Gemini III dual fuzz dissected

[digi2t]

OK people, Christmas is here! 

Here is the board tracings. There are 2 boards (upper/lower), connected by 7 bridge pins, so numbered on the drawings. So, pin 1 to pin 1, and so on.

The lower board is double sided, so blue trace lines are one side, red the other. The pale highlighting is the components.

There are 3 unknown trannies on the lower board, the writing has been sanded off, but they're silicon. They'll need to be diciphered by the experts here.

Some resistor values were very hard to decipher, because the resistors are so tiny, so I read the colors both ways, and highlighted my thoughts.

There are 2 trimmers on the upper board, 25K and 100K.

The transformer, only the primary side is used. I wrote the resistances I read on the poles on the drawing.

Without further ado;





It was real delicate work, like eye surgery, but I think it should be correct. I'm 90% sure about the traces to the 5n6 cap (next to pin 1) on the lower board. It was near impossible to see under the dual pot, but I don't think there is any other traces going to this cap.

I know it may seem that the right side of the images here are cut off, but if you download it to your computer, it will appear.

It would be nice if someone could now draw up a proper schematic. From that, I can do a vero layout.

Let the dissection begin!!

[digi2t]

Alright, I had a slow shift at work  

There are 3 transistors that had the facing erased, to hide they're identity, but with some research I think 2 of them are for the Millenium bypass. I labeled them as such.

The third one (marked ) is troubling me though. Again, my internet diving is pointing me towards a type of jfet buffer, but I can't be certain. Personally, I would put my money on the jfet buffer, though I couldn't say which jfet would be the right one. A BS170, or 2N7000 might be fine here, but I would appreciate an expert opinion on that.

As for the transformer, I'm getting resistances of 216 ohms on the secondary, and 463 ohms on the primary, so I've indicated the Xicon transformer that comes closest (200/500).

Here's my work to date;



So, what's the verdict on that last tranny? R.G.? Paul? Mark? Anyone?

I'm going to record voltages next.  

*EDIT* - I corrected 3 resistor values on the schematic. The 301K is actually 308K, and the two 25.5K's are both 36.5K's. These corrections ARE NOT reflected in the hand drawn trace. You've been warned.

[Bill Mountain]

Please forgive me but what's this all about?

Edit:  I just did a search and you appear to be reverse engineering a pedal.

I just felt like I walked in the middle of a conversation.  Please carry on!

[digi2t]

This the (unverified) schematic of the Spaceman Effects Gemini III Dual Fuzz Generator. It's a sweet parallel fuzz, germanium on one side, silicon on the other, and you can blend the two together. Some videos on Youtube.

They're as rare as the Pope's sh*t, since there's only about 160 or so in existance.

[digi2t]

Please forgive me but what's this all about?

Edit:  I just did a search and you appear to be reverse engineering a pedal.

I just felt like I walked in the middle of a conversation.  Please carry on!

No worries Bill. It's all good in the hood.

[slacker]

Looks cool, never heard of it before, I'll have to check out  youtube for videos.

The last transistor could be virtually any NPN transistor, it's a booster, presumably to make up for volume lost in the tone controls and mixer before it.

[Bill Mountain]

Looks cool, never heard of it before, I'll have to check out  youtube for videos.

The last transistor could be virtually any NPN transistor, it's a booster, presumably to make up for volume lost in the tone controls and mixer before it.

Wouldn't it be a buffer?

Edit: Nevermind.  I was looking at it upside down!

[slacker]

Mmmm, Fuzzbox Girl

[digi2t]

Looks cool, never heard of it before, I'll have to check out  youtube for videos.

The last transistor could be virtually any NPN transistor, it's a booster, presumably to make up for volume lost in the tone controls and mixer before it.

Hmmm, I thought that as well, but what's throwing me off is the fact that they ground the info off the tranny. Actually, they did the same with the fet and tranny for the millenium bypass, but tracing the circuit, it became obvious to me what they were.

Problem is that I don't have enough smarts yet to negotiate weather it could be a jfet or transistor yet. With the info I do have, I'm scouring the net looking at similar circuits. One I did find, I thought fit the bill;



But, it's not quite right. Looking at the layout again, I'm backtracking to the transistor again. Especially looking at this;



What's really throwing me off too, is the lack of a resistor opposite the 4M7 to bias/voltage divide. I think I might have to just construct this part of the circuit on the breadboard, and see what I get. Like I said, I don't have enough smarts to decipher this circuit just by looking at the lines and  numbers.  

[slacker]

I think you're getting confused because of how you've drawn it. Try standing on your head and it will make more sense.

[digi2t]

I think you're getting confused because of how you've drawn it. Try standing on your head and it will make more sense.

+1.

How about this?



Which gives me this;



I guess I'll have to breadboard this to find out which transistor gives the best results. Considering the fact that they ground the facing off the transistor and jfet to hide their identity on the Millenium II bypass circuit, it wouldn't suprise me if it's some mundane transistor like a 2N3904, or something like that. It's funny that they went to the trouble of grinding the numbers off this on as well, but I guess in my case, it seems to be giving me a good fit.  

BTW, I came across a great little demo on simple transistor biasing, and what it does. The guys voice is a bit monotone, but IMO his demo is excellent. Really good for a "visual" creature such as myself. 

[digi2t]

Alright, I breadboarded that circuit with a 2N3904, and it works. I didn't have any 4.7M's, so I used two 2.2M's in series to get close.

I've come to the conclusion that the unknown transistor is probably any low power small signal tranny (2N3904, 2N2222, etc.), used as a final booster. Especially since this pedal can be very loud at full tilt. Right now I'm at unity with the volume at about the 9 o'clock position.

I'm going to update the drawing, and leave this open for anyone to try different trannies to taste.

*EDIT* - Drawing is updated. Just takes some time to make it's way through Photobuckets servers sometimes.

[slacker]

Yeah, that's it. Like you said pretty much any NPN transistor will work in there.

[digi2t]

Thanks for your input Ian. Much appreciated.

BTW, I'm triple checking the schematic vis a vis the layout, and I spotted a couple of errors. They're now corrected, the schematic should update shortly.

[Earthscum]

On the transformer, I'm willing to bet any 10k Pri. transformer would work. The Xicons usually have the number stamped on the frame, don't they? I was looking at the 42TU012, which is 10k:5k, and 425R:220R. The 42TM019 that Steve sells should work fine. It looks to me like it's just a pickup sim, but I'm not sure if it's going to play differently given that it's a higher DC resistance, i.e. smaller dia. wire. I'd say the 42TL019 would be the best. Still a 10k primary (assuming the primary is 10k), but the DC is 450R, much closer to the target reading of 465.

[digi2t]

On the transformer, I'm willing to bet any 10k Pri. transformer would work. The Xicons usually have the number stamped on the frame, don't they? I was looking at the 42TU012, which is 10k:5k, and 425R:220R. The 42TM019 that Steve sells should work fine. It looks to me like it's just a pickup sim, but I'm not sure if it's going to play differently given that it's a higher DC resistance, i.e. smaller dia. wire. I'd say the 42TL019 would be the best. Still a 10k primary (assuming the primary is 10k), but the DC is 450R, much closer to the target reading of 465.

Yeah David, you're right. As long as it's 10K, and as close to the primary 463 ohms, you should be in business. I think that there is quite a high tolerance in this little transformers anyways. Besides, at 2 dollars a piece, I would maybe buy 2 or 3 different ones, and test to see if there is any difference.

[digi2t]

For the sake of clarity, I've done a redraw.



It's a bit easier to follow now. I did forget to update the transformer part number though (sorry David  ). I'll fix that later today.

[dorrisant]

Wow!! Thanks for the post and all your effort!!

Gotta build this one!

How did you get your hands on one of these? Gutshots?

Tony

[digi2t]

Wow!! Thanks for the post and all your effort!!

Gotta build this one!

How did you get your hands on one of these? Gutshots?

Tony

No prob. I did manage to get my hands on one. It's a great sounding pedal.

Heads up though!!! Some errors were brought to my attention, and after further inspection, digging, etc., I've made some changes to the schematic. Notably, one tranny is an AC176, and NOT an AC128. Huge difference, I know, but it was an assumption on my part since the writing on the can was on the underside. A PM by another DIY'er made me go back and really take a good look. I managed to spot that in the reflection of the tranny in the tinned ground plane. Two cap values have been corrected, misread on my part. A resistor value corrected, that was a copy/paste typo. The transformer part number has been updated as well to match to a closer model.

Here's the new drawing;


That should be it. When I get some time, I'll whip up a vero, unless someone feels like beating me to it, be my guest.

[digi2t]

Alrighty then...

Took me some time, I've been rather busy of late, but I managed to get all the voltages from the trannies. They follow the numbering of the latest schematic, so without further ado;

POWER SUPPLY VOLTAGE = 9.15vdc, NO LOAD ON CIRCUIT, BOTH GAIN SWITCHES SET TO MINIMUM (POSITION 1).
Q1-
E=2.017v
B=1.882v
C=1.491v

Q2-
E=1.522v
B=0.946v
C=0.769v

Q3-
E=1.547v
B=1.586v
C=1.672v

Q4-
E=1.271v
B=1.863v
C=3.164v

Q5-
E=1.752v
B=2.182v
C=2.207v

Q6-
E=1.184v
B=1.752v
C=2.207v

Q7-
E=161.8mV
B=0.694v
C=1.507v