Rothwell Lovesqueeze debug

[Renegadrian]

1. Name of the project: Rothwell Lovesqueeze.
2. Schematic and Layout
3. Modified that layout a little to follow the following omissions
4. voltage doesn't go thru diode and resistance, omitted that resistance and had the diode between neg and pos. tracks
had no 22r, so put 12r instead. omitted the 200r at the input and output. had no 1N5817, put 1N5819 instead.
5. What it does and does not do: even if wired as millenium 2 as usual, sound gets thru in both positions with no battery.
it DOES work, LEDs light up, but with pots at miminum I have no usable effect level, as soon as I turn the pots more I get distortion and crazy oscillation. Led light up accordingly.

Anyone had similar problems?" How do I get it to work properly?!

[samhay]

The fact that it passes signal without power suggests something is very wrong.

Once that is solved - I would guess that you aren't getting any compression and by substituting the 22R for a 12R (R13 in the linked schem), you have doubled the max gain to about 400.

Not entirely sure what you have done to the envelope detector. In reference to your schematic, which components have you omitted and/or moved? As a start, the 22R (R9) is almost certainly not optional.
Did you get your FET from somewhere reliable?

[Renegadrian]

THX for the reply!
First thing I am gonna try is change DPDT, I used the same millenium 2 wiring I use all the time, so it's correct.
btw, I have a 12r for both resistors that should be 22r, R13 and R9. the other 22r on the 9V line has been ditched totally.
other than that, I figured the two 200r resistors at in & out shouldn't be mandatory, so omitted those as I wrote above.
FET is ok AFAIK, usual transistor dealer (tayda I guess)

[Renegadrian]

Another strange thing I figured out is that the schem calls for 1N5817, but the pic on la revolution deux shows two blue diodes, they look like BATs, not the black 1N5817.

[samhay]

I would put 2 12Rs in series for R9 and R13.
The 200Rs on input and output are good practice but not essential. I take it you either jumpered these or moved you input and output wiring appropriately?
R5 is for power supply filtering, so can be left out if your supply is not noisy.
We had a chat about the diodes in this circuit a while ago - http://www.diystompboxes.com/smfforum/index.php?topic=101129.0. Using Schotkies (BATs) will reduce the threashold, and should give you more squish.
I think 2N5457s from Tayda are susspect - http://www.diystompboxes.com/smfforum/index.php?topic=97600.0. Can you try it in a Fetzer or similar to check it is actually a 2N5457?

[Renegadrian]

I never had problems with FETs from Tayda...Guess I am gonna try to solder two 12r res in place of those 22r...
thx for suggestions.

[midwayfair]

4. voltage doesn't go thru diode and resistance, omitted that resistance and had the diode between neg and pos. tracks

I don't understand what you're saying. What part number in the schematic does this refer to, and what did you do? Do you mean, you eliminated D1 and R5?

Samhay -- cutting R13 won't double the gain. Remember that the resistance across the FET isn't 0! It's about 300-400 Ohms. The extra 10 Ohms is well within tolerances of the FET itself.

A few things are going on. First, it sounds like your effect isn't compressing. That's why you've got distortion -- the op amp is gained up so high that it will distort unless the FET's resistance is allowed to increase.

Three things to do:
Measure the DC at the anode of D3 with a guitar signal. You should get a few volts.
Measure the resistance between the gate and ground of Q1 without a guitar signal. It should be about 300-400R.
Now measure the resistance between the gate and ground of Q1 with a guitar signal. It should jump up several K.

These will tell you if your envelope is working.

Make sure you've made all the cuts. This has several that can't be seen on the layout that would severely break the effect.

[samhay]

Samhay -- cutting R13 won't double the gain. Remember that the resistance across the FET isn't 0! It's about 300-400 Ohms. The extra 10 Ohms is well within tolerances of the FET itself.

Good point - Sorry, I've been think about MOSFETs as of late.
I'm guessing R9 has an effect on the attack, so it is probably a bit more critical. That said, it shouldn't completely disable the envelope detection, so there is something more fundamental that has gone wrong.

[Renegadrian]

4. voltage doesn't go thru diode and resistance, omitted that resistance and had the diode between neg and pos. tracks

I don't understand what you're saying. What part number in the schematic does this refer to, and what did you do? Do you mean, you eliminated D1 and R5?

Yeah, R5 is out and the diode is between positive and negative (first two rows of the vero) parallel to C5

[Renegadrian]

Me stupid...
jacks weren't grounded, as soon as I grounded them ckt works as it should, and switch bypass too obviously...
Had some 22r in so it's a fast swap to be closer to the original schem. Hope the guy is gonna like it!

[samhay]

Excellent.
As Jon mentioned, changing R13 is unlikely to make any difference. R9 might.

How do you like the effect?

[Renegadrian]

not bad...cool pedal, subtle but effective...