Sharkfin - this one weird issue...

[mjg]

I've recently finished a Sharkfin (2015) build, and it works great, except for a niggling little issue or two.  I can work around them, but I'm curious as to the 'why' of what is going wrong. 

I wired it up with the top mounted connections, and a 3PDT board at the bottom. 

Problem 1) Works fine with the effect engaged, but when in bypass, there is a high pitch squeal noise over the signal. 

I fixed this by replacing the 7660SCPAZ with a LT1054CP with pin 1 lifted.  The squeal went away in bypass mode. 

Problem 2) Works fine with effect engaged, but when in bypass, there was a 'tick tick tick' in time with the speed of the effect. 

I tried a few things to fix this. 

- I disconnected one of the grounding pins on the 3PDT board, just to check that wasn't an issue - didn't help. 

- I removed the top mounted jack connections, and instead used some shielded wires to connect the jacks to the 3PDT at the bottom.  Disconnected the 'I' and 'O' wires from the board to the 3PDT board.  This has fixed the problem. 

So it seems that using the top mount connections on the board can somehow cause interference with the bypass signal.  This had me scratching my head for a while.  Is the signal on the PCB passing too close to something and getting interference?

Problem 3)  Effect works fine with the 'Rate' LED connected on the board in situ.  I decided I wanted to have the rate LED at the bottom of the enclosure, so ran a single wire from the +ve connection for the LED, down to an LED at the bottom of the enclosure, and used the ground connection from the 3PDT. 

The 'tick tick tick' noise is now back, when the effect is engaged, but it's more of a mellow 'wump wump wump' noise now.  It changes with the speed of the effect. 

Playing around with shielded wires didn't have much luck, until I finally tried using the ground connection from the LED near the top of the board, rather than the ground connection near the bottom.  The noise went away. 

So I know how to fix the problem now - run a ground wire for the LED.  But what I want to know is why this is happening.  Is it some strange ground loop?  And now that I've narrowed it down to a grounding issue, I'm guessing that the problem (2) above was the same thing - changing from the ground at the top of the board to the ground at the 3PDT seems to fix the problem. 

I've checked the ground for the LED and the ground on the 3PDT - they have 0.03 ohms resistance between them, which seems pretty normal. 

I'll post IC voltages and pictures of the board in a minute...

[mjg]

IC voltages.  This is with all pots turned to about 3/4 up. 

Input DC:  9.04v

IC 1: 

1  -7
2 -0.7
3 0
4 0
5 0.16
6 -8.1
7 0
8 -1.1
9 ~0.5
10 0
11 8.3
12 0
13 0
14 0
15 -0.8
16 -7

IC 3:

1 -8.55
2 0
3 0
4 -8.55
5 -8.55
6 0.14
7 8.8
8 0.01

IC 4:

1 -8.55
2 0
3 0
4 -8.55
5 -8.55
6 0.14
7 8.8
8 0.01

IC 5:

1 -8.5
2 fluctuates
3 -4.4
4 -8.5
5 -8.5
6 ~ -6.5
7 8.8
8 0.01

IC 6:

1 2
2 4.7
3 0.01
4 0.01
5 -8.5
6 2.6
7 1.4
8 8.8

[mjg]

... and the pictures.

[mjg]

I should add that the problems occur whether the board is inside the enclosure or outside.  So it doesn't seem to be anything contacting the enclosure causing issues. 

Also - the thing worked perfectly when I tested it in my test rig... but I didn't try the top mounted jacks when in the test rig, didn't try bypass mode when in the test rig, and I also didn't try the LED wired down to the bottom when in the test rig.  So didn't see the issues until getting it in the box...   

So, my next move is to run a ground wire down to connect that LED.  I was kinda keen to have a bi-colour LED for effect/rate, so I might pull out the 3PDT board and redo the switching, keeping the ground for the LED(s) separate to the ground for signals.  That should do the trick.  But why?  That's a mystery to me.

[mjg]

I've redone the 3PDT switch, and separating the ground for the LEDs from the signal ground has helped. 

The ticking noise is all but gone - it is still there, but  very quiet on the bypass signal.  You basically have to have the guitar unplugged or turned right down to hear the ticking over the hum of the pickups. 

So it's good enough for me, but still wondering what has gone wrong. 

[WormBoy]

So it's good enough for me, but still wondering what has gone wrong. 

It is quite possible that you did nothing wrong: it is simply a finicky circuit. I built two FSHs on Tonepad PCBs, and both had issues with ticking and whining. The combination of LFO's and voltage doublers is probably not so easy to get right. I had some luck curing ticking by selecting low-power opamps. However, for the first one, I resorted to more drastic measures: completely kill the LFO in bypass and filter mode (I can probably retrace what I did exactly, but I think I grounded one leg of the speed pot). When the effect is on, ticking is not a big problem (especially in SH mode, where it just adds to the experience  ), but in bypass it should be dead quiet.

[mjg]

Thanks - that's a nice idea to kill the LFO when bypassed.  For now I'm happy with what I ended up with, but I'll keep that idea in mind. 

I was considering completely cutting the ground from the board when in bypass mode, but I suspect that would make for a big pop when switching the effect on and off.  Just grounding the speed pot seems like a nice compromise. 

[WormBoy]

Cutting ground from the board sounds like a bad idea to me ...

I checked my 'archive', and I think I indeed grounded leg 3 of the speed pot in bypass (easy to do, on the 3PDT switch, connect it opposite the ground wire for the indicator LED; pot needs to be grounded when the LED is not). I don't have the pedal here (built it for a friend), so I cannot check definitively. But, if you are happy with the way it is now, don't mess with it  .

The second one I built has no ticking whatsoever, but there is a slight whine, which you only hear when there is a distortion on after it ... Pity, as FSH into distortion sounds spectacular.

[mjg]

Yeah, I had the Sharkfin followed by Aeons distortion the other night. Sounded epic.

[AntKnee]

Any time you off board wire something, you increase the chances of picking up interference from the LFO. PCB layouts are often tweaked for a while to get the components and traces situated so that they don't pick up interference from being in close proximity to the LFO. Since the rate LED is part of the LFO circuit, if you move it and run a wire near an audio trace or audio wire, you likely started picking up interference from the new rate LED wire.

Also, keep in mind that you can' t just simply use shielded wire and expect it to silence noise. You have to connect the shielding to ground for it to work right. I can't tell if you did that in the pictures. Even then, it may not do the trick.