I just finished 2 shin-ei octave boxes using the schematics at http://topopiccione.atspace.com/PJ15ShinEiOctave.html.  For the transistors I used 2n5210 for the NPN's and a 2n3906 for the PNP.   I used 1% metal films and poly film caps.  Each of these were done point to point by hand.  Once I fired them up I found one working well one not so well, ie. it was quiet and not well defined.

I got out the scope and waveform generator and found out that I had a perfect signal(when compared to the working one) up until the octave volume control potentiometer(p2).  Well it turns out that on the WORKING octave box I had NOT soldered C11(100n or .1uF) to ground.  On the Quiet and less defined octave box, it WAS soldered.  Once I removed C11 from ground BAM, beautiful octave down, with a volume out of almost twice the input.  All of the controls work perfectly.  SW1 adds the original signal to the octave and can make for some amazing sounds.
My electronoics knowledge is limited, so if anyone can explain why removing c11 helped, I'd like to know.  All I can say is thank god for mistakes. 

As with how it sounds, its phenominal.  I have only used this with bass guitar so far, but its definitely going to become a part of my rig.  The octave down can be dialed in to be sublte and it just basically fattens everything up with an extra overtone and is great for getting funky on a solo or lead like parts.  It can get extreme when cranked, to the point where I turn the bass eq down all the way on the amp and guitar and still get a monsterous low end.  The tracking is so much better than a blue box and its cleaner than it too.  The tracking responds each time you attack its rarely all over the place unless you are.  Thats one thing I have noticed with most analog octaves is that you have to exercise alot of control to get it just right.  Thats why I dig this octave box alot it follows you, not you having to follow it. 

C11 to ground will basically make the signal quiet and muffled. It's a cap to ground just as if you had a tone control turned all the way down with a large cap as the roll off.  Cool looking project. I'll have to come back to that one.

John

Fox,
i tried your mod on my Octavebox clone and, yes, the octave looks very angry and powerful, but I lost the clean signal mixing.
Nice mod, but I prefer a muddy octave!

What you might want to try is a.
.047uf or lower value cap for c11
maybe a value between .1uf and none might be what you want or maybe none is more fun.
Like John posted it is a cap in parallel with the fixed 1k and pot that forms a varing lowpass/limit

First off thank you everyone for replying and trying this. 

Zero: I dunno how you lost the mixing when you remove the .1uf(c11) from ground.  I still have it on both my boards and its loud and clear with  or without C11 in the correct position.  Check to see if you removed something else by accident.

  I do agree with Gus though.  A cap with a lower value is an all around better solution.  A .047uf worked to remove some of high end fuzziness that is present when C11 is not in place.  To me anything much higher started to muffle it again.  Maybe its just me.  One thing I do notice is that with the .1uf C11 IN PLACE the octave will not be as present as with a lower value cap.  But if you go with a lower value cap(or no cap at all) you will have a louder octave, but you can use the octave volume control to Dial its presence back, making it sound the same as if the .1uf was in its place.   

I think this is a killer pedal and sounds WICKED before a mutron-III, giving you monsterously funky synth like sound especially when using bass guitar.  I'll try to post a few sounds sometime soon. 

What type of diodes did you use? I have some 1n34a's will those work?

My sense is that the "original factory schematic" is simply in error.  A value of 100n gives such a low treble rolloff that it is a wonder that anything is audible at all, particularly since the filtering effect gets more intense as the volume goes below maximum.  A value of 10n or maybe even less would seem to be in order. 

You will note that C13 and C14 (both 100nf) help to pre-condition the signal for cleaner detection/tracking by providing a 12db/oct rolloff starting around 234hz.  Now, bearing in mind that this is not what you'd call a "brick wall" type filter (in the grand scheme of things, 12db/oct is not THAT steep), still some might want to drop those caps down to .047uf (47nf) or .068uf to provide a little more sensitivity to higher notes.

You're the first one I've heard of that's built one of these, so I was glad to hear about it. It's a fairly labor intensive circuit, and I wanted to hear about it being prooved before I attempted it. The Blue Box (at least mine) goes insane if I try to play it through my tube amp. It tracks decently through my SS Crate. So far, the best tracking DIY has been the Shocktave. I can't remember, is the the Shin Ei a 1 and 2 octave down both, or just a single or a double octave down?

Audioguy:  yes I used 1n34's and they worked great.  Be sure to match them, see http://www.diystompboxes.com/smfforum/index.php?topic=51151.msg383356#msg383356

In general do your best to keep everything in that section with D1, D2, Q9, Q10 matched, it will produce a better result.  I used and measure 1% metal films myself.

Good idea hammer.  I made one for bass and one for guitar, I'll have to tweak the guitar one this week to see if it makes a difference.

Well, I just finished building this box; however I wasn't able to get the octave down to work. The orginal guitar signal seems to come through pretty strongly, but there is no response when switching on the octave. I've tested each component to check if anything was grounded, but it all seems ok. I've also checked if voltage was getting to each component. Any ideas on where I went wrong?

Here's all of the voltage values for transistors and diodes to add to my earlier post.

input voltage: 8.6 V

Q1:
E = 0V
B = .625V
C = 3.46V

Q2:
E = 0V
B = .625V
C = 3.53V

Q3:
E = 0V
B = .640V
C = .640V

Q4:
E = 8.51V
B = 8.50 V
C = 0V

Q5:
E = 0V
B = 0V
C = 8.5V

Q6:
E = 0V
B = .1V
C = 8V

Q7:
E = .14V
B = .79V
C = .16V

Q8:
E = .14V
B = .16V
C = 8.47V

Q9:
E = 0V
B = .19V
C = 7.72V

Q10:
E = 0V
B = .63V
C = .03V

Q11:
E = .22V
B = .85V
C = 3.7V

D1:
A = .19V
K = 7.28V

D2:
A = .627V
K = .58V

D3:
A = .027V
K = .033V

Quote from: brown5629 on July 05, 2007, 09:24:58 PM
Here's all of the voltage values for transistors and diodes to add to my earlier post.

input voltage: 8.6 V

Q1:
E = 0V
B = .625V
C = 3.46V

Q2:
E = 0V
B = .625V
C = 3.53V

Q3:
E = 0V
B = .640V
C = .640V

Q4:
E = 8.51V
B = 8.50 V
C = 0V

Q5:
E = 0V
B = 0V
C = 8.5V

Q6:
E = 0V
B = .1V
C = 8V

Q7:
E = .14V
B = .79V
C = .16V

Q8:
E = .14V
B = .16V
C = 8.47V

Q9:
E = 0V
B = .19V
C = 7.72V

Q10:
E = 0V
B = .63V
C = .03V

Q11:
E = .22V
B = .85V
C = 3.7V

D1:
A = .19V
K = 7.28V

D2:
A = .627V
K = .58V

D3:
A = .027V
K = .033V

I would check the circuit around Q5...it shouldnt have 0V at the emitter AND the base.

Thanks for the help. I checked out the circuit around Q5, and it turned out that C9 was dropping all of the voltage across it. I took the capactior out of the circuit and discharged it. It looks like that would have done the trick. I put it back in the circuit and .6V was now present in the base of Q5. However, the circuit still does nothing. Am I missing something? Should I replace C9 or Q5 or both? What is the actual voltage drop supposed to be at the capacitor?

Edit: I meant to say that the C9 cap was still 0V after removing it. I replaced it with a new cap, but with no changes.

bump. I would really appreciate any help you can give. Does anyone have the true voltage values for this circuit? Thanks again.

Anyone?? Please!

Sorry for bumping the thread so many times. I rechecked the voltage values I posted. The collector for Q3 is actually 3.5. Can someone post the voltage values for each of the transistors? I know someone's done this pedal before. I'm guessing my problem lies in that area between Q4 and Q5, which contains Q6-10.

Q7 is not working properly: collector voltage is too low. I would check R25 and its connection to +9V.
Even collector voltage on Q10 is very low: check R31, R33 and R34.
As suggested before, Q5 got both emitter and base grounded. I would check for short cricuit and / or substitute this transistor.

I just built one of these myself. I added this C11 mod as a optional tone control using a 100n cap in series with a 500k pot (in place of just C11). It works great and seems to really add a lot of versatility when you switch between single and dual coil pickups. With the pass-thru switch option and the tone control - along with some careful Q10/Q9 matching - this is an awesome octave down - the best I have heard.

One thing, if anyone else wants to put in this tone-control mod - a 100kA pot would likely work best. The 500kB pot I'm using is a little tight on the useful range for this mod. I'll try a 100kA and report what I end up with when I have time to tinker with this. For now, I love it the way it is.

Thanks Alex (Zero) and Paul - Topopiccione dudes!!!

And also thanks to everyone for contributing to this thread - I really like this build with these mods.