Kingslayer whine issues

[Bassisst90]

Hello all. I have a big problem with the Mad Bean Kingslayer. I breadboarded the circuit and it worked quite well. So I etched the board and made no part substitutions, did not use the soft clipping diodes and switch, used a buffered bypass with a DPDT footswitch and now the circuit only makes noise. The circuit makes a whining sound constantly, like a signal generator. The dirt control acts as a pitch control and the other controls slightly affect pitch as well, as does the clipping switch. I'm at a loss. I made sure that the IC's were facing the correct way and I'm using a TC1044SCP. The one thing that I had a problem with is that when I made the board, the ground plane was cut off, so I had to connect all the ground connections with wire (basically 3 wires connected from each part of the severed ground plane pieces, thought I could get away with it)

Here's a link to the schematic and project: http://www.madbeanpedals.com/projects/Kingslayer/Kingslayer.pdf

Here are the voltages
Voltage from jack= 8.94v
IC1
1= 0.02  14=0.04
2= 0.01  13=0.01
3= 0.01  12=0.01
4= 8.11  11=-7.43
5= 0.01  10=0.01
6= 0.01  9=0
7= 0     8=0

IC2
1= 8.12  8= 8.11
2= 4.5   7= 3.97
3= 0.01  6= 3.44
4= -3.03 5= -7.43

Zener diode
A= 0.01
C= 8.11

Please let me know if you have any ideas as I'm willing to try anything.

[anotherjim]

Anything? 

IC1d - there's positive feedback from output to + input. That could lead to oscillation. Check R9 from Dirt pot wiper is really going to ground and ALSO C9 negative goes to ground.

[Bassisst90]

Thank you for your reply and suggestion anotherjim. So I checked R9 and it is connected with all the ground connections, however according to the schematic it seems that C9 does not go to ground. It seems that it passes signal from the IC1d output and leads to the tone control. Anyway, that didn't fix the problem. Also, the voltage from pin 5 of the charge pump is actually -8.5 and pins 1 and 2 is positive 8.5. I think something was grounding as I was checking the voltages.

[anotherjim]

My bad  - too early in the day when I replied. I meant C10 neg to ground. I think it's possible you have feedback around IC1d making it a phase shift oscillator if C10 isn't connecting to ground.

+ as there's a lot of gain, I'm sure you need screened leads for the audio connections. That's between footswitch, pcb and sockets. Sometimes pot wires have to be screened too.

IC1 voltages look good from your first post. Power supply balance isn't all that critical.

It's nothing to do with the problem, but that 9.1V zener across the power isn't worth much. Accidentally use a 12v supply and it'll probably fail. If it fails short circuit it'll short out the supply -  if it fails open circuit it won't protect anything.

As it's your own etch, put the meter on resistance and check off connections between components. It's possible to get very small gaps in the copper that the eye cannot see. If you find one, just run solder along the trace and "sweat" a thin strand of tinned wire into the solder by running the iron along it.

[Bassisst90]

I checked C10's connection and it is connected to ground. I've been looking all around the board and trying to make sure that traces which should not touch are not touching. I used an x-acto blade to make sure the traces are separated. I  really am at a loss. Is there anything else other than IC1d that could cause this oscillation? I'm using shielded cable for the input and output jack as well as the output from the board to the footswitch. I did not shield the buf and x wires from the board to the footswitch, but I'm not sure that much noise can be picked up from there.

[anotherjim]

Have you read this from the build doc?
"09.17 Update: A couple have people have pointed out that running the Kingslayer at the max gain
setting with both the soft and hard clipping diodes lifted caused some motor-boating and/or
oscillation. This is due to the higher gain produced by using 1M for R11. For the most part, you do not
want to use the max gain setting with no diodes, but it does sound pretty good up to the half-way
point (before harsh op-amp clipping sets in). If you want to avoid this situation altogether, socket R11
and try some lower values, something between 470-750k. Keep in mind that this will also reduce the
max gain when the clipping diodes are engaged, so it may require testing a couple values to find the
best balance for your needs."

Ok, you still have hard clip and this speaks of soft and hard clip diodes removed, but I can see this feedback problem might exist without the soft clippers only. So, you might have to alter R11.

Wire "X" is only connected by the switch when bypassed, so when the effect is active, the full signal out of the high gain amp IC1d is radiating out of the wire. I would use screened wire, but the screen only needs to be connected to ground at one end.

Make sure R13 & 14 are the correct values. 1k5 and 15k Colour codes are close enough to swap them around by mistake.

[JerS]

Although the specs on the TC1044SCP list the oscillator frequency above the audible range, in practice, it is sometimes much lower.

There was a thread at Madbeans on this:
http://www.madbeanpedals.com/forum/index.php?topic=6209.0

Try a Max1044 or something else and see if the whine goes away. Sometimes even a different TC1044SCPA is enough.

[Bassisst90]

Alright I'll give those a shot. Thanks guys

[chuckd666]

Yeah, I'd also recommend just trying a bunch of different charge pump ICs. I always get fairly mixed results with charge pump whine - sometimes they're quiet, sometimes they're deafeningly whiny.