Red Llama debugging

[Plexi]

"The third time's the charm" they said... Two Llamas, neve a trouble, until now.

It runed great a few minutes yesterday.
Today, connected again to try some mods (10M resistor in the first inverter to add gain).. and comes the troubles.
A huge amount of fuzz/gain on it: it run without any ground connection.

- Checked pcb, and no shortcuts and cold solder.
- Used a socket for the 4049UBE

Later I'll check the CMOS pins voltages.. but I guess the trouble comes from it.

[Plexi]

From 9.3v

1: 5.6
2: 2.2v
3: 2.2v
4: 2.2v
5: 2.2v
6: 0
7: 5.6v
8: 0

16: 0
15: 0
14: 5.6v
13: 0
12: 0
11: 5.6v
10: 0
9: 5.6v

[anotherjim]

Don't know exactly which inverter you have used or not used. Schematic?
Anyhow, pins 9 and 10 should never be the same voltage should they?

[marcelomd]

Hi,
Pin 1 is VDD, right? Are you using a resistor to lower voltage there? Should be 9.3V.

[Plexi]

From this schematic:


I'm using this layout:
http://1.bp.blogspot.com/-htBgOVhkQ5Q/VGp0pfVA4kI/AAAAAAAADGg/ysjjzdXhlaM/s1600/1590b%2BRed%2BLlama.png

And of course, the 1k resistor from the 9v in.
CD4049UBE https://www.diypedalgearparts.com/973-thickbox_default/cd4049ube.jpg

It sunds great a few minutes, and then the extremely high-gain fuzz, oscillation, and even some kind of octave on it.

[anotherjim]

If you have 0v on pin10 - its a spare output and it could be shorted to ground. It's an inverter, pin 9 (input) should be 0v then pin 10 should go high - same volts as pin 1.

A huge amount of fuzz/gain on it: it run without any ground connection.

It will work, but badly, with no ground on pin8 (power ground) IF any of the inverter pins go to ground - a ground connection will go around via one or more of the chips pin protection diodes in this case.

[Plexi]

Hi,
Pin 1 is VDD, right? Are you using a resistor to lower voltage there? Should be 9.3V.

You're right, asi it says in the schematic; the 1k resistor before 9v in

[Plexi]

If you have 0v on pin10 - its a spare output and it could be shorted to ground. It's an inverter, pin 9 (input) should be 0v then pin 10 should go high - same volts as pin 1.

A huge amount of fuzz/gain on it: it run without any ground connection.

It will work, but badly, with no ground on pin8 (power ground) IF any of the inverter pins go to ground - a ground connection will go around via one or more of the chips pin protection diodes in this case.

Thanks Jim, I'll check what you say closely...
I've listed the right side pins backwards, now corrected.

[Plexi]

Tried a brand new CD4049UBE (socket, of course) and the problem persist.
Replaced the 100pF and 51pF caps... and the trouble continues.
Double check 1M pot, PCB traces, continuity, and everything else.

What the h-ll.......!?

[anotherjim]

Confusing for others -  the scheme doesn't match the layout as different inverters are used!

All I can see now, and its typical of slowly changing faults, is one or more coupling capacitors could be leaky. That's the 33nF or the 68nF for the circuit. The output capacitor, if that was leaky, wouldn't be harming this circuit, but it could be causing DC blocking of whatever you have it plugged in to. Note the 10u cap negative goes toward output, or else even a good cap will leak DC.

It is always far more likely for an electrolytic cap to have a leakage fault than the film caps. Also, if you tested the chip pin voltages while it failed, one or both inverters will be at higher or lower voltages than the normal bias voltages you wrote down if either of those film caps were leaky.

If you have it plugged into a typical tube amp, then DC leaking onto the first tube grid will gradually change from constant clipping towards saturation as the grid charges too positive to control the tube. Most solid-state amps will have DC blocking capacitor on the inputs that will stop a DC leakage fault from affecting the sound, although your pedals output level control will probably sound scratchy when operated.

You can easily measure for DC on circuit signal inputs and outputs - since there never should be any at all. That should be part of your test routine with every audio circuit.

[Plexi]

Confusing for others -  the scheme doesn't match the layout as different inverters are used!

All I can see now, and its typical of slowly changing faults, is one or more coupling capacitors could be leaky. That's the 33nF or the 68nF for the circuit. The output capacitor, if that was leaky, wouldn't be harming this circuit, but it could be causing DC blocking of whatever you have it plugged in to. Note the 10u cap negative goes toward output, or else even a good cap will leak DC.

It is always far more likely for an electrolytic cap to have a leakage fault than the film caps. Also, if you tested the chip pin voltages while it failed, one or both inverters will be at higher or lower voltages than the normal bias voltages you wrote down if either of those film caps were leaky.

If you have it plugged into a typical tube amp, then DC leaking onto the first tube grid will gradually change from constant clipping towards saturation as the grid charges too positive to control the tube. Most solid-state amps will have DC blocking capacitor on the inputs that will stop a DC leakage fault from affecting the sound, although your pedals output level control will probably sound scratchy when operated.

You can easily measure for DC on circuit signal inputs and outputs - since there never should be any at all. That should be part of your test routine with every audio circuit.

Thanks Jim!

Sure.. schematic is only for general reference. I have the datasheet on hand.


All I can see now, and its typical of slowly changing faults, is one or more coupling capacitors could be leaky. That's the 33nF or the 68nF for the circuit. The output capacitor, if that was leaky, wouldn't be harming this circuit, but it could be causing DC blocking of whatever you have it plugged in to. Note the 10u cap negative goes toward output, or else even a good cap will leak DC.

Replaced all of them.
Even the resistors: there's not much components to replace here, to be honest.. better be sure. 


Edit: motorboat is gone, the ugly farting fuzz still there.