LED circuit works by itself.

If there is no output resistor after the output 10uF, the LED won't turn off. A resistor that's too big won't pull the PNP's base down enough to turn the LED off.

Quote from: menstrel on May 01, 2025, 07:25:28 PMIf you look at the schematic, there is no output pot but on the source of the jfet there is the gain pot.

So what..??
That gain pot has nothing to do with output switch..
As R.G. said, you need a resistor to GND for the p-n-p to be bottomed (V_CE < LED's forward voltage drop)
The value of this resistor can be easily calculated by taking into account Emitter current and DC current gain..
(when dealing with BJT switches you have to be conservative by making Base current much larger than the calculated one - hence Base resistor much smaller..)

Quote from: antonis on May 02, 2025, 04:50:10 AM

Quote from: menstrel on May 01, 2025, 07:25:28 PMIf you look at the schematic, there is no output pot but on the source of the jfet there is the gain pot.

So what..??
That gain pot has nothing to do with output switch..
As R.G. said, you need a resistor to GND for the p-n-p to be bottomed (V_CE < LED's forward voltage drop)
The value of this resistor can be easily calculated by taking into account Emitter current and DC current gain..
(when dealing with BJT switches you have to be conservative by making Base current much larger than the calculated one - hence Base resistor much smaller..)

So in the schemme I'm attaching now it's not possible to make it work?

Yes, thanks to R8

I had already tried and it doesn't work.

Edit:
It happens that in bypass mode the LED turns off but slowly turns back on

See reply #1..

Find a proper value resistor for R8 and come back to tell you new gain mods..

e.g. for R8 = 10k, halve R5 & R6 values..

Hey guys, now it works, i had connected R8 directly on the jack socket between tip and ground.

Quote from: menstrel on Yesterday at 07:59:33 AMHey guys, now it works, i had connected R8 directly on the jack socket between tip and ground.

Good for pull-down (anti-pop) purpose...

Quote from: antonis on Yesterday at 08:20:10 AM

Quote from: menstrel on Yesterday at 07:59:33 AMHey guys, now it works, i had connected R8 directly on the jack socket between tip and ground.

Good for pull-down (anti-pop) purpose...

As an anti-pop sometimes it didn't work for me, but in this case, why did the LED stay on since in the bypass the R8 still went to ground via the DPDT ?

Quote from: menstrel on Today at 04:03:11 AMin this case, why did the LED stay on since in the bypass the R8 still went to ground via the DPDT ?

Because in any case Q1 Base was floating (DC open)..

When effect ON -> No connection ..
When effect OFF -> Base to C3..