Nurse Quacky debugging

[Mark Hammer]

Then it would seem like the reason you don't get any sweep is because there is no guitar signal hitting the envelope detector.

[jimladladlooklike]

Okay so just realised that I used the wrong caps for C2 + C3. The bag stated they were 4n7 but when I checked the code online (471) I realised that they are actually 470pF.

Obviously I will now have to swap these out, but do you think this could have been the source of the problem?

Thanks!

[Mark Hammer]

It could certainly have been the basis of not hearing very much, because a pair of 470pf caps would have set the filter range up very high, likely beyond the actual frequency range of the guitar.

But they would not be the source of not getting any signal voltage fed to envelope detector.

So, you have two things to fix.  Getting there, slowly but surely.

[thermionix]

On my breadboard, one of the LEDs stayed lit, the other responded to input.  Are you getting that?

[rankot]

On my pedal (but it is combination of Nurse and Dr) one responds to input, another one is dimmed, as far as I can remember. So I used responsive one as a cigarette amber

[anotherjim]

Maybe relevant. In designs that use a BJT as a variable resistor and ranging the sweep seems difficult - do try swapping the emitter and collector over. Many old commercial designs did this. Thing is, you don't need a "gainy" transistor action in that job and the low Hfe obtained from the reversed connection (aka reverse beta) can better suit the job. I was reminded of this by seeing a Farfisa Wha that swept a twin-T filter with a reverse beta BJT as the ground resistor.

[thermionix]

combination of Nurse and Dr

I hope you named it the Nurse Practitioner.

[Mark Hammer]

On my breadboard, one of the LEDs stayed lit, the other responded to input.  Are you getting that?

The LEDs do not have to light up, and whether they do or don't is not a reflection of the functioning of the circuit.  They are being used as diodes with a given voltage drop.  If the ones you use happen to be efficient enough (i.e., superbright variety), there's a good chance one or both will light up.  But builds using older less efficient ones will still work fine.

If you do happen to use a superbright for the one on the output of the op-amp, you can panel-mount it and use it as a sensitivity gauge.

[thermionix]

Ah.  I don't know if I was using superbrights or not.  I just grabbed two 3mm reds from a drawer of random bits, and they were nearly exactly matched for Vf, so I used those.  Come to think of it, they lit up when measuring with the Fluke, not sure I've seen that before, so maybe they were superbrights.  Vf was ~1.8V, don't know if that indicates anything.

[jimladladlooklike]

It works!!! I swapped out the 470pF transistors, each for two 2n7 (or 2n2, I can't remember) wired in series, and it works a treat. Not sure why my measurements were not showing as they should (could I have been doing something wrong?) but the proof is in the pedal I suppose!

[thermionix]

Try them in parallel.

[Mark Hammer]

The frequency range of the filter is set by the two caps.  The larger/higher their capacitance, the lower the range of sweep.  Caps with 4n7 value are suitable for guitar.  For baritone guitar, 6n8 would be more appropriate, and .01uf-.012uf would be optimal for bass.

Putting a pair of 2n7 caps in series gets you an effective capacitance of half of 2n7 (=1n35).  That certainly moves the range down from 470pf, but nearly close enough to where you need to be.  As Thermionix suggested, put those caps in parallel.

[jimladladlooklike]

My bad, I meant parallel...

[Mark Hammer]

No probem.  That better explains your degree of satisfaction with it.  Ironically, sifting through a box of semi-finished circuits last night, I pulled out a Dr. Quack build.  Depending on other chores today, I was hoping to install a variable decay time.