Help debugging Madbean's Loophole

thom · December 17, 2014, 06:50:37 AM

Hello all,

Not the most common build, but I hope someone will be able to point me in the right direction.

What's good: the leds switch on as expected, it records and plays back.

What's bad: horrible background noise during playback, possibly a grounding problem?

Schematic: http://www.madbeanpedals.com/projects/Loophole/Loophole.pdf

IC3 has been replaced with an ISD2560 with pins 7 and 8 grounded (found here: http://www.madbeanpedals.com/forum/index.php?topic=3029.0).

Voltages (in brackets voltages when record switched on):

Ic1
1: 4.46
2: 4.46
3: 4.46
4: 0
5: 4.18
6: 4.48
7: 4.49
8: 8.91

IC2
1: 4.96 (0)
2: 4.99
3: 0
4: 4.99
5: 0
6: 0
7: 0
8: 4.70 (0.20)
9: 4.70 (0.20)
10: 0 (4.01)
11: 0 (3.95)
12: 4.96 (0)
13: 4.96 (0)
14: 4.99

IC3
1: 0
2: 0
3: 0
4: 4.96 (0)
5: 0
6: 0
7: 0
8: 0
9: 4.96 (0)
10: 4.96 (0)
11: 0.19 (0.14)
12: 0
13: 0
14: 0
15: 0
16: 4.98
17: 0 (1.45)
18: 0 (1.45)
19: 0.18 (1.46)
20: 0 (1.44)
21: 0.18 (1.56)
22: 4.98 (0)
23: 4.96 (0)
24: 4.70 (0.11)
25: 4.99
26: 0
27: 4.96 (0)
28: 4.99

IC4:
1: 4.46
2: 4.46
3: 4.46
4: 0
5: 4.27
6: 4.48
7: 4.48
8: 8.92

My perfboard layout:

What it looks like at the moment:

Thanks,

Thomas

anotherjim · December 17, 2014, 07:23:29 AM

Look down row 17 on the perfboard. There's what looks like an un soldered wire end and a blob of solder.
You have a lot of flux build up and with the camera light glare, it's hard to tell if there isn't shorting from the solder.
Scratch a scalpel blade through between the pads to make sure it's all clear.
Can't get the schematic pdf to load so I can't tell more. IC1 voltages look right for a dual op-amp, but they may be too good, as if shorted together.

italianguy63 · December 17, 2014, 07:52:05 AM

Middle of row 24 possibly too...

thom · December 17, 2014, 09:40:43 AM

Quote from: anotherjim on December 17, 2014, 07:23:29 AM
Look down row 17 on the perfboard. There's what looks like an un soldered wire end and a blob of solder.
You have a lot of flux build up and with the camera light glare, it's hard to tell if there isn't shorting from the solder.
Scratch a scalpel blade through between the pads to make sure it's all clear.
Can't get the schematic pdf to load so I can't tell more. IC1 voltages look right for a dual op-amp, but they may be too good, as if shorted together.

Thanks, that was just a dry blob that got caught before I took the picture (well spotted, though).
Here's the schematic:

thom · December 17, 2014, 09:42:23 AM

Quote from: italianguy63 on December 17, 2014, 07:52:05 AM
Middle of row 24 possibly too...

Thanks Mark. I've checked continuity and it seems fine.

thom · December 19, 2014, 05:46:12 AM

I hate bumping my own thread, but I'm at a loss.

I've redone the grounding to have it more as a star pattern as I thought I might have created a loop with the pots, but no luck.

Bypass is clean, so it sounds like something is wrong when recording or playing back. I could really do with a set of fresh eyes here

duck_arse · December 19, 2014, 09:42:27 AM

you've left off R15? is there something funny with the led wirings?

I can't see much except flux and unsoldered ic pins.

anotherjim · December 19, 2014, 10:15:55 AM

ok, the opamp values are good.

Can't say much for the recorder chip - don't know it. Risky not to have used a socket for it!

But, it's a digital chip and there's going to be a lot of high frequency noise radiating about.
2 of the opamps have high input impedance (the 470k resistors), so those inputs will be very sensitive to noise.

Can you make the noise go away or lessen with any combination of control settings (apart from power off!).
Does the tone control change how the noise sounds?

The problem could just go away when you box it up and make sure all the pot and switch bodies are grounded by the box and the jack sleeve contacts. To make sure that happens, and if you have plastic body jacks, solder a wire to the back of a pot (assuming they aren't all plastic body too) and solder the other end to where the power supply negative wire is on the board.

Definitely (never maybe) use screened wire for the connections to the 1/4" jacks (input/output sleeve/tip) the board and the foot switch.

Reduce the length of all of the pot wires to as short as needed to just reach for assembly into the box.

thom · December 19, 2014, 11:23:50 AM

Quote from: duck_arse on December 19, 2014, 09:42:27 AM
you've left off R15? is there something funny with the led wirings?

I can't see much except flux and unsoldered ic pins.

I removed the detuning option at some point (can't remember why R15 was also removed, but with or without, no change).

I think the led wiring is OK. I did also remove the leds during testing, so no issue there.

thom · December 19, 2014, 11:29:12 AM

Quote from: anotherjim on December 19, 2014, 10:15:55 AM
ok, the opamp values are good.

Can't say much for the recorder chip - don't know it. Risky not to have used a socket for it!

But, it's a digital chip and there's going to be a lot of high frequency noise radiating about.
2 of the opamps have high input impedance (the 470k resistors), so those inputs will be very sensitive to noise.

Can you make the noise go away or lessen with any combination of control settings (apart from power off!).
Does the tone control change how the noise sounds?

The problem could just go away when you box it up and make sure all the pot and switch bodies are grounded by the box and the jack sleeve contacts. To make sure that happens, and if you have plastic body jacks, solder a wire to the back of a pot (assuming they aren't all plastic body too) and solder the other end to where the power supply negative wire is on the board.

Definitely (never maybe) use screened wire for the connections to the 1/4" jacks (input/output sleeve/tip) the board and the foot switch.

Reduce the length of all of the pot wires to as short as needed to just reach for assembly into the box.

Thanks Jim, I'll try that.
At least knowing the voltages seem ok narrows down the possibilities.

bean · December 19, 2014, 12:09:25 PM

Not having used that chip it's hard to comment intelligently. Noise level on the stock one is very acceptable....quite surprisingly. I wish I had more to offer here....I do have a demo of mine on youTube so you could listen to that to get an idea of the end-noise you should hear) although it is pretty small so it probably doesn't translate well through the audio.

As much as it sucks, you might be looking at undoing some work to put in a socket and try a different chip. Actually, I may even have an undrilled PCB left over that I could give you but I would need to check.

thom · December 19, 2014, 12:26:05 PM

Quote from: bean on December 19, 2014, 12:09:25 PM
Not having used that chip it's hard to comment intelligently. Noise level on the stock one is very acceptable....quite surprisingly. I wish I had more to offer here....I do have a demo of mine on youTube so you could listen to that to get an idea of the end-noise you should hear) although it is pretty small so it probably doesn't translate well through the audio.

As much as it sucks, you might be looking at undoing some work to put in a socket and try a different chip. Actually, I may even have an undrilled PCB left over that I could give you but I would need to check.

That's very kind of you, Brian.
I'm going to have a go at shielding (although the amount of noise I get suggests something else, I reckon) and check everything one more time over the Christmas holiday and I'll get back to you.

Cool video btw

News:

Help debugging Madbean's Loophole