Has anyone successfully built a Shin-Ei Octave Box?

[t1redhands]

Hello!

I've been tackling making the Shin-Ei Octave Box OB28 for a little bit now without much success. Has anyone ever made a properly operating OB using these existing schematics:

Stellan Lehrberg's: https://www.tremolo.pl/Efekty/Octaver/Octave%20Box.gif
From the inside of the back lid (I think): https://postimg.cc/Sn34YFFJ
Compare to the schematic of the Colorsound Octivider by Rob Strand: https://postimg.cc/QB3cwmK3

My main question is this: what's the deal with the two 100k resistors in series? My circuit works well enough until this point. The volume drops significantly by the time it gets to the base of that transistor to the point of being nearly inaudible by the time it gets to the Octave Volume pot.





Is the problem that the gain stage leading to the Octave Volume pot doesn't have enough gain, or something else? Is it the two 100K resistors in series along with the 47K to ground?

Thanks!

[noisegrrrl]

My main question is this: what's the deal with the two 100k resistors in series? My circuit works well enough until this point. The volume drops significantly by the time it gets to the base of that transistor to the point of being nearly inaudible by the time it gets to the Octave Volume pot.

It looks like a voltage divider (https://www.wikiwand.com/en/Voltage_divider) which would explain the volume drop at that point. In this configuration it should be dividing the Voltage by 2.
Looking at the circuit though, it looks like it's being fed back into the circuit?

[pacealot]

I have successfully built an Octivider and had no problems at that stage (and I also tried the Shin-Ei and other similar cap values with no changes in efficacy). My limited, non-EE, perspective makes me suspect something fishy around the PNP transistor that precedes it (pinout? is it PNP as it should be? etc).

I imagine our actual EE-level-type friends around here might like to see voltages, especially from that transistor stage and the following one.

[Mark Hammer]

I made one, using this layout - http://effectslayouts.blogspot.com/2015/09/shin-ei-ob-28-octave-box.html - and it works great.

[Rob Strand]

I have successfully built an Octivider and had no problems at that stage (and I also tried the Shin-Ei and other similar cap values with no changes in efficacy). My limited, non-EE, perspective makes me suspect something fishy around the PNP transistor that precedes it (pinout? is it PNP as it should be? etc).

I imagine our actual EE-level-type friends around here might like to see voltages, especially from that transistor stage and the following one.

The voltages don't help so much around the colored area because the transistor are operating unbiased; sort of filtered switches.   Having traced that circuit independently photos and the fact all the schematics aren't far off from each other,. and also the layout Mark posted follows the same circuit, I think it's safe to say the circuit is like it is.

The question is, is the level drop abnormal?  I don't know.  I suspect there will be some level due to the size of the resistors.

It could be something simple like a mistake with the transistor pinout.   Check the pinouts on both the PNP and NPN as they might not be the same.

[PRR]

two 100k resistors in series? My circuit works well enough until this point. The volume drops significantly by the time it gets to the base of that transistor

Well, it has to drop a lot, because one side can swing 9 volts and the transistor base conducts hard at 0.6V.

However by guitar standards a 0.9V signal is not "inaudible". A builder error is possible.

6.8uFd there is pretty brutal. Is that for real?

[Rob Strand]

6.8uFd there is pretty brutal. Is that for real?

That's the problem my schematic shows 6n8 and so does the back of the box schematic!!!

[pacealot]

The schematic I used had 6.8nF there, and I think I changed it to 5.6nF just to make it track better a little higher up the neck. If OP's got 6.8µF there, that would certainly affect things!

[t1redhands]

With a signal going I have about 0.2V at the base of that transistor immediately prior to the "octave volume" pot.
I have 6n8 from the base to the collector of that transistor, not 6u8. That would be brutal for sure.

Builder error is entirely possible. I'm currently making an attempt at this layout: https://dirtboxlayouts.blogspot.com/2021/06/shin-ei-ob-28-octave-box.html

I've tried the EffectsLayouts one with similar results (very very quiet octave signal).
I haven't tried changing any values of the voltage-dividing resistors, but I've tested all of them and they're correct (and working, had to replace a transistor). The Dirtbox layout isn't verified but I've gone over it several times and it all seems to add up. Totally could be missing something, however.

[Rob Strand]

I've tried the EffectsLayouts one with similar results (very very quiet octave signal).
I haven't tried changing any values of the voltage-dividing resistors, but I've tested all of them and they're correct (and working, had to replace a transistor). The Dirtbox layout isn't verified but I've gone over it several times and it all seems to add up. Totally could be missing something, however.

Can you measure the transistor gain based on the pinout you are using for the build?   You never know.  Also keep in mind there are fake parts out there and there are suffix letters which can change the pinouts of some parts from the common parts (very annoying unless you check).

[Slowpoke101]

I've tried the EffectsLayouts one with similar results (very very quiet octave signal).

Did you read my posted notes near end of that layout posting. I did find an error with the circuit board. Also wiring the foot-switches is rather confusing. But the effect does work quite well.
Also note - the 6u8 capacitors are 6n8. It's just a misprint. I bread-boarded the entire circuit in 2020 before I built the PCB version just to confirm that the thing would work.

[Rob Strand]

Did you read my posted notes near end of that layout posting.

Can you post the link?

Ah I see it's your notes in the comments on Mark's link:

Slowpoke101April 13, 2020 at 4:51 PM

I bread-boarded the circuit. Used 6n8 (6.8nF) capacitors - the 6u8 is a misprint. Diodes used were 1N4148. The circuit works but I have not made the PCB yet. From your google-drive I used your wiring diagram but I found one problem. There is supposed to be a wire link on SwB linking terminals 2 and 4.
Reply
Replies

    Slowpoke101July 29, 2020 at 5:52 PM

    There is an error with both the perf and PCB versions. A 100K resistor that should be connected between Q9 collector and Q10 base is missing. The effect works correctly when a 100K resistor is installed and a wire link between terminals 2 and 4 of SwB is also installed.

Did Mark do the fixes to the layout?

[Slowpoke101]

Slowpoke101July 29, 2020 at 5:52 PM

    There is an error with both the perf and PCB versions. A 100K resistor that should be connected between Q9 collector and Q10 base is missing. The effect works correctly when a 100K resistor is installed and a wire link between terminals 2 and 4 of SwB is also installed.

Did Mark do the fixes to the layout?

[/quote]

Unfortunately no. Mark hasn't updated the project and the project remains unverified. But both Mark Hammer and myself have built working versions. I guess that other people have also built it as well.

[Mark Hammer]

I built mine a while back, and can't really recall if I incorporated any changes that were noted in the comments.  All I can say is that it booted up just fine, first time out.  That may or may not have been a result of component-value changes, but the actual layout was NOT tinkered with in any manner.

[duck_arse]

Builder error is entirely possible. I'm currently making an attempt at this layout: https://dirtboxlayouts.blogspot.com/2021/06/shin-ei-ob-28-octave-box.html

I've tried the EffectsLayouts one with similar results (very very quiet octave signal).

everyone can guess what I'm going to say next, can't they?


why not show us your build? post photos, please.

voltages. please.

as the D1 diode is going to gate the signal between those two 100k resistors, why not try lifting the diode, see if you get full volume like you expect. that will/should then clear the two transistors before the bass vol pot.

[Rob Strand]

I built mine a while back, and can't really recall if I incorporated any changes that were noted in the comments.  All I can say is that it booted up just fine, first time out.  That may or may not have been a result of component-value changes, but the actual layout was NOT tinkered with in any manner.

I'm pretty sure Slowpoke101's fixes are correct but I'm not surprised if some circuits still work with minor bugs.  Perhaps they don't track as well on some guitars - that happens with these types of devices, especially around dead spots in the neck.   I'd really need to look at the fine details to understand what the differences are.

[Slowpoke101]

My fixes only apply to the Effectlayouts project. The missing 100K resistor is in the flip-flop (divider) stage and without it the sub-octave effect cannot work.
The missing link refers to the foot-switch diagram that was mentioned in one of the other comments on that project. That diagram was found on the google-drive belonging to the person who made that comment.

I'm fairly certain that the circuit diagram shown in this thread is correct. It looks exactly the same as the one that I bread-boarded and I know that it works.

Rob - When you asked "Did Mark do the fixes to the layout?", I thought that you were referring to the fellow that runs the Effectslayouts website and not Mark Hammer (Sorry Mark). Mark would have had to have done something about that 100K resistor as the divider has to have it in order to work. Also the foot-switches' wiring is easily determined from the circuit diagram.

Like any octave divider, this one seems to work better when you play a bit higher up the neck - above the 5th fret at a minimum. On a Strat, use the neck pickup and on an LP style use rhythm pickup. You have to fiddle with the tone and volume levels for best results.

[Rob Strand]

Rob - When you asked "Did Mark do the fixes to the layout?", I thought that you were referring to the fellow that runs the Effectslayouts website and not Mark Hammer (Sorry Mark). Mark would have had to have done something about that 100K resistor as the divider has to have it in order to work. Also the foot-switches' wiring is easily determined from the circuit diagram.

Sorry I'm a little out of touch these days with the who's who.

[t1redhands]

Thanks for taking a look at this, guys.

Still no success. I found that the PNP transistor was faulty and replaced it. Removed the diode but the signal was still very quiet.
Voltage at Q5 with signal (between the PNP stage and the octave volume pot): E: 8.9V B: 0.15V C: 0V
Using a probe, the signal is loud and fuzzy coming off of the collector of the PNP transistor. Once it passes the initial 100K resistor to join the signal coming from the diode the volume drops a bit. Past the next 100K resistor at the base of Q5 it is basically nonexistent. I can hear a tinny fuzzy tone if I crank up my amp accompanied with a thump at random intervals. I get the same tone whether D3 is connected or not. This makes me think that Q5 is misbiased but I've tested every component around it (using a component tester: TC1) and tried several varieties of transistor (Q5 is socketed now, not shown in the photos). I do actually get a quiet bassy fuzz tone (maybe the octave tone?) and no knocking when I invert Q5.

[Slowpoke101]

That looks like the Dirtbox Layouts version. You do know that it is unverified? But not to worry, I'm sure that everyone here will help you to get it to go.

First thing to do is to go over the copper side of the board closely and make sure that there are no solder bridges or slivers of copper causing short circuits. Remove any that you find of course. Then make sure that you have installed all the components in the correct place and made the cuts to the copper tracks are in the correct places.

Cross-checking the circuit diagram against a strip-board layout takes time, so give us some time please. There may be errors in the layout itself. You may have not made any errors at all with your construction.