Fake transistors

[sureshot]

This might have been mentioned before, I'm not sure. Having bought voltage regulators from china that where fake, i don't bother with that anymore. I've got a trusted handful of ebay sellers that have NOS new old stock. But i thought I'd try a seller with some transistors that where home based here in the UK. They arrived today and check out fine doing a test with the diode function, but there not looking like the normal finish to me. The tab is more shiny, i filed off a little of the plating, sure enough its copper. But they just seem not quite right. I don't have a camera handy at the moment, but will put up a comparison later between legit transistor, and the one's I'm unsure about. I'm tempted to do a distructive test, but something is telling me to bread board one and test it under load.
I knew i might be buying fake, but at £1.00 each it was worth a gamble. The devices are TIP36C power transistors. I was hoping to get a bargain by buying a small quantity first. Then a few more if they checked out good. The voltage regulators from china where lm317k's i opened a fake one and could see the difference immediately,  they couldn't even handle 500mA of current. Anyone else been stung with fake transistors. Thoughts appreciated.

[Gyro]

Yes, fake transistors do exit, and in large quantities from certain locations (Ahem). It's much less likely from a genuine UK registered seller though. The simplest way to tell is to look at the seller's other items. From the selection it's usually pretty easy to judge whether they are selling genuine NOS / overstock or importing fakes.

[sureshot]

Yes i always checkbout the seller carefully, just to judge if its worth the gamble. This seller has 100% feeback and has sold quite a few, no negatives to speak of.
Below is the known good transistor on the left and the suspect transistor on the right. Both show virtually the same with multimeter diode check.

[Gyro]

Hmm, I see what you mean, it looks 'tinned' rather than plated. I wonder if those score lines on the back of the pins could be where new leads have been butt welded? I'm probably just being suspicious though!

[MK14]

One of the problems with fake power transistors, is that the silicon die in the transistor will probably be far too small, for the datasheet specification of the device.

Hence failures like shown in the video:

[sureshot]

That's it really, there's no way of telling straight off the bat. From looking at fakes on the web you can see the die is way smaller than the real device in many pictures. And I guess the readings would check out with a multimeter. It's not until it's put under load that it's true parameters would show. Thanks for the replys and for the video. That seller off eBay had over 25 confirmed sales of that transistor, nothing in the negatives. I checked a lot of other sales with components he's done, again it all looks good. I did file the top edge of the plating just to see if it was copper. That checked out ok. Just have to put them in a circuit and test them. Could be another project.... Test circuit with quick interface change out for npn and pnp devices.

[Jwillis]

Fakes Fakes Fakes .If your in doubt of where the unit is manufactured ,Look it up. ST Microelectronics has facilities in all over the world including China .Many companies outsource manufacturing around the world . Many of these out sourced facilities will have slightly different equipment for packaging the dies for a component.  Just because it looks different does not mean its not genuine .These companies do this to bring costs down for the consumer.
I myself have used components from many facilities around the world and have found that the expected tolerance to be the same no matter what the package looks like.
Do I support these practices? Of course not .But I won't get into a long economic, political argument for this isn't the place to do it.
All I say is if your in doubt  its easy to check where something is made.

[sureshot]

As I've always purchased from good known sellers, I have had little to nothing in the way of fakes.the UK company's I use are family run and are rock solid. I buy from them because there prices are fair, it supports my country's small businesses, and I've been buying form them for decades. This was only meant to be a taste it and see exercise. It said generic in the listing so I knew what I was buying. It was purely to see if something cheap and cheerful would come up trumps. If not nothing lost.
So I will taste them and see. The thread wasn't a " moan about it fiasco " on my part. I've not been hit by fakes much, only once. And that was from China and was voltage regulators. Most semiconductors don't look like cheap bling jewlery. But I'm fine with it, if they work with out failing it's a bonus. It might be wrong to think like this, but I find the lengths the fakers will go to fascinating.

[MK14]

It might be wrong to think like this, but I find the lengths the fakers will go to fascinating.

In that case, you might be interested in the following thread. Its OP, similarly, bought suspect fake transistors from ebay.
They thoroughly examined the transistor(s), both electrically and physically, as you will see if you visit the thread.

https://www.eevblog.com/forum/repair/warning-fake-mj16012-transistors-on-ebay/

[sleemanj]

Shiny legs doesn't necessarily mean fake...

  https://hackaday.com/2017/07/15/lets-play-spot-the-fake-mosfet/

but it's not uncommon for it to mean "recycled part", where it was pulled out of some e-waste, given a bit of a wipe down, solder dip, and sold to you.  The legs on the right-hand one at the back look like they have a cut/join, I have heard before of splicing new legs on recycled parts, but I couldn't readily google a reference for that.

[sureshot]

Shiny legs doesn't necessarily mean fake...

  https://hackaday.com/2017/07/15/lets-play-spot-the-fake-mosfet/

but it's not uncommon for it to mean "recycled part", where it was pulled out of some e-waste, given a bit of a wipe down, solder dip, and sold to you.  The legs on the right-hand one at the back look like they have a cut/join, I have heard before of splicing new legs on recycled parts, but I couldn't readily google a reference for that.

I found that in my web searches, and the hackaday one as well, thanks for posting them though.
The two transistors I bought I will try out next. Might be a while, but I will be back with the result, good or bad.

[thermistor-guy]

Yes i always checkbout the seller carefully, just to judge if its worth the gamble. This seller has 100% feeback and has sold quite a few, no negatives to speak of.
Below is the known good transistor on the left and the suspect transistor on the right. Both show virtually the same with multimeter diode check.

It's a half-decent copy. Notice with the fake that (a) the logo isn't quite right, and (b) it uses different fonts (e.g. the capital I in TIP, the non-italic e in e3). Give thanks that the frauds in China tend to be lazy about details.

[wraper]

Shiny legs doesn't necessarily mean fake...

  https://hackaday.com/2017/07/15/lets-play-spot-the-fake-mosfet/

but it's not uncommon for it to mean "recycled part", where it was pulled out of some e-waste, given a bit of a wipe down, solder dip, and sold to you.  The legs on the right-hand one at the back look like they have a cut/join, I have heard before of splicing new legs on recycled parts, but I couldn't readily google a reference for that.

They cut old legs off close to the body and then weld new ones on top. Look from the back strongly suggests that legs were welded. I remember someone posted x-ray pictures of this on the forum.

[sureshot]

Thanks for the heads up, that xray would be really interesting to see. I'm unsure how to do a distructive test so as not to ruin the end look and see result. I know acids can disolve the package material, but there are tough rules in the UK for such chemicals. And even then I'm not sure if it would destroy the die. I'm happy to do a distructive test on both good transistor and suspect transistor, but would want to get it right. Just thought I'd post this, i just purchased some large snap in capacitors. As i bought a few i thought I'd do a destructive test on one, the seller had one single comment on these in his feedback claiming fake. For all the world it looks fine to me 63 Volts 4700uf snap in electrolytic capacitor. Bargain £1.00 each. Worth sacrificing one capacitor for the war effort on fakes. I was expecting 20uf soldered inside the can, or something similar. But looks fine to me.

[wraper]

As i bought a few i thought I'd do a destructive test on one, the seller had one single comment on these in his feedback claiming fake. For all the world it looks fine to me 63 Volts 4700uf snap in electrolytic capacitor. Bargain £1.00 each. Worth sacrificing one capacitor for the war effort on fakes. I was expecting 20uf soldered inside the can, or something similar. But looks fine to me.

Something like this unlikely to ever happen. They will just put a counterfeit sleeve on some crappy capacitor. Even capacitance most likely would match, voltage rating probably not.
BTW, I think I found the listing you bought it from. When buying NOS, you should consider they easily may be 2-3 decades old.

[wraper]

Your capacitor seem to be 15 years old.

[b_force]

ah jeez, fake transistors.
They gave me quit some big headaches and sleepless nights.
Certain types are notorious of being copied, even being sold from well known distributors.
We had a whole production badge that totally failed quality control.

We couldn't find the problem until we decapped a few of these transistors.
The die was very different, especially MUCH smaller, so they couldn't get the heat away.

[sureshot]

The age of the capacitors might be an issue, but it's a gamble worth taking. I've used a lot of old but unused stock before from a family company in Sheffield here in the UK. Semiconductors are generally safe bet, but with capacitors, yes it's the chemical state of the electrolyte. So how to break the case of a TO218 case with out destroying the die ? Just so as the have a look see. I've got a far few tools to attempt it.

[sureshot]

ah jeez, fake transistors.
They gave me quit some big headaches and sleepless nights.
Certain types are notorious of being copied, even being sold from well known distributors.
We had a whole production badge that totally failed quality control.

We couldn't find the problem until we decapped a few of these transistors.
The die was very different, especially MUCH smaller, so they couldn't get the heat away.

That's not good, remember the capacitor fiasco in the pc industry, although it did effect other products as well. Some one in Taiwan stole some electrolyte that was unstable I read at the time. That ended up in the bad capacitors. Not sure that story is 100% accurate.

[wraper]

So how to break the case of a TO218 case with out destroying the die ? Just so as the have a look see. I've got a far few tools to attempt it.

Only by dissolving plastic by nitric acid, some very nasty stuff. By breaking plastic off the metal you still can see silicon die size despite it breaking.

[MK14]

That's not good, remember the capacitor fiasco in the pc industry, although it did effect other products as well. Some one in Taiwan stole some electrolyte that was unstable I read at the time. That ended up in the bad capacitors. Not sure that story is 100% accurate.

My understanding is ... (probably from the bad caps website)

They stole the formula, for making the capacitors. But unfortunately, they stole the wrong formula and/or did not realise, there was more to it. Otherwise the capacitors would only last a rather short time.
The rest is history.

I.e. There was a mistake in the formula, which they used to copy/clone the capacitors and/or the formula was right, but needed another step/process to make it last a long time. But without that final bit, it did not last very long.

Edit: searched for info:
https://www.badcaps.net/forum/showthread.php?t=425

1.1. What is the story behind the BadCaps phenomenon?

Well, the rather dubious story behind this problem is that a scientist working for the Japanese company Rubycon left the company and began working for Luminous Town Electric in China. He developed a copy of the Rubycon P-50 water based electrolyte which is what they use in the Low-ESR caps that are present on all motherboards. Unfortunately his staff left and stole his electrolyte formula. They started producing tons of the electrolyte and supplied many Taiwanese capacitor manufacturers. Unfortunately the formula was incomplete and did not contain the additives that prevent electrolysis from occurring inside the capacitors and releasing hydrogen gas which bursts the capacitor at the vents at the top or at the bottom of the capacitor can.

On the other hand, maybe the story is not 100% correct/true and/or telling the full/latest story ?
Because...

1.2. Should we believe this story?

It has gone on far too long to be a result of this one incident of industrial espionage. It still goes on today. Now bad capacitor manufacturers are just using cost cutting measures in manufacture and choice of raw materials in order to achieve the best price. As a result the capacitors fail early. One method used is to dilute the chemical mix of the electrolyte with water. Another method is to use natural rubber bungs instead of more expensive synthetic rubber ones. The natural rubber bungs go hard with age and temperature and cause electrolyte to leak from the bottom of the can. Furthermore impurities in the aluminum foil, the ph of the electrolyte and reduced aging procedure in manufacture are other reasons for failure.

[wraper]

They stole the formula, for making the capacitors. But unfortunately, they stole the wrong formula and/or did not realise, there was more to it. Otherwise the capacitors would only last a rather short time.
The rest is history.

Stole or did not steal, Nichicon made their share of bulging low esr capacitors as well (HN and HM series). Nippon chemi-con was affected too (KZG, KZJ).

[Lionered]

I hope I will now become a victim of fake transistors. It's really terrible to purchase something that will just damage our appliances so it's really important to purchase only from the trusted sources.

[szoftveres]

Does anyone recognize the logo of the "assumed" manufacturer of this transistor? It is the main transistor mounted on the back of a cheap adjustable power supply ($27 shipped) so I have no illusions of this transistor being genuine, just curious whose logo they printed on it.

https://www.newegg.com/yescom/p/0J6-0032-00042?item=9SIA8SK3S25897&recaptcha=pass

[wraper]

Does anyone recognize the logo of the "assumed" manufacturer of this transistor? It is the main transistor mounted on the back of a cheap adjustable power supply ($27 shipped) so I have no illusions of this transistor being genuine, just curious whose logo they printed on it.

IMHO some small Chinese company logo. Most likely with a very tiny die inside. Are glass insulators blue?