Boss OC-2/3 Clone getting faint static crackle and thinking the schematic is bad

[bushidov]

Well, I did the C25/C26 mod, but I think you meant C26/C27, but I got what you were saying. Take C26's negative lead and pull it to 0V instead of the divider and shrink C27 down to a 10uF.

Unfortunately, that didn't fix it either.

I removed those suspected VBIAS connections. I removed C23, R33, R35, and R39.

Well, now the crackle is completely gone, which is good. But now I don't get any sub-octave or 2 sub-octave. On either of the pots, I get what sounds like a low-pass filter with a low cut point on the oct 1, and an even lower low-pass filter with a lower low cut point on oct 2.

If I turn the blend knob to 0 and run on either of the OCT knobs up, it just sounds like my regular note, but very muffled and extremely not-bright.

Did that shake something loose?

[bushidov]

So, I put R35 and R39 back in. No crackle, but no octave either.

I added R33 back in. Crackle returned, but so did my octaves.

I put C26 and C27 back the way they were originally. Crackle remained, and so did my octaves.

So, it appears that R33 has something to do with this. With the resistor in place, I get my octaves, but I also get crackle. With the resistor pulled, my crackle is gone, but so are my octaves (I get super low pass filters on the octaves, but not actual octave down notes)

Any ideas?

[Rob Strand]

I'll have to get back to it after I've slept a few hours.

Sorry about the cap designator stuff-up.

So, it appears that R33 has something to do with this. With the resistor in place, I get my octaves, but I also get crackle. With the resistor pulled, my crackle is gone, but so are my octaves (I get super low pass filters on the octaves, but not actual octave down notes)

I suspect what's going on here is withouty R33 it kills that part of the circuit.  No pulses to the 4013 so no crackle.

So, I put R35 and R39 back in. No crackle, but no octave either.

I added R33 back in. Crackle returned, but so did my octaves.

So R33 is necessary.

You could try R33 in but R35 R39 out.    It's a long shot.

So if no progress on that put everything back to normal and try either pulling the 4027, or cut the track to 4027 pin 13 then connect it to ground.

The idea here is to see if it's the 4013 or the 4027.

[bushidov]

The idea here is to see if it's the 4013 or the 4027.

Well, the ICs are all socketed, so when I pull out the CD4027, it still crackles. If I put the CD4027 back in and pull out the CD4013, the crackle goes away.

[Slowpoke101]

Just a thought...
Short out R46 so that the 4013 gets full supply voltage and see what happens.

[bushidov]

Short out R46 so that the 4013 gets full supply voltage and see what happens.

Just tried that. Didn't make a difference. Not a bad thought, though.

[Rob Strand]

    Well, the ICs are all socketed, so when I pull out the CD4027, it still crackles. If I put the CD4027 back in and pull out the CD4013, the crackle goes away.

    Quote
        Short out R46 so that the 4013 gets full supply voltage and see what happens.

    Just tried that. Didn't make a difference. Not a bad thought, though

Well at least all that narrows things down.   It's an interesting result just the same.

The biggest dilemma of all this is:
- It's seems to be related to the 4013.
- Crackle is getting into the audio when the pots are on zero.
  The way that could happen is through +9V, VBIAS, or a layout issue
- but:  the 4013 has it's own power and shouldn't affect the power rails or VBIAS.

So an easy thing to try is to replace the 4013.   It's a bit of a cop-out but it is easy to try.

Another thing I noticed is that some other designs, including the OC-2, use higher valued resistors for
R37 and R41.   I'm seeing 22k to 47k,  not 1k.   Maybe you could try changing those.

As for things you could check:
- You could check is the DC output on U2A and U2B.   Perhaps with no signal and with signal.
- You could connect a cap to the output of the input buffer U5A and see if you can hear the crackle there
as well.

Debugging from cause and effect is starting to run out of options.
An oscilloscope would certainly help.

[duck_arse]

Are you sure you have the vdd and ground pins of both the 4013 and the 4027 connected to the power and ground?

Verified

Perhaps the next step is to check the voltage on the power rail for the 4013.

Voltage Supply reads 8.93V
Pin 7 of the CD4013BE reads 8.90V

pin 7 at what volts?

I've been following this thread, but might have missed where you posted all the pin voltages for the two flipflops ...... is there a chance that those 1M resistors at R42 and R43 are too big? I've had pull problems w/ cmos and 1M's before.

[Rob Strand]

pin 7 at what volts?

Good catch.   Maybe need to confirm voltage on both pin14 and pin7.

[bushidov]

Good catch.   Maybe need to confirm voltage on both pin14 and pin7.

Is there a chance that those 1M resistors at R42 and R43 are too big?

I could try swapping them out. What value do you suggest?

So an easy thing to try is to replace the 4013.

Crap, I'd need to order another. That's not a chip I keep a lot of (maybe I need to be?)

R37 and R41. I'm seeing 22K to 47K,  not 1K. Maybe you could try changing those.

It wouldn't be the first time I copied and pasted something in a schematic and forgot to change the values afterwards. I'll give that a shot.

You could connect a cap to the output of the input buffer U5A and see if you can hear the crackle there
as well.

Can't seem to hear the crackle here.

- You could check is the DC output on U2A and U2B.   Perhaps with no signal and with signal.

No Signal
U2A Output Pin 1: 0.007V
U2B Output Pin 7: 7.64V

With Signal (hard strum of all chords)
U2A Output Pin 1: Starts at a couple volts (2.5V-ish) but then as it "crackles" it decreases till it falls back to 0.007V. Basically for decay of signal.
U2B Output Pin 7: Starts at a few volts (5.5V-ish) but then as it "crackles" it increases till it hits to 7.64V. Basically for decay of signal.

Did that help?

[Marcos - Munky]

Nothing to add to help you, I just want to say that's the best way to show voltage measurement I've ever saw. Why don't people thought on that before?

Hope you get this one fixed. I'm keeping an eye on this topic.

[Rob Strand]

Crap, I'd need to order another. That's not a chip I keep a lot of (maybe I need to be?)

They are pretty common.  It depends what you want to play with.   Like if you play with octaves it will come in handy.

No Signal
U2A Output Pin 1: 0.007V
U2B Output Pin 7: 7.64V

With Signal (hard strum of all chords)
U2A Output Pin 1: Starts at a couple volts (2.5V-ish) but then as it "crackles" it decreases till it falls back to 0.007V. Basically for decay of signal.
U2B Output Pin 7: Starts at a few volts (5.5V-ish) but then as it "crackles" it increases till it hits to 7.64V. Basically for decay of signal.

Did that help?

Hmmm,  it was very useful because it forced me to look at that rectifier + comparator circuit a bit more.    That's given me a big headache which way to go!

As to how the crackles are getting in.   I still don't know.   I'm starting to think it's a layout issue.   However it could be more than that:  It could be the circuit is misbehaving and exposing a layout issue which might not be a problem if everything was working fine.

So up front there's a problem with the rectifier + comparator circuit (all of U2 upto the U3A).    Your circuit is more of a Behringer UO100 than an Boss Octave 2.   It's fairly clear the Behringer UO100 is copied from the EBS Octabass.   However, to me it looks like they stuffed up!   They took the EBS Octabass rectifier circuit and changed the flip-flop (U3 on your circuit) from a 74HC74 to a 4013.   The 74HC74 which has negative going Set and Reset inputs.   The 4013 however has positive going Set and Reset inputs.  It appears they did not rewire the opamps U2A and U2B to compensate.  That means the Set and Reset signals to the 4013 are the wrong polarity.  Next it looks like they made C23 bigger to patch-up the problem; Behringer UO100 is 100n and the EBS Octabass is 10n.     The 4013 Set input overrides the Reset input so having the inputs flipped will in fact screw up the way the circuit works.

I'm assuming the EBS Octabass is correct.  We don't know that but by looking at the Boss OC2 we can see two comparator opamps are at least consistent in that the comparator inputs are flipped to compensate for the differences between the 4013 and 74HC74.    Have a look at polarity of the comparator opamps on the Boss OC2, the EBS Octabass and the Behringer UO100.  We expect the Boss OC2 and EBS Octabass to be wired different because they use different flip-flop chips.   However the Behringer UO100 is wired like but with a 4013 like the Boss OC2, making it wrong.

I'm kind of assuming this stuff-up is promoting the problem.  I could be totally wrong.   On one hand we could argue Behringer got it to work (maybe) on the other hand we could argue the EBS Octabass and Boss OC2 make a lot more sense.

For sanity's sake, and the fact it's getting hard to debug this without an oscilloscope,  I think it would be very worthwhile lifting the C24 side of R36 and lifting the C25 side of R40, then,  wire R36 to C25 and wire R40 to C24.   That effectively makes the signals at the input of the 4013 correct.

See what happens with that change.

Next change C23 to 10n (EBS value) and see what that does.

If that's still no good, try flipping the connections of U2Aand U2B to R37 and R41.   Off hand I don't think it matters but with those flipped it should be closer to the Boss OC2.

Please convince yourself of what I have said is true (this tricky problem isn't the best thing for my tired little brain today  ).

[duck_arse]

me again. why isn't pin 4 at 0.000 volts? maybe check the resistance to ground on that pin.

[Rob Strand]

I'm still puzzled how the crackle is getting in since some the previous tests discount the most common causes.

So one layout issue which might not help is as follows:
If the R20 is not located near U3B then the track between R20 and U3B could couple crackle into the audio, especially if it passes by more sensitive audio tracks.   Similar for the track between R22 and U4B.  If R20 and R22 are placed so they are near the digital IC's it could help reduce the track to track coupling.

FWIW, I know you traced the UO300 but I have seen the UO100 schematic and it seems to have the same problems in the area of the rectifiers and comparators.  In other words I don't think it's a tracing issue.

[plague.doc]

Hello and thanks to you all for all the replies! I've got pretty much the same issue with that project http://vulcanofx.blogspot.com/2014/09/chopped-boss-oc-2.html
I'll check all solutions you guys mentioned earlier soon, but maybe you have any progress on that problem now? Thanks!