Prescription Electronics COB original transistors

[jerb]

Hi!

I am looking to build a COB clone based on some schematics found online. I could not find any info regarding the transistors used in the original. 2n3904 is suggested often but I have seen many different suggestions but want to know what the original used.

Also, are 1n4001 diodes used in the original? Wouldn't some other type, perhaps schottky, provide better and cleaner octave?

Thanks!

[antonis]

Hi & Welcome..

2N3904 (as well as any other jellybean BJT) should be fine..
(no need for either high gain or low noise transistors..)

As for the diodes, they should be as close as possible in forward voltage drop for best octave effect..

[Mark Hammer]

Agreed.  I built one, and it's a nice octave-up.  Do note, however, that the octave itself is NOT clean.  The name really should be Clean/Octave Blend, so that people understand it blends between the clean boost and octave fuzz.

Ideally, GE or Schottky diodes are used.  It will "work" with silicon diodes, but may not respond as well to delicate picking.  Try to use a pair that have very similar forward voltages (which should not be hard to do)

[jerb]

Agreed.  I built one, and it's a nice octave-up.  Do note, however, that the octave itself is NOT clean.  The name really should be Clean/Octave Blend, so that people understand it blends between the clean boost and octave fuzz.

Ideally, GE or Schottky diodes are used.  It will "work" with silicon diodes, but may not respond as well to delicate picking.  Try to use a pair that have very similar forward voltages (which should not be hard to do)

Yeah, that's what I was thinking. What diodes did you use in yours? And do you by chance know what the original used? Im trying to get as close to the original as possible.

[Mark Hammer]

I'd have to do some disassembly to tell you what diodes I used.  Not likely a "today" activity.  But I imagine you'd be happy with GE or 1N914, albeit with a little more volume from the GE.

[pacealot]

To the technicality of the question of what was actually in the original pedals, I was only able to find a couple of useful images, but this one from a listing on the reverberant website shows at least one 2N3904 and what also appears to be a pair of 1N4001s (and so conforming to the commonly-accepted trace results):



However a different listing shows diodes which look more like germaniums to my eye:



(The trannies on that board are only visible from the top, so no real help there.)

[antonis]

Yeah, that's what I was thinking. What diodes did you use in yours? And do you by chance know what the original used? Im trying to get as close to the original as possible.

Original circuit (as many other well designed ones) relies its performance upon its design rather than items individual properties..

[mozz]

I'm sure there are many makers of 2N3904, picture is a bit blurry so can't see who the original maker is.

[antonis]

I think it's Motorola..

P.S.
I've seen COBs with BC107,also..

[moosapotamus]

I have an original and it is loaded with 2N3904 and 1N4001.

[duck_arse]

given that that looks like a Motorola M, and that there is E B C markings, I'd say 100% Motorola.

[mozz]

First i thought there was another company that had a symbol like that.National semiconductor?





Then looked for Motorola 2n3904, most in my stash have gold leads and the EBC is embossed. You also have the top, where the package is molded and snapped off, these seem to be offset, where my Motorola are all centered. Then looking through another file cabinet, i found one that's probably a match, i guess later ones they dropped the gold plating and embossing and just made them with printing. I'm agreeing 99% they are Motorola. Hey you said " Im trying to get as close to the original as possible. " Then you have the hole in the back, some have symbols in there...........

[PRR]

a symbol like that.National semiconductor?

That image is indeed very early NatSemi, with the bent-arm "N".

[Rob Strand]

That image is indeed very early NatSemi, with the bent-arm "N".

I was just about to post the same thing.  The national semiconductor logo changed a few times.  I remember that old logo but I can't remember the exact details for transistors.   

I'm not 100% convinced the transistor is real.  It could be a fake.  I'm not sure at this point. 
The thing that alarms me is the diagonal line on the real logo has points (ie. a flat on each axis forming an "arrow" point) whereas the lower part of the diagonal on the transistor only has a vertical flat.
I have a transistor with these markings.  However this is the full N and S logo.  You can see the diagonal line on the N is different, in fact the "arrow" point is at the lower end.



Online I found this one, which cannot really be judged as 100% verified.  This one has no arrow head points at all.  The S is equally dubious.

[PRR]

diagonal line on the real logo has points

Yeahbut, the brown logo is from color literature with all the lithographer's art behind it, and the transistor may be a rubber stamp. Maybe made in haste (even from a preproduction sketch) or already used a million times. Our FBI used to claim they could identify the specific typewriter used in a crime, and this has been discredited; plastic(*) transistor marking may not even be that good.

(*)I did once handle presentation-grade 2N2222 from 1964. Beautiful metal cases (gold leads) and very careful printing. They don't even make CPUs like that anymore.

[duck_arse]

a symbol like that.National semiconductor?

That image is indeed very early NatSemi, with the bent-arm "N".

I was going to say, AM going to say - I've never seen a transistor with that particular NatSemi marking. however, I was going to add, WILL add - in my experience, the NatSemi parts were always slightly less better finished than you might want. then I looked again PRR's image, saw the flash between the legs*, and I'd now say that is 100% natsemi.



* not a euphemism.

[duck_arse]

Online I found this one, which cannot really be judged as 100% verified.  This one has no arrow head points at all.  The S is equally dubious.

I have one only BC178, gold pins, date code 7615, has that exact logo, but in black, and, obvs, not quite entirely fit on. the top is missing for the roll of the can.

[Matthew Sanford]

a symbol like that.National semiconductor?

That image is indeed very early NatSemi, with the bent-arm "N".

...then I looked again PRR's image, saw the flash between the legs*, and I'd now say that is 100% natsemi.



* not a euphemism.

So you're not saying he is Old Gregg, but still can't confirm he is not in possession of "The Funk"?

[Rob Strand]

Yeahbut, the brown logo is from color literature with all the lithographer's art behind it, and the transistor may be a rubber stamp. Maybe made in haste (even from a preproduction sketch) or already used a million times. Our FBI used to claim they could identify the specific typewriter used in a crime, and this has been discredited; plastic(*) transistor marking may not even be that good.

Nonetheless, fine details of the fonts and logos is where fakes often fall down.

I have one only BC178, gold pins, date code 7615, has that exact logo, but in black, and, obvs, not quite entirely fit on. the top is missing for the roll of the can.

Interesting.  If we put aside the arrow detail, the thing that stands out is the *single* old style N, essentially a partial logo.

I've found the complete "NS" logo in various forms on various packages:  button heads (TO-105) and metal parts (TO-39/TO-5, TO-18).
I have not found any examples with a single old logo N. 
However, I have not found any (plastic) TO-92 packages with old logos of any sort.

FWIW, the transition from the old "NS" logo to the two wiggles was within year 1976.   That might explain the absence of a lot of TO-92 examples.
(eg. https://www.diystompboxes.com/smfforum/index.php?topic=124599.msg1183026#msg1183026 )