millenium bypass V1.1 (Ge version) - problem (some fuzz content)

[duck_arse]

I have a layout and vero, cut and ready to solder, for the "Add" fuzz, circuit as shown here and currently living on the bredbord:

http://www2.marist.edu/~jzmc/AFTYEGOB.htm

original thread:
http://www.diystompboxes.com/smfforum/index.php?topic=577.0


mine has input cap and pulldown, feedback bypass (12pF) cap, less the 2 resistors following the vol pot. and seeing as the case already has a led hole, and seeing as I don't have 3pdt f/s's, we need a millenium of some description. and so, this:


and for the colour averse, as usual, this:.

the odd, low split supply causes a slight problem, solved by the (selected) J201 and the shunting pnp transistor. (the 2N990 is a v. low leak, low gain, Ge planar; it was handy/fortuitous.) it wasn't until I hung the cro on the mill that the problem became apparent.

humm. the jfet seems to turn-on on humm, which appears on the source resistor as 1V p-p, which then switches the shunt at 50Hz. prodding about with caps shows a ~220pF from gate to V+ or ground will stop the humming, and brighten the led. but hanging caps here will then cause dis/charging pops, won't it?

so, my options?
- harden the f up, build it as is and the metal box will cure all ills;
- add a cap to ground, tell the user to harden the f up, they all pop;
- some other solution you, dear reader, are about to impart, which still fits with the odd battery supply.

thanks for any and all suggestions, and anyone interested in the fuzz itself, let me know, I'll tell what I've found.

[anotherjim]

Maybe D100 needs to be more leaky. Have you tried a Ge diode?

[duck_arse]

this in advance, you'll know why when you see it.


today's fiddle shows the answer is harden the f up and solder it. it seems the problem was "a" ground, cause using screened to the mill cut the humm substantially at the gate, less so at the source. then mounting a f/s on a plank of metal, and a copper-clad under the bb, the hum could be gone-d. and brought back, but I don't know which was the magic wire.

the board is 90% soldered now, so the smoke test tomorrow. and the lesson:

when using the jfet versions, make all your grounds solid, because if you build it, they will hum.

[duck_arse]

and a final update: the circuit is 100% electrical now, functioning correctly, no AC on the led. and, as an extra special added bonus, i also have a 'jack is inserted, batteries are fitted, fx is still bypassed' indicator - the led glows very dimly when the pedal is bypassed. I didn't notice it on the bb, don't know why.

look to the wood thread for the final built pedal, soonish.

[PRR]

> led glows very dimly when the pedal is bypassed

May be working-as-designed.

Q101 could shut-off the LED. But somehow R102 got in there. Now Q101 can only divert some-but-not-all the LED current. Residual glow.

The odd supply voltages may have upset the original figuring. Making R102 zero should shut-off the LED (if Q101 has ample base current, not checked). Making R102 some intermediate value between 0 and 5K gives an intermediate residual glow. You have assorted parts, you can find a happier value. That is, if it is not happy-enough now.

[duck_arse]

my intention with R102 was to reduce the current drawn when the circuit was bypassed. I'm stingy like that. it was a case of bredbord joggle-ing to arrive at that value, but it seems I should have been gazing straight down into the led while testing. it only highlights/warms the edges of the die, it has so little current.

[PRR]

> reduce the current drawn when the circuit was bypassed.

Ah!! Cuz you using batteries! Makes perfect sense.

So you know where to go if the glow is a Problem.

Expanding the box, you could add another stage and get current to zero when not active.

However even with AA-cells I don't think the drain matters.

And as you say, it verifies that the batteries are in and not dead, before you go for the effect.