Has anyone successfully built a Shin-Ei Octave Box?

[duck_arse]

two of your resistors down near the bottom right show 6 colour bands. assume we can't read them due to colour shifts in the photos - can you please tell us the colour bands and what the meter reads them as?

[t1redhands]

The six-band resistors are 100K and have been tested to be as such.
I've traced the layout against the schematic and it looks correct to me. However I'm sure there's something obvious right in front of my nose that I'm missing if you guys have had success with these schematics.
There aren't any solder or sliver bridges (or at least there better not be). Along with cleaning up the rows with clean solder tip I use a jewelers saw to get between the copper. I've tested each row and cut with a multimeter for continuity and found none.
It's gotta be something obvious at this point.

[duck_arse]

please, educate me. what brand are those resistors, what is the extra band indicating, etc? as long as I have been at this, the six-banded resistor remains a mystery.

[t1redhands]

please, educate me. what brand are those resistors, what is the extra band indicating, etc? as long as I have been at this, the six-banded resistor remains a mystery.

I don't know where I got those 6-band resistors. I mostly get my stuff from Tayda but these could be remnants of a resistor kit I got years ago. The sixth band indicates the temperature coefficient value, evidently.

[t1redhands]

Found it - it was an error in the layout. The Darlington pair wasn't connected correctly: the layout has the base of the first transistor going to the base of the second. It should instead (obviously) be the collector of the first transistor going to the base of the second. Added another cut and a jumper and voila we have a sub-octave.
Thanks everyone for all your help!

I immediately reduced the 100k resistor in series with the octave pot to 47K to get a higher suboctave signal relative to the clean.

[Slowpoke101]

Well done!    I haven't had much time today to check through the layout and now you've saved me from having to do so, thank you. I also see that you've posted an update to Dirtbox Layouts letting them know of your findings, which is always a good thing to do when you find a problem with a layout and what you did to fix it. Now all that remains is for you to do is to enjoy your new effect. Have fun.

[t1redhands]

All's well that ends well, but that was a very trying build. Photos of the finished build for posterity, featuring old DALL-E 2 art.

Thanks again, everyone.

[amptramp]

Found it - it was an error in the layout. The Darlington pair wasn't connected correctly: the layout has the base of the first transistor going to the base of the second. It should instead (obviously) be the collector of the first transistor going to the base of the second. Added another cut and a jumper and voila we have a sub-octave.
Thanks everyone for all your help!

I immediately reduced the 100k resistor in series with the octave pot to 47K to get a higher suboctave signal relative to the clean.

Don't you mean the emitter of the first transistor connected to the base of the second transistor and the collectors connected together?

[Rob Strand]

All's well that ends well, but that was a very trying build. Photos of the finished build for posterity, featuring old DALL-E 2 art.

Good result.  If you don't give up these things end up getting solved.

[t1redhands]

Don't you mean the emitter of the first transistor connected to the base of the second transistor and the collectors connected together?

I could be dyslexic, especially after staring at the schematic for several hours. This is the pair I was referring to:

[amptramp]

Don't you mean the emitter of the first transistor connected to the base of the second transistor and the collectors connected together?

I could be dyslexic, especially after staring at the schematic for several hours. This is the pair I was referring to:

That's fine, but it's not Darlington.

[Rob Strand]

It's a Schmitt trigger using 2x BJT's.   A simplified version without any resistors on the base of the second transistor.   It helps remove noise from affecting flip-flip trigger and hence improves tracking.

[Rob Strand]

Something interesting is the Schmitt trigger has quite low voltage thresholds so won't it trigger unless the input is quite strong.   That gives the unit a built-in threshold which only passes the octave signal when the input is reasonably strong.  It provides a natural aversion to fizzing out and bobbling in the tails of the signal which then gives the impression of good tracking.

(On my schematic, I'm not sure if Q11 blocks the signal before the Schmitt trigger.   It's 20 years since I looked at it in detail.)