Carcass / Carcosa insane octave up

DigitalZombie

New member
I haven't seen anyone else post about it but does anyone else get really REALLY pronounced octave up on their Carcass build? I don't just mean when you've got the controls boosted, I mean ALL the time. It similar to running a Green Ringer in front of it. I kept watching video demos online but they always have the volume cranked going into a dirty amp so I figured maybe that's just how it sounded with my amp. My obsession got the better of me and I ended up ordering an actual DOD unit off eBay. It just showed up and it sounds WILDLY different. No octave, very subtle and usable tones. I figured maybe I made a mistake and just checked all of my resistors and caps and sure enough it matches the build doc precisely. I used 5089's (measuring around 500 on my meter) like the build doc suggests as well.

So yeah, before I pull apart this Carcosa that just showed up, anyone have any ideas?
 

Feral Feline

Well-known member
INTERESTING!

Please take photos and carefully document your build, an accurate trace of it would be much appreciated (maybe a part is out of spec, maybe a resistor value got mixed up, 470k instead of 4k7 etc. Whatever it is — I'd love to have my bass going through your build! It would be a good learning experience to compare and know how your pedal is doing that vs the original.
 

DigitalZombie

New member
Feral,
if you've built one could you tell me which transistors you used at what gain? Not sure what other documentation you'd want other than what I've already stated. I guess I could note Q voltages but I don't have a benchmark to compare to unless I trace this SMD factory unit.

I'll try to remember to post pictures of my board later today.
 

DigitalZombie

New member
This was before I realized the magic of cleaning the flux off the board after population, so please reserve that advice. As I mentioned I doubled-checked the printed value of every resistor against the build doc...
 

Attachments

  • 20210723_100158.jpg
    20210723_100158.jpg
    999 KB · Views: 16
  • 20210723_100347.jpg
    20210723_100347.jpg
    1.1 MB · Views: 16

Feral Feline

Well-known member
Feral,
if you've built one could you tell me which transistors you used at what gain? Not sure what other documentation you'd want other than what I've already stated. I guess I could note Q voltages but I don't have a benchmark to compare to unless I trace this SMD factory unit.

I'll try to remember to post pictures of my board later today.
It's in my last order, which hasn't arrived yet.

I mean "document" in the verb sense, and just like I said, make note of any discrepencies: If you find a faulty transistor or one installed backwards or a cold joint on one leg, if a pot was inadvertently swapped with another such as a 5k and a 500k, incorrect resistor values etc, it could be a combination of things, a "perfect storm" — I'm very interested as to why you're getting this octave up and wish to replicate it. If it's a faulty part, replication will be impossible. However, if all parts are good, it may be possible to replicate that perfect storm. Is the octave up useable?

Even if it's not, I'm still interested in what exactly is causing a "really REALLY pronounced octave up" in your build. 😺

Okay, pics... looking at the pics...

A100k Output looks like it has some cold joints, test with a DMM for continuity, if one of the pins isn't connecting properly, document which one, then reflow the solder there, and test with the DMM again. Mark it down in your trouble-shooting doc whether you resolved the continuity and whether that killed the octave up.

I try to mark everything down while I'm building, potential problems or parts substitutions. However, I find I'm never thorough enough as when I revisit the document later during troubleshooting, the stuff I trusted to memory — "oh I'll remember that, it makes sense" or whatever — with a time gap I forget what I was thinking in the moment. So while I'm notating the troubleshooting, I'm marking everything down and try to be even more thorough. Stuff that seems obvious or not necessary to mark down, often bites me in the butt later.

Back to the pics...

A5k High-Cut is bare-backed so put some tape on it or a pot cover to eliminate the possibility it's shorting.
SW looks like there's some solder trailing off the middle lug, but might be a reflection off the flux (now I get why everybody here is a stickler for a clean board). That could be shorting the switch.
C5 & C4 may be suspect, but I've melted caps much worse and they worked.
C9 looks okay on the solder side but the component-side shows the pad — worth checking the solder joint there.

Check all your resistor values, make sure they're correct.

I've got some parts that just came in to the local parcel centre, so gotta go pick those up. I'll be back on this later tonight.

Cheers,
FF
 

Feral Feline

Well-known member
Haven't had time to revisit this, after all.

Sorry, I missed in my other post that you'd already double-checked resistors.
 

Feral Feline

Well-known member
Next suggestion is to do an audio probe with it. Start at the input, then work your way through the schematic. Find where the octave up is introduced.

I'm thinking a couple of diodes can introduce an octave up, as in the Fuzzy Foxx, so might have something to do with those diodes.
 

DigitalZombie

New member
Next suggestion is to do an audio probe with it. Start at the input, then work your way through the schematic. Find where the octave up is introduced.

I'm thinking a couple of diodes can introduce an octave up, as in the Fuzzy Foxx, so might have something to do with those diodes.
If possible would you mind posting photos of your unpopulated board so I can see the traces? It would make it easier for me to audio probe the circuit.
 

DigitalZombie

New member
OK, so the octave definitely starts at Q3c. Q3 is somehow way mis-biased. I tried rebuilding on my breadboard which is usually how I can quickly identify a problem, and with the After control turned all the way down (full resistance) the voltages come out within spec if I try to calculate based on the formula for an LBP-1 (which this section basically is). So I should be getting over 8 volts at Q3c but I'm only getting 4.5. Also 5v at Q3b… what?

I also don't understand why VBase changes when the After control is turned up though. From what I've learned that's determined by the two biasing resistors, (Vin x R2) / (R1 + R2). The resistance off VEmitter isn't factored in, so why that changes is beyond my understanding.
 

Attachments

  • Screen Shot 2021-07-26 at 1.18.04 PM.png
    Screen Shot 2021-07-26 at 1.18.04 PM.png
    22.9 KB · Views: 3

DigitalZombie

New member
Figured it out! The 220k bias resistor had a bad solder joint, so it wasn't connecting to ground. Confirmed when I pulled the 220 off my breadboard it's octave city. I would recommend this as a mod - maybe put it on a toggle, or even a 2nd footswitch to disconnect it from ground.
 

Feral Feline

Well-known member
Bravo! Well done.

I will definitely play around with R14, thank you so much for that info!

I'll play around with some values, might be better to find the minimum ohmage it takes to get the squeal, don't want to cause unnecessary damage to anything. So...
Maybe somebody more knowledgeable can weigh in and tell me why it's a seriously (or not) bad idea to have R14 220k on a switch.
 

Bricksnbeatles

Well-known member
Bravo! Well done.

I will definitely play around with R14, thank you so much for that info!

I'll play around with some values, might be better to find the minimum ohmage it takes to get the squeal, don't want to cause unnecessary damage to anything. So...
Maybe somebody more knowledgeable can weigh in and tell me why it's a seriously (or not) bad idea to have R14 220k on a switch.
Careful— if you put a 110k resistor in, it’ll end up being a tritone up ;)
 
Top