My Transistors, original or copy?

Claim made

Alright, I made a return request.
Seller finally replied to my message about all JFETs being out of spec.
His reply was (translated from German) "No returns accepted! Read the conditions! Good day!"

Informative reply about the terms, but not so much about the issue / bad specs. :D
I sense some guilt.

At least here in Finland selling something defective (unknowingly) to someone, claiming it to be in order, makes you responsible for making it up to the buyer. Selling something with a defect you know about, but do not disclose, makes it a serious crime.

Let's see what happens.

For your info: 10 x 2SK170BL / Ultra Low Noise / J-FET / N-Channel / Neu ! | eBay

Hi all,

We all know that eBay if full of fake J74/K170.
Anyhow I went and bought some from Germany. 20pcs for like 15€ to be exact.

I must say I have not seen this kind of K170BL before.
Thickness: ~3.65mm
IDSS: 14-19mA (on the 4 I measured so far)
Yfs: 30-35mS (tested 2pcs, 200mV change in VGS gives 6/7 mA change in IDS 6mA/0.2V = 30mS, right?)

Laser marking looks like top quality. There are no signs of sanding. No tiled front e.g.

Burr is not Toshibalike. (long and centered) It is short and towards the side, but also pretty much all around the device and thus hard to distinguish from photo.
If you look at the leftmost 2 you should see the burr on top of the devices.

I have received fake toshiba JFETs multiple times, but Yfs has always been way below the Toshiba specified 22mS typ.
What could these be???
Not many JFETs have that high Yfs.

I wanted to ask the experts before I make a claim :)
 
Member
Joined 2009
Paid Member
I recently bought a couple of 2SJ74GR from Reichelt.com (not eBay, big german online retail dealer!). As suspected they were fakes.
Crime? Possibly, but who is going to make a cause for a couple of bucks?
Toshiba certainly does not care as these are obsolete parts.
 
Sorry for talking about Finland all the time, but here we have something called like "consumerprotectionoffice" they would help if I got fake parts here.

But on a more global scale...what does Linear Systems say?
This fake junk certainly hurts their business. And they have probably purchased some rights for something if they make LSJ74 and LSK170 today... Just a though. I could ask them.

I recently bought a couple of 2SJ74GR from Reichelt.com (not eBay, big german online retail dealer!). As suspected they were fakes.
Crime? Possibly, but who is going to make a cause for a couple of bucks?
Toshiba certainly does not care as these are obsolete parts.
 
Member
Joined 2009
Paid Member
But on a more global scale...what does Linear Systems say?
This fake junk certainly hurts their business. And they have probably purchased some rights for something if they make LSJ74 and LSK170 today... Just a though. I could ask them.
I think they are just happy with that:)
New customers scared of fakes and believing that LSK/LSJ are equivalents to the Toshiba parts:D
 
Administrator
Joined 2004
Paid Member
Well, everything I have had from Linear Systems was excellent. I don't think they claim they are the same parts at all. They are close and they do spec their parts separately on their own. So there is full disclosure on their end.

ISC and like firms always made me nervous. Who knows who actually makes their products and what QC is like. It seems like anyone can hang a shingle and present parts to the world maked with whatever part numbers they care to use.

For other Japanese parts, both Fairchild and On Semi made great parts under license (?). They did measure identical to the original parts I've tested, these were all BJTs. Now it's all On Semi as they bought Fairchild.

-Chris
 
I get opa2604 but I am afraid these are not original because there are signs it is reprinted.

better picture I can not make. Came from eleleparts ebay seller from china.

regards

Here I have some better pic of mine order, I did see on the internet that it has different prints. pic two are the original ones.

Maybe I can test the specs.

regards
 

Attachments

  • IMG_09281.jpg
    IMG_09281.jpg
    9.5 KB · Views: 328
  • s-l1000.jpg
    s-l1000.jpg
    58.6 KB · Views: 332
Last edited:
Here I have some better pic of mine order, I did see on the internet that it has different prints. pic two are the original ones.

Maybe I can test the specs.

regards

Based solely on the looks, I would take the counterintuitive conclusion and say that the first photo shows the genuine item of more recent production. Yes, the ones with the "wrong" BB-logo. I will explain later.
 
Later has come now.

We have been using this opamp in a particular product for about 15 years or so, and they've always looked like this:

ScO5IZT.jpg


I know, the BB-logo looks odd, but they're always this way and they're always acquired from authorized suppliers. This is on the most recently produced batch.

But you're probably more used to seeing the BB-logo this way:

9iMjG48.jpg


Left 2008 vintage, right 2011 vintage.
2017 vintage ICs look like the one on the right. Any other exterior, and I'd assume them fake until proven otherwise. Especially the OPA627 is prone to being counterfeited as they are pretty expensive, even when bought in large numbers.

Do you notice anything about the lot trace codes on these ICs?
 
Last edited:
But beware... looks alone may be deceiving as the better counterfeiters will try to match them as closely as possible.

Even through this may look similar to the OPA350 in my previous post, the rough edges make me think this might be a counterfeit too..
And the lot trace code seems a bit too similar to that of the right OPA627 to be a coincidence...

L45Hbby.jpg


Buy from Mouser or Digikey and you'll likely get the real thing, buy off fleabay, and fakes are what you will get...
 
Last edited:
But on a more global scale...what does Linear Systems say?
This fake junk certainly hurts their business. And they have probably purchased some rights for something if they make LSJ74 and LSK170 today... Just a though. I could ask them.

Linear Systems does indeed continue some products that the big manufacturers made obsolete, but that doesn't necessarily mean they make them to the same standards. One particular product we use is the 2N4117 low leakage FET.

Those are not without problems. The application calls for extremely low leakage, and the A-brand predecessors (Silliconix, then Vishay) never gave us any problems. The Linear Systems version are reasonably OK, but some just have too high leakage and need to be replaced.

We also had a bad batch in which the bonding wire between the metal case and the pin wasn't connected in a large percentage.

Also from a certain date, the marking is missing a numeral, and this continues to go unnoticed now for years. LS QC isn't up to snuff...

But LS is just as powerless as any other brand to do something about counterfeiting. All they can do is advize people to buy from authorized distributors. And frankly, in the end it is our responsibility too. If we all stop buying fakes, there's no market and it will collapse. Yes, Mouser and Digikey and the likes are expensive, but that is the real cost of real parts, not just markups.
 
Last edited:
Absolutely right, nobody should buy fakes. Unfortunately there are many bad reasons to buy fakes :(
I sometimes have to buy parts of eBay. I always check them carefully and also claim them when needed.
Unfortunately eBay policy makes claiming fakes very hard and time consuming. (also expensive)
If component can be found from official supplier then I 100% buy from them!

I reported a few K170 listings for a few days back, I guess eBay is not gonna do crap about that.
How about we all report obviously 100% fake components on eBay. Would that motivate them to take action?

Linear Systems does indeed continue some products that the big manufacturers made obsolete, but that doesn't necessarily mean they make them to the same standards. One particular product we use is the 2N4117 low leakage FET.

Those are not without problems. The application calls for extremely low leakage, and the A-brand predecessors (Silliconix, then Vishay) never gave us any problems. The Linear Systems version are reasonably OK, but some just have too high leakage and need to be replaced.

We also had a bad batch in which the bonding wire between the metal case and the pin wasn't connected in a large percentage.

Also from a certain date, the marking is missing a numeral, and this continues to go unnoticed now for years. LS QC isn't up to snuff...

But LS is just as powerless as any other brand to do something about counterfeiting. All they can do is advize people to buy from authorized distributors. And frankly, in the end it is our responsibility too. If we all stop buying fakes, there's no market and it will collapse. Yes, Mouser and Digikey and the likes are expensive, but that is the real cost of real parts, not just markups.
 
Hello guys !

I have two TIP3055 from 2005, are those with thermal grease inside the hole. I bought more TIP3055 and TIP35C and TIP36C from several stores, but someone seems a fake, you can see a different terminal design in the exit of case, and more "sharp" at the end, the back are very bright, and others are more opaque and have a different design. Its really counterfeit or they are diferent batch/site of production

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 
A word of warning if somebody happens to stumble upon the (ample) stock of 2SK170BL / GR @ the Dutch distributor's Huijzer.com - I bought the lot (+ some other Toshiba parts), and promptly returned it, as what I received was... [see the attachment].


The font and the GR parts' weird date code were a dead give-away that these are not genuine, but I did measure a dozen or so of each - the GR parts' IDSS was in range, but the BL were all *way* above 12mA + both had poor transconductance [13-15mA/V].


The "manufacturer" even went as far as faking the packaging - the boxes / antistatic bags looked like Toshiba's, but all references to Japan were replaced with Thailand. :yuck:


Huijzer didn't agree with my findings, but I did receive my money back in the end. However, the stock is listed on their website again, so buyer beware...
 

Attachments

  • Font_comparison.jpg
    Font_comparison.jpg
    313.8 KB · Views: 404
Last edited:
Member
Joined 2007
Paid Member
I'll post more details shortly, but I ordered a small amount of BF862s from China. They were sold as "unmarked" BF862s but with my jeweler's visor, I could make out the markings and they did not match the NXPs I bought from Avnet a couple of years ago.

I have a locky_z curve tracer and I took a sample of each. The ones from China weren't even close and there was no typical curve for Id vs Vds at constant Vgs. The NXP graphs were what I expected. :mad:
 
Administrator
Joined 2004
Paid Member
Hi petoluk,
Yes, the laser engraved numbers are a dead giveaway as well. Any of these old transistor numbers, BJT or JFET, are fake if the numbers are neatly laser etched.

The K369 and right hand K170 lettering looks correct. I didn't check the date codes. I usually buy a low quantity to test, then buy more if they check out okay. They almost never do. Buyer beware!

What do I do with the fakes? I use a 4 pound hand sledge hammer that does a fine job of distressing the cases. :) It's not worth having fakes turn up unexpectedly, or having another hobbyist "score them" from my garbage. I assure you, mine are not useable if they are fakes. It is very important that you destroy fakes. Even if you pay a lot for them (I know it hurts, but not as much as having to troubleshoot non-working projects because of a part you knew was a bad one).

So what should someone do? Use a currently manufactured device from On-Semi or Fairchild (bought by On-Semi). Or use a current product from Europe.

-Chris