Man I'm soo sick of fake transtors

[defaced]

As TO-92 parts go obsolete, they will become counterfeited. TO-92 J201s are a recent example.

I also suspect that doing small samples prior to a large order will be fruitless because there are no rules to this game. They may be making up fakes as orders are placed, or keep a stock of working parts on hand to selectively sprinkle into the buying pool to keep faith up, or mix fakes and good parts together. So on one order parts may test good, but that may not apply to subsequent orders.

Given that this problem has wide reaching implications in industry and military (and subsequently profit for those making the fakes), I don't suspect it's going to go away any time soon. Dated, but relevant (IMO):
Fake electronics becoming military danger - CBS News
http://defensetech.org/2011/11/08/co...p-8-posiedons/

[diydidi]

Fake people are like fake transistors. They don't last long. - diydidi

[soundguruman]

Fake people are like fake transistors. They don't last long. - diydidi

On one hand, there are plenty of fake parts from China.

On the other hand, a part is only as good as the technician who installs it.

[diydidi]

Any of you come across some more fakes? Share the part numbers and sellers..
I took a chance and bought some STD03N and STD03P transistors from Hong Kong. When I got them, I tested them and cracked a couple open to check the size of the die compared to a known good one. NOT FAKE!!
Seems like the fakers haven't gotten round to faking these 5pin devices yet.
Also there is a Chinese company copying a whole bunch of transistors. They are called ISC. They look and actually work great-
so far the only "other" manufacturer I trust.

[tedmich]

FYI Rod Elliott has a great page on this, though not too recently updated
Counterfeit Transistors

[soundguruman]

The ECG and NTE cross reference books, and the Japanese transistor manual, are some of the best resources.
These will help you find transistors with similar or matching characteristics.

Not necessarily that you have to use NTE.
But it helps you find two different numbers, that will do the same job.
And it helps you find the voltage rating, current and pin-out diagram.

After some practice, it will be easier to find substitutes that work acceptably.

So, the number usually does not have to be an exact match, to work right. If the rating is correct ...

Then you no longer have to buy transistors from China...that might be fakes.
You can buy good ones...that will work as a substitute....from USA, and Japan...etc...

And it does take some practice and patience and careful research.
But, eventually, you can get pretty skilled at it.

FETs in some circuits will be harder to substitute. The characteristics are more critical...
and there are a lot of special purpose FETs, that are hard to substitute.

In one of my oscilloscopes, the inputs are FET coupled. And there was no available exact match.
There was no substitute that worked correctly, critical application.
So eventually I found a surface mount FET, with the same exact specs, and soldered leads to it.
When I installed it, the problem was fixed.
That's an example of how you "could" substitute an FET, when nothing else will work.
And here it is, 9 years later...and it's still working.

A 35 cent part fixed the problem,
but it took 16 hours to find a part (in stock) with the exact same specs as the original. (which was the only part that would work in a critical application)

So, you will probably learn how to do that, as you go along.

[g1]

The ECG and NTE cross reference books, and the Japanese transistor manual, are some of the best resources.
These will help you find transistors with similar or matching characteristics.

Not necessarily that you have to use NTE.
But it helps you find two different numbers, that will do the same job.

Trying to "back-cross" using NTE or ECG data can be dangerous. They use subs. that are rated same or better, so you may end up thinking a generic part you have is higher rated than it is.
Here's an example:
generic part 2N5XX3, actually 5A 100V ; NTE says use their NTE2x3 which is 6A 120V.
Now say I don't have either of those but in my bin I find generic 2N4YY2 which NTE says NTE2x3 is the sub.
Now is 2N4YY2 actually 6A 120V? No, it turns out it is actually 4A 80V. I put it in the amp and it blows up.

NTE guide will NOT tell you the spec of your generic part! It will almost always tell you a higher rating.

[tedmich]

this place sometimes works

All Transistors Datasheet. Parameters and Characteristics. Transistor Database.

[soundguruman]

I've done it for decades.
How did I "ever" get by without your negativity?

"I put it in the amp and it blows up."

I put it in the amp, and it keeps working.
So, there's the difference.

[gui_tarzan]

Back in the early 80s when I was fiddling around with ICs and transistor projects I used the NTE & Jameco cross references quite a bit. The only source I had at the time besides mail order was Radio Shack and if I remember correctly there was a good xref there too. Of course that was back when RS was still a hobbyist's hangout, now they mostly sell phones and computer crap. The backward incompatibility might explain why some of my projects didn't work when they should have and why some of the components blew up. I hate the smell of burnt silicon, it smells like wasted money.

SGM, there's a difference between ignorance and stupidity - I'll let you decide which one fits you the best.

China wasn't a player back then though, we at least had some good components to work with if not the actual replacement parts.

On a side note, there was a discussion in another thread about Chinese quality, or lack thereof. I recently repaired a Marshall made in India (I think?) and a Vox made in China. Granted the circuits were likely designed by American engineers to be as cheap as possible, but the build quality (although weak on both fronts) was terrible on the Vox. It was embarrassing. I even fixed an assembly flaw in the cabinet, that's how bad it was.

One factual thing we know is the Chinese aren't afraid to copy whatever they want and sell it as equal to the original. Our government seems to be afraid of China when it comes to faulty or illegal copies of products and we pay for it. It's disgusting.

[gui_tarzan]

How much more effort would it take to make them correctly?

I have asked that question about model car kits (another of my many hobbies) for years. One example: Revell released a 1969 Dodge Charger kit as a "new tool" and they had to replace the body and windows because they made the roof wrong. If I'm in charge of a half-million dollar tool I'd be darned sure it was correct before the first kit was released to the public. It frustrates me but it is what it is.

[km6xz]

The bashing of China does not make sense. Sure they make cheap gear and parts because that is what most companies want to sell to under their own brand name but there are other companies which spec a higher quality and get it. Board level reliability actually improved in the industry when China came on-line as a contracting source. Equipment in most fields got a lot more reliable. There are several reasons but one that stands out is the national trade policy that established that China would dominate electronics production by strong central government support of loans and grants for highest tech automation. No one can really compete with China because the length of time a production line retains a technology type is very short compared to companies in the west which need to amortize the value of production equipment over a longer period. That is what killed Mackie. Remember when the 1604 came out, it killed the competition, 1/2 the price and twice the performance due to Mackie investing in higher end board fabrication equipment. The precision of boards allowed them to be mounted in a way that simplified and strengthened the boards instead of vertical strips like was used by the popular UK boards of the time. One board for the whole mixer. The problem for Mackie was they had a major investment in production automation that became obsolete compared to the Chinese pc board makers which kept up with the latest hardware, which continually lowered cost and speed of production. Behringer and others were making more margin on cheaper boards.

Don't blame China if the brand you like is specifying the lowest price and level of component grade. Other companies are not cheapening out and are spec'ing high quality assemblies.
The point is, China will make as good or as poor as you contract with them to build. There are not very many brands of anything that do not use pc boards made in China, from cheapest toys to aircraft or F1 race cars. I have had gear made in China and would do it again. I needed guide radios, audio systems for our guided tours and for renting. No one made one optimized for guide systems that were strong enough, reliable enough and cheap enough to use every day, get dropped, stepped and abused. I designed the receivers and transmitters 6 years ago and contacted some companies to bid on it. The one who won was based on the sample boards they sent. They were sturdy fiberglass, thick double sided boards about the size of a pack of cigarettes. I got the first batch of test runs in only 4 weeks and the first installment of 1000 radios in 2 more months, and 2500 more 6 weeks after that (They all could have been sent but we did not have the money all at once). They have been tough and profitable, with only a few each year needing repair. The rental business took off because they were the most reliable around. Another feature that made the rental business take off was a talk over drinks at a party with a upper management person of the Hermitage museum to whom I was talking about how much of a better experience it would be if all larger tour groups had guide radios so none of the guides would have to talk so loud. That was something they had a lot of complaints about. The made a rule that any guide with a group of 7 or more people had to use a guide radio. There are other rental companies but these units have the best reputation because of how reliable they are. Mostly, they are reliable because I targeted a budget of $32/receiver and $29 for the, about twice what brands of similar radios were willing to pay for their radios that they retailed for over $120 each.
If a amp company wanted to increase the quality of their models, they could do it very easily by cutting down their magins a little (which are 3 times as wide as they were when they built their own gear). Figure a Bugera amp has a manufacturing cost of about $50 and the they retail for $500, they have a little room to step up the quality ladder.

The claim that all they do is copy is bogus as well. It takes some serious engineering chops to make something so cheap yet with increased performance and have less than 4% return within the warranty period. They are graduating more electronic and CS engineers each year than the US has in its entire population.

[J M Fahey]

I've done it for decades.
How did I "ever" get by without your negativity?

"I put it in the amp and it blows up."

I put it in the amp, and it keeps working.
So, there's the difference.

Man, you deserve the

[Justin Thomas]

Off topic, but I couldn't help myself...

Homer "Hey! That word you keep calling me... what is it?"
Artie: "Ignoramus?"
Homer: "Yeah. It means I'm stupid, doesn't it?"
Artie: "There is a difference between ignorance and stupidity."
Homer: "NOT TO ME THERE ISN'T!!!"

I generally stay out of the pissing contests, but that one's a davorite debate exchange. I'll shut up now since I don't know anything about transistors...

[bob p]

I think they concentrate on "the great amp transistors" because these are commonly bought for repair and DIY purposes by small tech shops and hobbyists who will be fooled easier than, say, Dell would by a batch of fake RAM chips.

I think you've got a point there, Steve. It looks like they do focus on the "great transistors" from the "golden era" of high power transistor amps. And they probably do this because they are exploting the knowledge gap between design engineers and small service shops and between design engineers and hobbyists. There are a lot of people out there performing repairs, who just look for a specimen of the original out-of-date transistor rather than spending the time to pour over data sheets to see what type of current production transistor is a proper replacement.

When I decided to rebuild a bunch of phase linear amps, I decided that it wasn't worth the effort to try to source and match a bunch of extinct Japanese silicon. Instead, I spent some time studying the data sheets and I found a current production, exact drop-in replacement that had twice the SOA. Once I realized that I could populate the amps with a transistor that was a drop-in replacement (didn't require the driver circuit to be modified) and met performance specs while having twice the SOA, I decided to buy a couple of trays direct from the manufacturer. I ended up ebaying the orginals and the guys who knew what they were scarfed them right up. I probably could have charged twice as much for them if I were greedy.

If the hobbyists and small tech shops were willing to spend the time looking at data sheets and doing the math, there would be no market for the counterfeit "great amp transistors." The only reason that the counterfeiters even have a market to sell into is because there are a great many people who don't know enough to know that they shouldn't even be looking for the "great amp transistors" of yesterday. Today the smart guys have figured out that with today's smaller die sizes and better thermal profiles you get double the SOA. that means that the "genuine article" isn't worth using any more.

If everyone knew this, there wouldn't be a market for the genuine "great amp transistors" or for the counterfeits. To me it seems that the counterfeiters are expliting the buyer's lack of knowledge in two respects: 1) inability to differentiate betewen authentic transistors and fakes, and 2) inability to realize that you don't really want the authentic transistors in the first place.

Just my $0.02.