Fake small signal transistors?

Hi everyone.

I thought I'd pick up some misc small signals for repairs and playing around. I found kits with typical used transistors on Ebay/Banggood for almost no money.
But I am worried they are fakes, because almost all transistors have the same datecode/batchno "331" even when they are very different types. Also, i do not see any manufactorer markings.
I bought some 2n5551/2n5401 that also have this same marking. Unfortunately I noticed this too late.

Should I be worried, or is there an explanation for this. Please see the attached picture ( from banggood )



  • Fakes.png
    321.1 KB · Views: 742
Last edited:
I would rate them as very likely fake ¯\_(ツ)_/¯

Did you receive them yet? Are they shipped?

I bought and payed for the 2N5551/2N5401 yesterday. They will probably be here january. See attached picture ( from Ebay )

At first I thought that since most manufactorers are moving more to smd, they might want to empty their old stock for the most common parts. I am probably wrong here :)


A side note, but related:

For the laughs of it, see this link. It's for sanken power transistors. Very cheap price at $2.60 for a set !. But if you scroll down, you will se a red note, saying:

1pairs OR 2PCS Transistor SANKEN MT-200 2SA1295-Y/2SC3264-Y 2SA1295/2SC3264

Important notice:These products were made in China,not made in Japan.If you require high quality products which made in Japan.They don't match your requirement.Pls don't buy them in this case,or we don't accept return and refund.

And just below this read note, it says:

1. Own storage with large stock components, 100% genuine.
2. Top quality and Best price

1pairs OR 2PCS Transistor SANKEN MT-200 2SA1295-Y/2SC3264-Y 2SA1295/2SC3264 | eBay


Should one report that they are knowingly selling fakes?


  • Fakes 5551 5401.png
    Fakes 5551 5401.png
    238.7 KB · Views: 535
for almost no money
What else did you expect?

Ever heard of Mouser, DigiKey, Farnell and similar "serious" suppliers?
I buy nowhere else, period.

"hey, but these were 1 cent each, with free shipping"
Then live with it.

Hi JMFahey

You are offcause right. I made an honest ( stupid ) mistake. I am however not expecting much. It's mostly for playing around anyway. I won't be ruined if they blow up in my face ( unless i go blind :) )
These 5551/5401 will come no where near our amp projects :D

We have been lucky before though to get genuine parts from Ebay. Mostly we use Mouser or Farnell. Haven't tried DigiKey
Oh, I´m not dissing E Bay per se, there must be thousnds of good sellers too.
But some sellers, well ... look beyond dishonest.

Funny thing is they boast of:
99.3% Positive feedback
Get fast shipping and excellent service from eBay Top-rated sellers.

BUT ... how did they get that?
Simple, thy arm-twist buyers to rate them 5 stars, no matter what (which to me is way worse than just selling dubious parts):

Seller does not accept returns
We offer 14 days’ unconditional return.

do not leave negative FB without communication,we will block and no more business.

If item(s) arrive in satisfactory condition, please take your time to leave feedback to us,we hope to win your all 5-star positive feedback.

Pls DON’T leave 1, 2, 3 or 4-star Detailed Seller Ratings because they mean "you are NOT satisfied", equaling negative feedback.
International shipping
Economic Int’l shipping :has no tracking number,without insurance.

Oh well.
I always try to find a seller that has free returns, good ratings, and has been around for a long time, regardless of what I am looking for.
This is by no means a guarantee that you actually get what you buy, but it might help a little.

I also have a tendency to believe, that sellers who deal with all sorts of things from bicycles to clothing and makeup are less reliable, than one who sells only a certain type of merchandise.
By reliable I mean to have genuine stuff… components in this case.

If you want cheap genuine transistors you can always salvage them. Most of the common types don't tend to degrade unless they had a fault to begin with or something egregious happened to them. There are particular types however which are almost always bad due to some error in the manufacturing process allowing moisture ingress or similar.

It's probably luck of the draw, but I harvested about 50 BC547, BC557 and similar from an old TV just recently. All tested good, although a few had quite low hFe I got some which were over 600, and they still have nice log legs, I guess chopping them short wasn't the fashion. I sorted them all by hFe for later use.

There were also a few high voltage variants, a beefy driver and output which might work for a power supply at some point. And loads of high temp tinned stranded wire.

Don't get me wrong it's pretty time consuming but for me this is a hobby and that's kind of the point anyway [emoji28]
If you want cheap genuine transistors you can always salvage them. Most of the common types don't tend to degrade unless they had a fault to begin with or something egregious happened to them. There are particular types however which are almost always bad due to some error in the manufacturing process allowing moisture ingress or similar.

I used to do that a lot earlier. But now a days there isn't a lot of equipment coming through here anymore.
I am however not expecting much. It's mostly for playing around anyway. I won't be ruined if they blow up in my face ( unless i go blind :) )
These 5551/5401 will come no where near our amp projects
I wear safety glasses when first powering on, after the TO3 metal caps bounced off the ceiling.
Seriously, farnell has sales on end of bin or out of fashion transistors. 2n5401/2n5551 are too popular to ever get below $.35, but MPS8099/8599 have similar specs, same package, and I got rolls of 100 for $8. Use the max/min on vceo, Ic, package, see what you come up with. I got MPSA55 (60 v) for $2.30 a hundred in the great On semi TO92 shutdown when they bought fairchild. 2 watt TO92L MPSW56 were $11 the hundred, but to match them for drivers I have to buy MPSW06 for $.25 ea from a surplus house (which are probably dodgy)
Watch the specs; I bought from farnell 50 On semi MJL21194 for $1 with a 250 mw rating. It wasn't a misprint, they were obviously rejects. KABLAM! on 60 v Vceo.
Stepping further afield, by using farnell max min selector, if I need 270 ohm 500 mw resistors, I can pay $6 the hundred for first line vishay 20%, but I can get a hundred 1% 272 600 mw for $.80.
I don't see similar bargains at digikey or mouser.
Last edited:

Hi again

After looking a bit more into it, i stumbled over a posting of 2sa733 that has the same x331 marking as those in the picture above.
According to seller, these are made in China, by a company called "Wahkitsing"

They have a homepage here: 华捷盛电子有限公司

I tried to find more info without much luck. Maybe one of you know more?
Their online order system looks funny...
Should I be worried, or is there an explanation for this. Please see the attached picture ( from banggood )
I have bought the same, from the same source, not expecting too much considering the price, but I have been pleasantly surprised: it could be that most of the types differ only by the laser marking, but basically they are generally up to their specs: I have used them in projects (they are posted here, you can check for yourself) for more than 1 year, and I didn't encounter real problems.
In fact the opposite might be true: since they are one-size-fits-all, they are overspecced compared to the real, original parts.
The basic ratings are right (voltage, current ...) but secondary parameters not directly specified in the DS are generally much better than the originals: transition frequency, Early voltage, etc.
This could cause problems in some circuits not designed for such performance.
In fact, they are a modern process, having nothing to do with the original, but broadly complying with the DS.

As an example, I have tested a S8050 in the Nonoiser project, and it did better than an authentic Philips BC337 on noise, I have built a HV power supply (Elektria) using MPSA42 and 92, a discrete HV opamp working from +/-135V with the same, etc.

They might be a bit weak on dissipation, because the amount of material in the case and leads is minimal, but other than that, they are OK, if you know what to expect (if you use them as a replacement of older devices, you might have problems due to the higher gain and Ft).

These observations are for transistors of one source, and it does not mean that everything you find on Ebay is going to be similar: in fact members sent me samples because they suspected they were fake -and they were-.
They were medium power transistors mainly
Hi Elvee.

Thank you for our answer. That was good to know. You mentioned testing.
And I came to wonder if any of you know of any diy curve tracers that are simple and affordable.
Preferably one that can save it's results to a spreadsheet or a usb pen for later processing. Would be a nice tool for matching bjt, jfets or checking for fakes.

Here is a link for such a project https://www.instructables.com/id/Transistor-Curve-Tracer/

Unfortunately it does not do all i need, but it's close and does not need a oscilloscope
Last edited:
One cheap test for sub-standard transistors is Vceo test. One doesn't have to test at the actual Vceo rating. I have tested using a 12 vdc supply, 3 clip leads, and ma microa scale of a DVM.
Found some fakes today, out of a pile of TO92 transistors removed from a 2019 replacement board from an organ supply for a 1990 swell engine. I found one leaking in circuit, power up reset never timing out. Tested 2 removed parts today, npn with 12 v from c to e base open, was drawing 2 ma through a 47k resistor.
Nah, goes 3 ma b-e forwards. There is a label on the part for these weird little TO92 ECB transistors. Replaced them with BD639/640 from Newark, all 34 of them. BC639 were full price $6.10 a hundred, but 640 were starred sale items, $2.70 a hundred.
By contrast 2n3772 from M date 9131 leaked 1 microamp. Keeping those in equipment.
The lady that built us this board is not an engineer, all those guys retired. She probably took the old number from the BOM and went hunting on google. You can buy any number you want on a TO92 package, but they probably are not transistors. Shade motor drive was trying to go forwards & backwards, fast & slow, all at the same time. No discrimination from decision network when the transistors all leak. The old board, the transformer wire insulation was pierced by a sharp connector pin back & caught fire. Burned a big hole in paper board. Pretty alarming when something in a pipe organ is smoking, a whole room full of wood.
Last edited:
Were those the funny little “top hat” TO-92-ish looking things, with a round flange at the bottom? The ones with a hard plastic case that the lead frame/die assembly goes into, and the filled with “goop” to seal it? Every single one I’ve ever seen goes bad eventually, probably from moisture ingress. I see them in input stages, and when they go bad it makes the amp sound like a thunderstorm.
These looked just like real black TO92, probably bought this year right after we ordered replacement board. No round part at the bottom. Label square 414, next line 0007, next line ECB. Looked just like the bc639/640 I got from newark, only those were real from "multicomp", newark house brand.
Organ parts co only charged us $169 for board, real nice but no electronic engineers anymore. 4th call they sent me the schematic so I could trace out the problem. Step 1, PUR stuck in wait forever.
Last edited:
Hope that was just a typo (in post #14) -- 2 mA from 12V through a 47k.;)

A 6000 ohm resistor across 12V draws 2mA. A 47k resistor only 255 microA ..

Was attracted to this thread 'cause for a while I've been wondering how often an alternative pinout factors into cries of, 'Fake, fake, fake, fake ..'. Seems to me there are quite a number of different 'generic' TO-92 p/n's that are available 'flipped' (still center Collector), but with Base and Emitter on opposite sides. It was consumer electronics service in the early 80's before I ran into the center Collector; we started calling it 'the Japanese pinout'. But I had no idea at the time that there was a 3rd option popular in Europe.

How 'bout sharing us some wisdom, youse guys ..?

More transistors

Hi. Hope you all had a good xmas/newyear.

Before xmas i also ordered some 2SA1145 and some 2SC2705.
They finally arrived, and look like you can see in the images.

Comparing to Toshiba's datasheet the mechanical measurements are
within spec. So are the markings. Hfe is also within spec ( done with a cheap multimeter )

What bugs me is, that npn and pnp have the same batch no. I have actually never seen this, and can't tell if they are genuine or not.




  • thumbnail_IMG_20200120_174942.jpg
    167.3 KB · Views: 225
  • thumbnail_IMG_20200120_175112.jpg
    150.9 KB · Views: 217
  • thumbnail_IMG_20200120_175208.jpg
    143.9 KB · Views: 213