Real Motorola 2N3055 or Fake?

The die in the OP is 3.5 mm square, per the attached picture.

Per the IEEE history paper I posted earlier, the die was 4.6 mm square in 1960. I'm not sure if the die shrank heading into the early -mid 70s.

But the scratched up top plate and the wrong font tells me my batch are not Motorolas. I think someone sanded off the original label and imprinted the new label.

So in the trash they go. Thank you all for the help.  Luckily they were very cheap.

Edit: the roughed up aluminum top plate and font both appear to be consistent with vintage Motorola 2n3055 after all.

stockrock said:

The die in the OP is 3.5 mm square, per the attached picture.

So in the trash they go. Thank you all for the help.  Luckily they were very cheap.

Click to expand...

Hang on with that, as the mystery thickens.

Figured I'd sacrifice one for the team, so off came the top of a Motorola '76 3055, the one in my earlier jpeg. The die is smaller than yours, a lot smaller. Measures about 0.104" square, about 2.65mm across.

I was not expecting that. The emitter fingers even look similar to yours.

I have dozens of pulls and rarely use them, I don't mind cutting more tops off, if anyone is interested in certain years or brands. Plenty of RCA's in the pile.

Perhaps yours were meant for a specific client that spec'd the date codes written different. If a client has a custom spec for an order of 100K devices, they'll print whatever the client wants on them, even pictures of naked ladies, I suppose.

Hmmm, hfe indicated by breast size? It'd be fun to match them up. ;D

Gene

Wow, thanks for that. Ok, not going in the trash yet, pending more information.. do you have a capacitance meter, and have you measured capacitance between B and E on your known good Motorolas?

Looks like the pins on mine are gold, and the pins are yours are silver-colored.  When did Motorola switch from gold to silver pins, or vice versa?

And hope springs eternal - here is an old ad from this site with Motorolas installed on cards, and they have the same font and date code as mine.  I attached the file, and linked the thread, below.

Mine were sold to me as 1974, and they came out of old looking styrofoam trays.  It looks like heat sink grease (that has hardened over time?)  was smeared inside it, which I have read is an earlier feature of authentic 2N3055s.  I believe. And the leads appear to be aluminum, but the B and E posts are copper (aka "gold" pins).  I wonder if there are any clues in a 3 digit date code vs a 4 digit date code as far as timeline.  The problem with a 3 digit of course is that it repeats every 10 years. So my 1974s could be a decade later, or a decade earlier.


https://groupdiy.com/index.php?topic=62660.0

This is a side by side of the vintage Motorolas on that old card, and mine.  The font looks similar. Not sure if they vintage motorolas are 1962 or 1972, but they also have a 3 digit date code.  The 2s, 3s, and 5s and Ns are very distinctive and the same font, and spacing.

Digging a bit, I get this from a Google Groups discussion:

Motorola 2n3055s  are epitaxial, and always used aluminum bonding wire connecting the silicon B&E to their pins - this was so that overstressed/failed devices could disconnect themselves and limit damage.

Motorola used Aluminum Cases until the mid 80s, so mine are consistent with 1974.

I am still not sure about the gold pins, but I am getting comfortable that the font on mine is ok for an original Motorola 2n3055, and that the date code is 1974.  Overall more comfortable that mine are valid - they came out of a batch in Switzerland, from a large Swiss company.

This thread here has a picture of 3 Motorola 2n3055s, and the one with the 3 digit code has gold pins - AND has the same font as on mine.
https://groupdiy.com/index.php?topic=56395.20

I wonder if three digit date codes and this font might go with gold pins generally? Or maybe this Swiss company just ordered them with gold pins and 3 digit date codes, and this font was used on some special orders?  Just getting more and more comfortable mine are authentic, which is nice.

stockrock said:

Wow, thanks for that. Ok, not going in the trash yet, pending more information.. do you have a capacitance meter, and have you measured capacitance between B and E on your known good Motorolas?

Click to expand...

I can do that, if I have a cap meter that has a test voltage of less than the B-E forward voltage, I never actually checked what comes out of my cap meters. I imagine getting into B-E on, would screw measurements up. Otherwise, I'll have to rig up a bridge.

Just curious, if you are looking for a larger B-E capacitance, can you simply hang a cap between B-E?

Might not be tomorrow, but I'll get after this soon, as I wanna know too.

Gene

Hi Stockrock,
yeah that was the document I was talking about .

What I noticed about the motorolas is that the the top cover is aluminium, most of the others Ive seen are steel on top. Ive seen the older RCA types available in reasonable quantity ,they have the lightening strike logo, and date from the 60's,so definately the older die. I did read a fairly comprehensive document by another guy who swore the older ones sounded better in guitar amps .Maybe its possible the older ones in a neve output stage have a nicer overload charachteristic too. Ive tried both kinds in BA283 ,but I did no specific test on transient overload.

Could anyone tell me what might be an appropriate signal for such a test, would injecting a burst of two frequencies like when testing for intermodulation distortion be the most appropriate ?  The usual steady state sine wave test displayed on a scope ,doesnt reveal much. Say if using an arb sig gen and DSo ,what type of signal could be used effectively to give a better representation of the overload charachteristics of a circuit to make comparrison between different output devices possible?

I simply measured capacitance - positive to Emitter, and at 1 pf resolution the test frequency with my tester is 820 Hz.



And got high 400-500 pf on the batch in question.



The one in the top right is a brand new 1976 Motorola that I popped out of its package to test. So my batch largely tested higher capacitance than the reference.  Maybe because the die got a little smaller moving from 1974 to 1976?

Also, the test batch all have gold pins, vs the aluminum/silver pins on the 1976 reference.

Gene Pink said:

I would bet "fake".

Click to expand...

Note 4-digit date codes, and the "3" in the number is a different font than yours. Also, stamped numbers around the mounting holes.

Click to expand...

100% originals of mine 

Seriously what is the goal here?  If we are trying to make a classic (old) circuit that used 2n3055 work as good (bad?) as the legacy 2n3055s available back then, it seems to me the fist step is to characterize the transfer function of the circuit in question both using a legacy part, and modern (better) part.

This may be much ado about nothing or very little. Inferior parameters associated with the legacy parts could probably be mimicked with compensation.

JR

PS: Of course the only way to secure pure mojo is to use crazy expensive original parts.  :  Caveat, old gear may have been repaired at some later date so may contain newer replacement parts.  I applaud the attempts to use bench measurements to confirm legacy, just question why measure the parts out of circuit, when in circuit performance is what you are trying to recreate? 

stockrock said:

Wow, thanks for that. Ok, not going in the trash yet, pending more information.. do you have a capacitance meter, and have you measured capacitance between B and E on your known good Motorolas?

Looks like the pins on mine are gold, and the pins are yours are silver-colored.  When did Motorola switch from gold to silver pins, or vice versa?

Click to expand...

None of mine have gold pins, but I think I may have seen that somewhere.

Another picture of some oddballs, and some measurements. The oldest I have with a real date code is these military ones, 1972. and I ain't cutting these up. Both of my cap meters are less than a diode drop p-p, so should work.

Westinghouse: 1383 pF

Big S, Signetics(?): 570 pF

Military : 841 pF

The '76 motorola I cut up: 394 pF.

Hope that helps, seems like yours are real after all, and I apologize if my first post misled you.

Gene

I just found these on a power supply for an industrial laser thingamajig.....

Just another view of I guess some real ones......

https://s20.postimg.cc/ay2tcol71/20180704_134817.jpg


just noticed the markings around the screw holes....... I guess it's maybe a plant thing???? I don't see them on all varieties....

just use something that nobody would bootleg,  like a 2N3054 in the TO-66 case.

or get pre-sorted 2N3055's out of a peavey carcass,

2N3055A = SJ-6392  = 70483100  = 100 V-ec

or


2N3055 = 62792 = 84140 = 140 V-ec

Early Neve cards all used the Newmarket 2N3055, but no hype around them...yet!
I have also seen RCA branded ones and also other type of transistors to on original cards.
But Motorola became the more consistent choise when they started to use the BA283 cards in the early 70´s.
Many different fonts and datecodes used on the Motorolas over the years.