My Transistors, original or copy?

One of the other problems in the "fake" market is the manufacture of "onspec" devices. A company goes to a manufacturer and asks them to do a run of say 100,000 unlabelled NPN and specify the simplest factors like VCE HFE and Dissipation.

Then they go and label them themselves with comparable type part numbers. Sometimes they're OK in the end user app. other times they're a disaster.

One of the clues with the "nonumber" parts is a different printing for the date code and the label. But you can't rely on that either, because a lot of transistors are similar and are manufactured from the same basic dies and are then selected and labelled according to the tests.

It's like buying in the "interchange" market. There are subtle differences between similar numbers, like BC182, BC183 and BC184 (BC107, 108 and 109) transisitors, but you'll often find them all under a single interchange number which for many applications, they are, but with different VCE and HFE and gain bandwidth products, you could come unglued.
 
Is it possible that there also are fake TO92 devices? I recently got a couple of 2sa1123 transistors, to replace a couple of noisy input-stage 2sa942 ones. I got about 8 or so in total. A few were dead on arrival, a few died after installing, and the ones that do work, have widely differing Hfe values.

These are the original and the replacement:

2sa1123+2sa942.jpg
 
2sa1123 is manufactured by Panasonic (Matsushita) and has the standoff blips on the leads and a factory lead offset like the 2SA942 you took out. Your 2SA1123 has neither. I'd say it is a fake since it appears that it doesn't even meet the specs.

Moreover, Panasonic has superceded the A1123 with the A1123S
 
2sa1123 is manufactured by Panasonic (Matsushita) and has the standoff blips on the leads

What exactly are standoff blips?

and a factory lead offset like the 2SA942 you took out

Like so?:

to92.jpg


edit: I assume the 2SA872 also has an offset middle pin?

Moreover, Panasonic has superceded the A1123 with the A1123S

Since when is this?
 
Not quite ...

Look at your A942 ...

The outer two legs bend outwards and then down. That's actually done in the transistor production and not during assembly into a board. This appears to be standard on Matsushita (Panasonic) TO92 case transistors.

Then also as you come out of the bottom of the A942, the legs get fatter and then thinner again (blips). This is so that the transistor cannot seat all the way down on the board when inserted into a PCB (It "stands off" the PCB)

See here ... http://www.ortodoxism.ro/datasheets/panasonic/SJC00011BED.pdf
 
Administrator
Joined 2004
Paid Member
Hi sbrook,
Lead forming occurs when components are supplied taped. Bulk transistors are typically straight with no bends. Component insertion machines need the bends done before they ram the part into the PCB. Just because the leads are not formed does not mean they are fakes.

I use offset pliers to form the leads into that shape.

The 2SA872 matches the specs of your 2SA942 more closely.

-Chris ;)
 
halfgaar said:
Is it possible that there also are fake TO92 devices? I recently got a couple of 2sa1123 transistors, to replace a couple of noisy input-stage 2sa942 ones. I got about 8 or so in total. A few were dead on arrival, a few died after installing, and the ones that do work, have widely differing Hfe values.

These are the original and the replacement:

2sa1123+2sa942.jpg

I think that TO-92 devices have by far the largest fake market, although the quality sometimes is not as deceiving as with high power devices.

If some devices were faulty on arrival, you should return all them, get your money back and try another source. According to what you tell, all seem to be fakes (unless you managed to damage them in some unexpected way).
 
In response to my notification of possible fakes, I got some more free ones, unfortunately, all fakes as well. This was back then. Just now, I sent them a message asking for the 2SA872, and reiterating the high possibility of fake 2SA1233's. They are friendly folk there, and are as much a victim as I am. They might be able to complain to their suppliers.
 
Well, the 2SA872 are manufactured by Hitachi (now Renasas)

http://www.ortodoxism.ro/datasheets/HitachiSemiconductor/mXyzyuuu.pdf

The data sheet only shows stand offs ... not offsets.

Chris (anatech) is correct for some manufacturers - the lead formation is done later, but I think for these it is a part of the manufacture process, just as Philips (Mullard) did with their LokFit transistors (BC147/148/149 etc.) They were designed for automatic insertion.

While we can never be sure, it sure sounds like the ones you got were definitely of questionable origin and I would be suspicious of the new type if they don't look like those on the data sheet. Also, check the printing on the labelling ... if it looks like the ones that you got last, be very suspicious and check them out for HFE and polarity before installing.
 
Also, check the printing on the labelling ... if it looks like the ones that you got last, be very suspicious and check them out for HFE and polarity before installing.

What aspect about the labeling would be suspicious?

And even the polarity can be wrong? Perhaps that's why I initially thought they were NPN transistors, because that's what the multimeter said...
 
Administrator
Joined 2004
Paid Member
Hi,
For Japanese transistors, only one manufacturer makes a number. Another device from someone else is assigned a different number, even if they are identical parts.

Hi sbrook,
In all my years of servicing, I've received transistors with bent legs depending on whether they were delivered on tape or bulk. I've never once noticed any brand in particular and don't even think about it. taped transistors are almost always bent for Japanese parts.

Panasonic (Matsushita) may be different. I can't say.

-Chris ;)
 
anatech said:
Hi,
For Japanese transistors, only one manufacturer makes a number. Another device from someone else is assigned a different number, even if they are identical parts.


Which sure makes it difficult when the manufacturer doesn't actually display how he marks up his transistors! One of the good things about most Euro/N.Am. mfrs ... they actually show their typical markings so you can get a better grip on what you're looking at!

It also makes it difficult to find replacements. Goodness knows it's hard enough when you HAVE got multiple manufacturers manufacturing to similar specs.

In all my years of servicing, I've received transistors with bent legs depending on whether they were delivered on tape or bulk. I've never once noticed any brand in particular and don't even think about it. taped transistors are almost always bent for Japanese parts.

I believe you ... though it still adds "?" marks as to the authenticity of the devices.
 
Administrator
Joined 2004
Paid Member
Hi sbrook,
I believe you ... though it still adds "?" marks as to the authenticity of the devices.
The failure rate alone does that, and measured specs. Japanese manufacturing is to a very high standard.

It also makes it difficult to find replacements.
Not really. Get the Japanese spec books and look them up. There will be a few popular devices in use that will be close. You can even "google" them.

-Chris ;)
 
Hello there people.. i have been doing much reading of this forum recently and learned much from you all. in particular i have been reading up on these counterfeit transistors....

pictured are 2 5200s and 2 1943s... for each, 1 genuine, and 1 presumed counterfeit.

my question is only this.. The ones with the clear markings in the smooth rectangle have been mentioned many times here and elsewhere on the web, but does anyone know for certain if Toshiba have ever manufactured transistors in this exact packaging?

if someone with more knowledge and experience than me could clarify this once and for all, it would be helpful to many people to have such an obvious difference clearly acknowleged.

apologies for the quality of the photo.. i really did try to get a good clear one.. but photography is not a skill i have.

greets...

steve.
 

Attachments

  • transistors2.jpg
    transistors2.jpg
    64.8 KB · Views: 1,301