Strange problem with XOR gate fuzz

[John Lyons]

I've been working on trying to debug this fuzz and I found something strange.
When I put in a 0.1 cap on pin one as on the schem there is no gating and a full over the top gain is present, background noise when I'm not playing, which is to be expected with any fuzz at full volume.
After about a minute of playing the gating starts and there is not much sustain and a pretty hard on off gating. 
When I replace with the same value cap the sound comes back, no gating. Thwen after a minute or so the gating cleeps back. I have to let the caps sit for a few minutes before they "work" again.

Any ideas on this strangeness ?

I replaced the XOR chip and get the same results. Different value and material caps have been tried with the same results

Here's the schematic and some notes and a sound clip     http://www.mrdwab.com/john/XORUS.html

[Paul Perry (Frostwave)]

Well, pin 1 doesn't have any direct connection to a fixed bias point, so I guess it is free to wander around (influenced doubtless by input signals overloadign the input).
Whenever digital based stuff has this 'delayed gate' thing, it's due to either a pin being uncomnnected & the voltage floating around, or lack of stable bias, in my experience.

[ethrbunny]

Is this the 'xor ultra sick'? Heh. I still have mine unboxed and hoping for life. Maybe its time to dig it up and scare the kids again...

[John Lyons]

Ok, so I should look for some weak link in the bias. Looking the the schematic, the 4047 gets full 9v power at pin 14 and is connected straight to ground at 2 and 7.  So those would be the ones to check concerning bias?

This thing is killing me. I've had a hard time getting this thing to behave and be stable.

Yes ethrbunny. the XORUS as you said. Man, want to get this thing in a box and move on!

John

[John Lyons]

I checked the voltage and at pin 1 I get 9.5v. Is there supposed to be voltage at any of the other pins?
When the cap seems to stabilize and the gating starts I can remove ground from pin 7 and the gating stops. When a new cap is put in and no gating. remove pin 7 from ground and it dies...(gating) !!!

There has got to be something happening when I put the new cap in to remove the gating.
Any other things to check?
As far as bias, there are 2 direct conections to ground pins 2 and 7, and one connection to 9V pin 14. What else coul be affecting this?

Thanks folks

John

[Paul Perry (Frostwave)]

I can't guarantee anything, because I never built it, but if it was on my bench I'd put a 100k(or greater) pot across from + supply to ground, and run the wiper to pin 1. Then I'd tweak the pot & see if there was a position that made it OK. That's my best shot, anyway.

[John Lyons]

Holy smokes! It works!
I tried a 1meg pot and it worked at one tiny section of the rotation. Tried an 250K, 100K and a 25K . Seems that the voltage divider ratio is what does it because each pot only worked at a tiny place in the rotation... On the 25K pot it was a 14K/11K divider. I'll measure the other post when it I dial them in and see what the values are. Then try to find those exact value resistors or combonations to make the right values.

So, was is that the voltage was unstable somehow.I havent done any digital logic studying but I've never heard of applying voltage to the input pin. The original circuit was Tim Escobedo's Digital Octaver Fuzz and this doen't have any voltage referenc at pin 1. But then there is a lot more going on in this circuit.

Here's a link to the chip pinout. http://en.wikipedia.org/wiki/XOR_gate

Thanks so much Peter!

John

[Paul Perry (Frostwave)]

It's strange, I have more success guessing how to fix other people's circuits, than working on my own!
Maybe it is because, if it is somebody on the net, then I have to THINK HARD before posting, instead of just grabbing the iron & wrecking something else as well. Just a bit of philosophy there, folks!

But the moral of this particular episode remains: an input pin to a CMOS chip, if left free to float around (that is, with no direct or resistive connection to ground or +) will eventually float to somewhere you don't want, pick up hum, radio stations, static fields etc. And drive you mad.

[ethrbunny]

So that killed the gating?

[John Lyons]

Yes, The gating is gone. The pot/resistive divider mentioned above works on the ratio of the divider I believe. So any pot will work, connect the power supply to one outer lug, ground to the other and wiper to the first pin of the XOR chip. There was some hum and odd noise at certain spotsof the rotation, but you have a little wiggle room to adjust the ration of the divider/pot.
Adjusting for a small amount of gate was good because the background noise gets cut off.
I was hoping for fewer knobs on this beast. Now I have a "gate" knob. I'm sure I make it a trim pot...

This may be the reson Jeff Musser had problems getting his crazy sounds once and not again. If the circuit was unstable (as the case with my dying caps above) then at least now it's more stable and you can get some consistancy with it.  I might try smaller value pots to tighten up the range of sweet spots, some of the ranges on the knobs swing are too wide. You have to turn so far to get a change...

I'm happy this beast is at least in line enough to get things sussed out.

All hail, PETER!

John

[ethrbunny]

Excellent. Let me know if you get it down to a resistor value.

I still have two knobs that don't seem to effect the sound v much. Of the four on the original 'ultra sick' how many do you still have?

[John Lyons]

Right now I have all the original 4 pots and a tone control.(Search "Stupidly Wonderful Tone Control") Since this circuit has the ability to get so out of control with the Hi Frequency fuzz and strange fizzes etc. I put in a tone knob to cut some of the razor edge on some settings.
And now I have the "Bias" pot.

So I have 7 knobs total. I'll make the bias a trim pot on the board since it will get the least use. Maybe a switch to select gate or no gate.

With the bias pot just connect what you have and measure i to get the resistor valuest. Then put in those values. The pot makes a resistive divider between the power supply voltage and ground. With the junction of the resistors connected to Pin 1.

Some of the controls are subtle. The thing is that they all play off each other. At one setting any certain pot may not do much. But at another setting the pot will have a good bit of change on sound. Make sure your pots are all wired correctly. I found a couple mistakes on mine initially. All the pots wipers are connected to each other. There are som pot to pot connections that don't go to the board that are tricky and I missed them the first couple times on the schematic.

[ethrbunny]

Would my pin-1 bias resistance be different from yours? Can you tell me what it should be?

[John Lyons]

I would just hook up a pot and set it for your circuit. I would think the Bias would be the same but you might want a little gating as described above.

Here's how:
Get a 100k or which ever pot you have around.
Connect one side to ground and the other to the 9V source.
Connect the wiper/middle lug to the pin 1 of the chip. (4070)
Plug in you guitar and have the amp on, volume onj guitar up but not playing it. 
Turn the pot slowly until you hear the gating stop you get the loudest sound. Turn back and forth to find the best sounding spot on the pot.
Disconnect power to the XOR and measure the resistance from one side to the middle lug of the test pot.
Then measure the middle to the other side.
Find similar resistors and solder these at the same points of the circuit as the test pot. Or keep the pot as an adjustment. Or use a trim pot of roughly the same value and solder it to your board.

There are a few usable places that the pot sounds "good" just try this and pick which position sounds good to you.

John

[ethrbunny]

I just went back and re-read this in prep for doing it.

This gating test is done with *no* sound running through the board?

[John Lyons]

You have to be plugged in and playing through the pedal. Have the amp and guitar turned up but not playing it. When you dial in the pot mentioned above, there will be a place in the pots rotation where the volume will be the loudest. If yours gates now you will hear the place where the gating stops because the volume will turn on al of a sudden. I set mine just below the loudest setting because it was a little too crazy with hum and buzz! WHen you find the place you like then measure the pot (as above) or make it a trim pot so you can adjust it later. If you want the gating effect you could putin a switch to cou out the pot.

[ethrbunny]

If yours gates now you will hear the place where the gating stops because the volume will turn on al of a sudden

Yah - mine gates now fer sure. Volume of what? N/m. Ill just do it and see if it makes more sense afterwards. Fun!

[John Lyons]

When the circuit gates it's like a switch....when you get the pot/voltage divider set right the gate opens up and you can wave bye bye to the gating effect that the circuit had.  So when you get the pot set right there will be no gate and the pedal volume will always be on and not gated...

John

[ethrbunny]

Okies. I set this up (after fixing a mistake).

I found the spot where the volume turns on like you said. Have you set yours to somewhere near this now?

[John Lyons]

I'm working on a PCB layout of it right now. I don't feel like fooling with making it on perfboard, too many connections to the 4070 chip.
I'll be including a trimmer pot to bias the input as described above.

Right now I still have the circuit on a breadboard with a pot connected for the input bias.

John