Interesting video on fake chips

[Onion Ring Modulator]

https://youtu.be/k72SFBOZ_lw

[mac]

A few weeks ago I bought BF245B "fets" ... hfe: 250, NPN, much like 2N3904 

mac

[Rob Strand]

It's a great video.  IIRC, it's the one Mark Hammer posted in the Lounge a few years back.

[CheapPedalCollector]

I bought some 1n270's from Smallbear or Antique Electronic Supply (I don't remember which), fake. Some TL074's I bought from Smallbear were fake too.

[blackieNYC]

I've never experienced any fakes from Smallbear.  He always had a well-curated selection. 

[FSFX]

As regards the video, I watched it with great interest and took out my stock of Yamaha synth chips to see if the looked real or fake. Fortunately they all looked real according to the video.

[CheapPedalCollector]

I too have never gotten fake parts from Small Bear until then, they are really well faked (TL074), but fakes none-the-less.

The 1n270's were obviously fake, I'm pretty sure I got those from Antique.

[Phend]

So would getting, for instance, TL074 from Mouser, Newark or Digikey have a better reliability in not sending fakes ?

[Rob Strand]

So would getting, for instance, TL074 from Mouser, Newark or Digikey have a better reliability in not sending fakes ?

The biggest advantage the larger companies have is reliable sources.  They need to have traceability back to the source. I have no doubts they have been stung in the past. They will have processes in place as a second line of defense.  For example, a goods inwards inspection and people who know what to look for when they vet the parts.   They will also be aware of the latest faking tricks and key features from the current fakers.

As you move down the supply chain your part volumes become lower, your suppliers become smaller and less consistent.   More hands along the way so more risk.  Vigilant small suppliers will have reliable sources too but those sources could occasionally get screwed over.    That's exactly why hobbyists have the highest chance of getting fakes, they are the smallest customers.


Here's the type of things the bigger companies can afford to do.
I'm not pushing Mouser here just giving an example.

https://www.mouser.com/quality/

[Note 2012]
https://www.electronicproducts.com/mousers-anti-counterfeit-practices-among-industrys-best/

[Phend]

Cheap is cheap, what is your time worth, get real parts and pay the OMG 6 bucks in shipping.
Makes a real effect and let's you sleep at night.

[CheapPedalCollector]

Yes I buy most of my parts through Mouser as they are always genuine. But stuff that is surplus and EOL can be harder to obtain. I don't use those parts in new designs, just for repairing vintage things when there isn't a newer version available.

[Phend]

Question:
Are the fake ones always made from "old" ones or are they sometimes made "new" and just don't meet the original spec. ??  If made new why aren't they just made "correctly".

[FSFX]

Here are some links to some ways of detecting counterfeit chips:

https://www.circuitinsight.com/pdf/methods_detection_counterfeit_components_smta.pdf
https://www.electronicdesign.com/technologies/test-measurement/article/21185936/saelig-detecting-counterfeit-ics

[Rob Strand]

Are the fake ones always made from "old" ones or are they sometimes made "new" and just don't meet the original spec. ??

You can get just about any combination.   The video goes through most of the use use old cases.  Another option is to take any chip of the correct size and mark it with any part number - for the end user, those might as well be plastic with nothing inside.  I guess it's easier to mark real parts than to make convincing looking new ones.

As for new not meeting spec.  In most cases it's not that the part was produced as the target part and was tested and found to be out of spec.  Back in the 70's, 80's you got stuff like that.   These days is just outright marking a similar part with any old part number.  Hence all the JFETs with totally wrong parameters.   In some cases BJTs are marked with JFET part numbers, it sort of test out OK but under a finer test it doesn't.   This idea work for remarking old parts as well.

If made new why aren't they just made "correctly".

The bottom line is greed.   The companies have no respect for their own company or their customers.

They just make any crap because that's what they are set-up to do.   If you test it and it sort of works then that fends off some of the screening.  A lot of hobbyists would not be able to work out their op amp is not what they bought.

Look at many products today.  They are useless garbage that isn't really fit for purpose.  When I was a kid toy versions of products were better quality than the real thing now.  A large hardware outlet here sells many plastic products buckets, containers, plastic spray bottles.   They are all designed to fail.  The plastic on the buckets is so thin the rim and handle collapses when filled with water and in no time something fails.  The spray bottles have thin brittle plastic exactly where the stresses are.   It would take very little engineering and very little additional plastic to make better products but they don't.   IMHO, it's essentially selling counterfeit products (in that they bucket itself is counterfeit).

[amptramp]

At one company I was at, we subscribed to the GIDEP alerts (Government-Industry Data Exchange Program) which were set up to warn people of known fake devices and where they were obtained.  If you think fake components in stompboxes are a threat, you should see how bad it is in the military industry.  We got in one 74LS74 once that had 16 pins!  This is a 14-pin device.  This problem goes back a long time - in 1961, tube manufacturers were running into competition that was taking used tubes and selling them as new, so the industry agreed to make the bond between the ink used to mark the tube type and the tube itself so weak that it would wash off or disappear over time due to handling.

Trust no one.  Do a thorough incoming inspection and verification of the parts.  I usually test passive devices before I put them on a board.  So far, I don't think I have any bad semiconductor devices, but who knows?

[Paul Marossy]

OK I am going to preface this by saying I am an old timer that built his first working guitar effect Oct, 20, 2001. So here is my perspective on this:

That was very interesting and informative. Also very sad that this is going on these days with EVERYTHING. I built a DIY handheld o'scope a few years ago, a thing called "DSO Shell" and apparently the Chinese have cloned the hell out of something that appears to be a product that was originally made in China. So much so that you can't find a genuine DSO Shell anywhere. I mean the PCBs were exactly cloned, the plastic enclosure, everything. It works but you can never update the firmware (also cloned), because if you do, it will just become a light paper weight. I found out mine was a fake after I was already done building it. It works and I use it for some things.

A few years ago I bought a few LM386s off eBay and they all had oscillation problems. The ones I had bought at RadioShack ten years prior had none of those problems. I suspect they were pulled from old junk, were faulty ones they somehow got a hold of (the trash can behind the REAL factory?), or are poorly cloned chips made in a factory somewhere in China. 

Whenever you get an NTE "replacement part" this is exactly what you are getting, a re-marked component - but at least they aren't claiming to be selling NOS parts. I've never had any bad NTE parts which I was forced to use for one reason or another.

Personally, I find this BS of counterfeit everything these days to be very annoying. I see all the time posts in guitar related hangouts where people are asking if their MIM Strat is legit. I mean 10 years ago I would have said "who dafuq wants to clone a MIM Strat?" but you can't really say with any certainty that someone out there isn't actually doing that today. It's like deepfake videos where you aren't sure if what you're watching is real or not. I had no idea of the massive scale of this stuff going on now.

The non-PC term "Shang Hai'd" comes to mind in all this stuff. SMH.

So the take away from that video for me is that 50% of the stuff on eBay is likely fake, and about 10% of it won't work because it went bad during the re-labeling process. Good to know.

[CheapPedalCollector]

I've bought good parts from ebay, but you really gotta be familiar with markings, packaging, and other factors to tell if something is real or not. Especially markings on the bottoms of chips and the finish of the chips themselves and not just the laser etching. Real parts are usually laser etched and not painted for ex except for really really old stuff. Even then it can be hard to spot fakes.

Also some real parts can be dodgy looking, like panasonic caps with digitally printed labels can look sus.

[Paul Marossy]

I've bought good parts from ebay, but you really gotta be familiar with markings, packaging, and other factors to tell if something is real or not. Especially markings on the bottoms of chips and the finish of the chips themselves and not just the laser etching. Real parts are usually laser etched and not painted for ex except for really really old stuff. Even then it can be hard to spot fakes.

Also some real parts can be dodgy looking, like panasonic caps with digitally printed labels can look sus.

Yeah I have found good parts from eBay too, but it was stuff like old NOS germanium transistors and the like. Sometimes I buy just a few of something on eBay because I don't want to spend $15 to have my $15 Mouser order shipped when I can get it shipped to me for free on eBay. I think that bit me in the rear at least once.

They apparently also re-label chips by baking the fake surface on top and then laser etching it. Those ones are much harder to spot. The acetone trick will catch the poorly done fakes.

[Kevin Mitchell]

Even then it can be hard to spot fakes.

I've learned recently in my "Looks too new" thread, even fake parts can be laser etched nowadays.

You've mentioned getting fake parts from smallbear. If this was before they sold the company I'd challenge that claim as SB was known for purchasing parts directly from the manufacture when possible. TL074s are for sure within that category.
-Just thought it was an odd note. The chances would be the same as getting fake stuff from mouser.

What was the sign? Slow slew rate?

[Phend]

Latest News : March 16, 2023
South Korea to build ‘world’s largest’ chip center with $230 billion investment from Samsung