More fake transistors

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
EUVL said:
I have 50 MJ21194s DIRECTLY purchased from On Semi about 2 years ago, and they look EXACTLY like those posted in Post#1. My On-Semi datasheet from 2001 also indicates different case markings. For those transistors that I have, the markings also come off if you wipe it with a strong solvent, so I am afraid that it is no solid proof until someone would take the trouble & expense of actually opening up a genuine MJ21194 and see what the die size is.

what solvent did you use?

will the markings of original and/or fake ones come off with acetone/ lacquer thinner?
 

taj

diyAudio Member
Joined 2005
Eva said:
These transistors look genuine due to the manufacturing style, I bet they are actually made by On-Semi, but they are definitely *not* MJ21194, I think that they are some other lower power model REBRANDED.


I'm having second thoughts about agreeing with this. Referring back to the photo in post #1, the die is mounted squarely, but in all the real ON Semi / Motorola devices I've seen cut open (admittedly only a handful) the die has been mounted diagonally.

Also, though it doesn't show much in the photo, the die is submerged in a pool of clear epoxy-like glue in these devices. I've never seen that in a real On Semi/Motorola device either.

..Todd
 
taj said:



I'm having second thoughts about agreeing with this. Referring back to the photo in post #1, the die is mounted squarely, but in all the real ON Semi / Motorola devices I've seen cut open (admittedly only a handful) the die has been mounted diagonally.

Also, though it doesn't show much in the photo, the die is submerged in a pool of clear epoxy-like glue in these devices. I've never seen that in a real On Semi/Motorola device either.

..Todd

I got some MOSPEC branded MJ15015 and they are genuinely rated but they do have those heat dissipating coins and the die covered with rubbery silicone. the main difference is that the heat dissipating coins are smaller.

attached pic shows a 'rebranded' MJ15003 (guessing according to die size) from an amp kit on the right and a MOSPEC MJ15015 on the left.
 

Attachments

  • trans.jpg
    trans.jpg
    91.8 KB · Views: 518
I find the best way to open TO-3 cases is with a small bench vise. I squeeze the can portion until the weld breaks, the rotate it and squeeze again. When I have enough of an opening, I stick a screwdriver in the break and pry the can off.

Over a lot of years I've see so many variations on cases, printing, dies, angle of mounting, and just about everything else, to have little faith in my ability to detect a fake. The clear RTV coating was common, though I don't know if it still is. Because suppliers are many and scattered, you can't assume any given part will always look the same. Even die size can change, though the example given is pretty scary. Ah, for the good old days when manufacturers published an image of the die, and you could simply put it under a scope and compare the pattern and measurements!
 
Taj,

How do know that these are fakes. Did you actually hook them up to a power supply? I would tend to think that the beta would be oncorrect if they were [fakes]. Likewise, I suspect that they would self-destruct when made to operate at the published maximum ratings if they were not the real thing. Being a tube guy I can attest to the fact that the same part # may look different from one manufacturer to another (compare an RCA 6SN7 with a Tung-Sol 6SN7), but remain electrically equivalent.
 

taj

diyAudio Member
Joined 2005
Hi 6F6,

I'm not 100% sure what you're getting at. We're only talking about one manufacturer here, (ON Semi/Motorola) not other brands or generics or anything. The REAL (not counterfeit) MJ21194's have a die that is 4-times the size of these ones. The date code is 6 years ago, and much newer examples I've seen of the same part# still have the large die, so they unfortunately haven't learned how to reduce it to 1/4 the size and still handle 250 watts (yet).

I'm not putting these into a working circuit, nor would I bother with any current load tests. The outcome is pretty predictable.

However, I have a bunch here you can play with and test if you want, and all it'll cost you is shipping, if you're interested.

..Todd
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.