FAKE TRANSISTORS (CB or QRP type finals).. testing being done by Earl Andrews VE3AB.. to make
sure I don't end up using (or selling) bum transistors. New Articles just being written 22 December
2013. This article is not finished yet..neither is my testing! I will be doing much more on this topic in
the next few days and weeks ahead.
I bought two batches of 20 2sc2075 transistors for a little over $1.00 a piece from a large supplier on ebay. That was over a month ago. I just started
testing them now and the results of the test proved that these transistors are highly suspect. Basically they are going to be held ready for destruction. I
may try using some of the ones that tested "ok" in experimental circuits..but the ones that are completely bad (no power output) and the ones that gave
only "some" power output -- will be held in a bin for destruction and marked clearly with a big X scratched into the plastic. -- When I worked in Fed Govt ..I
think the stock parts used to be (quarantined) and held under a stock holding code-- Iam not the supply system but I do supply some hobbyists and ham
radio people with parts..so my system will be to first test them thoroughly to make sure they are bad or flaky and then scratch a big X across the plastic
body next to the part number.
Pink market transistors are flaky ones that only put out about half or less of the power
output of the good tested transistors. One transistor is already marked with an X above the
part number. All transistors in this bin are bad ones I have tested thus far. Im only part way
through the order of parts I received over a month ago. I have not sold any thus far.
THE good transistor bin. The three in the middle with green marking are the ones that were
tested from the batch of 40 I ordered from ebay over a month ago. The MRF476 to the left is
one I pulled from an Argonaut 515 final output board. A qrp final PA transistor. The
2SC2075 is apparently the same or very similar to the MRF476. The MRF 476 in my testing
is being up held as the 'ideal transistor'of the lots that I am testing. I know it works.
The other 2SC2075s (C2075)s with the letter T on them..seem to all be good. I bought them
several years ago from Dans Small Parts in USA. I had ordered MRF476s but he substituted
2sc2075. They all seemed to work and Ive used them in my ARgonaut 515 as final PA
transistors and they performed well.

The MRF 476 was pulled from the board pictured. I dont know when it was manufactured but
it works well and I am comparing all others I am testing..to it.

I retested the green ones the next day and actually had to reclassify two of these good ones
to duds and color coded them pink.
Good output. Full output for my testing. This is the OHR wattmeter on the 1 watt scale. I am
experimenting with the circuitry right now and decided to run the transistors at a lower power
setting. It gives good test results and I verify the test results through a verification process of
plugging transistors (good and bad) alternating often to make sure the test is accurate.
Output from a "sick" transistor -- I color coded them pink. Most of the
transistors in the batch of 40 I ordered from the orient (ebay) were almost
zero output on the meter. Id say right now..(early into the testing process)
at least 60 percent of the 2sc2075s were complete duds. Maybe 20 percent
gave some reduced output and 30 percent (roughly) seem to be good...or
at least produced the full scale power output in my testing. CAUTION..THIS
IS EARLY INTO THE TESTING PROCEDURES. -- I plan to order some more
transistors to test and as well..I plan to perhaps modify the test circuit and
as well..try some of the good (new ebay transistors) in the Argonaut 515
final PA (if I can scrape up a pair with similar Hfe test results).
Test circuit is a simple class C transistor circuit. I use a plug in
transistor socket. The RFC for the collector is about 55 Uh.
Test frequency (at this time of testing. 7.050 into a OHR qrp
wattmeter) and a dummy load.

The Oscillator (transmitter) is a Ten Tec Argonaut 515 transceiver.

I plan to move the frequency up to 10 meters for the next round
of testing on all these transistors. I had read that some fake
transistors seem to work OK at lower frequencies but then the
performance tails off at the higher frequencies..ie the CB band
and 10 meters.

So alot more testing will be done soon and results will be posted
in this article (soon).

If you go to buy transistors from me..I want you to get good tested
ones. I may not be able to test each and every single part..but I
want to do some good spot testing to find bad batches.

This batch of 40 from ebay I bought a little over a month ago is a
BAD BATCH! -- NONE OF THESE 40 TRANSISTORS WILL BE
OFFERED UP FOR SALE.

I also plan to test some 2SC1969s and some used pulled
transistors from old cbs and other transistors in my collection.

I'm really looking forward to this testing and Im really curious
about how many bad parts are being flogged!!!
Here are some of the LOT of 40. There are 4 containers of 10. Packaging is good. Price
was low..a little over a dollar each transistor. -- Quality looks to be very poor and certainly
not trustworthy. I will be studying some videos on you tube concerning this problem and
reading about it in the next little while. Id be interested in hearing from you on the subject.
MY
email is LINK HERE..thanks and 73 earl ve3ab
www.earlandrews.com or www.hamelectronics magazine.com (mirror sites). operated by Earl VE3AB since
2005. I took early retirement and these activities keep me active and interested in things.

I sell tested parts. I buy parts too and work with them (as you can see from the above article). To see my
inventory of parts for sale (and trade) go to my main web page and look up the LINK
TO MY SUPERLIST of
PARTS.. (link here).. HERE IS A LINK TO MY
MAIN PAGE..ham radio experiments and antenna experiments too.
I will publish the schematic of my test jig when I get it more
refined. Right now it works but some refinements may be
needed..in particular for higher frequency testing.
23rd December 2013... continuing on..testing the rest of this batch of 40 transistors I bought from
ebay.. (2sc2075s which are the equivilent of MRF476s (I believe the 2sc2075 is the Japanese
version of the MRF 476).. only 1 out of 40 transistors provided proper amplification!! a real bad
batch of transistors!! They are still being offered on ebay by the same company ..a lot of 20 for just
over $20 -- the price is certainly right ..BUT THE QUALITY IS NOT THERE!
Today I am using an Argonaut 505 as the test transmitter. The 505 is a bit different than the 509 and 515. For one thing..the CW transmit power is back
panel adjustment. Since in this test..I have to adjust the drive power (7.050 mhz in this round of testing) to the point where the transistor in the test
puts out 1 watt full scale (that is for a good working 2sc2075) ..so I have to sometimes make adjustments to the drive and I put a knob on the back panel
power adjust pot. You can see this knob in the photo.
The test jig is pictured above. In the picture to the right..you see a 47 ohm 2 watt resistor. This was to connect from the BASE of the transistor
(also the RF input from the test transmitter) to ground. I ommitted the connection and the test circuit worked fine without the 47 ohm resistor
connected.  .01 uf capacitors couple the input and output signals. The 13.5 volts to power the circuit goes through the 56 uh choke. I noticed
several times during my testing run..that this choke got a bit hot. When this happened ..I gave the circuit a rest. Its a very simple class C
amplifier. RF input turns the transistor ON and output is monitored by the OAK HILLS RESEARCH qrp wattmeter set on the 1 watt setting.
MY GOOD TRANSISTORS (the MRF 476 old Motorola one with gold leads) and the batch of 2sc2075s with the letter T on it all work well and the
meter on the output of the qrp wattmeter goes to full scale. Maybe that T stands for Toshiba..I dont know. All I know is those transistors all work.

THE BATCH OF 40 TRANSISTORS (HOWEVER)...a horse of a different color!! ONLY ONE of the 40 produced a near full scale reading. About half of
the 40 produced a rather anemic power ouput of 1/2 to 1/4 scale and about the other half of the 40 produced little or no output at all!

Actually-- this is much worse than I would have expected! -- The only good news in this story is that I didnt sell any of them to anyone! I try and
test my components before I send them out. I dont (usually) test every component ..but in the case of CB or QRP type RF transistors..it seems
like I should. -- There are reports Ive read on YOU TUBE and on various web sites about counterfeit components. These transistors I just tested
go beyond counterfeit. I had one technical fellow tell me..he didnt mind counterfeit components if they worked in the circuit! -- Well these just
plain are trash!

I was cataloguing and working with MIL SPEC COMPONENTS where I used to work in the military electronics components division. MIL STANDARD
883 TESTING requires testing of comonents like these ones (ie transistors and ICs) in a test jig. Used to be 1,000 hours of burn in in a circuit.
Then they modified the burn in time to only 160 hours (something like that) at a higher temperature. I think it was 125 degrees C test
temperature. --- So these components would not hold a candle to the MIL SPEC components.
The one transistor that did put out some power
in the class C amplifier test circuit is pictured to
the left.

It did not quite produce as much power output
as my GOOD TRANSISTORS..but it was close. 90
percent or so of full power.

This was at 7 mhz.

I plan to run through some of these transistors
again ..this time testing them up around 28 mhz.

From what Ive gathered so far on the topic of
fake transistors (from you tube and various web
sites)..the fake transistors (especially for CB
people) tend to drop off power output as the
frequency goes up.
In other words..at the CB range of
frequencies..the transistors just dont give any
amplification.

So..very soon..(it is Christmas)..and I will be
relaxing myself..and not working so hard!! but I
will soon in the next week or so..try these
transistors at 28 mhz.

Also in the works..is a batch of 2sc1969s that I
bought from ebay over a year ago. I want to try
them in a test after I finish with the 2sc2075s.


24 December 2013.

Just went on Yahoo Search and did a search for
"fake transistors"-- came back with a whole slew
of hits..

Here is but one comment " There is also a ton of
easily recognisable fakes on eBay with bad
lettering and obvious respraying, ...

I will be doing more work in this area in the very
near future as I screen my own stocks.

I SELL TRANSISTORS but I will be
SCREENING/Testing them first.


FAKE PHONEY TRANSISTOR PICTURED BELOW
I test my transistors..so you dont have to worry ..if you
go to buy some from me..they will be tested.


I am always looking to buy old stock transistors. I would
really like to buy old stock from before the Chinese
started manufacturing them and all this fake transistor
problem was not nearly so bad.

You can email
me earlinclayton@yahoo.com

thanks..73 and happy holidays..Earl Ve3ab
The ones color coded pink..did put out some power during the test. Usually less than 1/2 the power and often much less. The ones on the
right side all put out no power at all.

As I said before..MUCH WORSE THAN I EXPECTED!! -- Now Iam glad I only do small orders and I keep my buys small! -- Usually I only go out
and buy small amounts at a time. There have been exceptions however. One time I bought 600 MV1662 varactor diodes from ebay. I took a
chance but I tested some of the varactors and they tested fine. This batch of 40 transistors is a dead loss.
Here's a close up of one of the 40 bad fake counterfeit transistors. I
have noticed..on ebay..some parts have M for Motorola brand or the
two squiggly lines for the NATIONAL SEMICONDUCTOR logo or perhaps
some other top of the line logo.
These are the transistors I try and buy. In with these good brands there
will be other listings on ebay with funny looking logos that I dont
recognize. I avoid these listings. -- I still buy from China. I have chosen
a few companies that seem to have good track records and I have dealt
with sucessfully in the past.

This lot of 40 - 2 lots of 20 - 2sc2075s was a departure of mine. I went for
it because of the low price and thought Id try them. I wont name the
company ..because I dont know the legal repercussions of doing so.
I dont know this LOGO on this transistor. Not sure who made them. All I
know is I will avoid this logo from now on.

C2075 is a short form part number. The whole complete part number is
2SC2075. I beleive this is a Japanese part number- or at least it was
originally.
Next up for my testing will be
these 2sc1969s.

These are common transistors
used in CB radios and QRP radios
as RF amplifer for the transmitter.

Sometimes they are used in pairs
and sometimes alone.

I have about 20 I purchased over a
year ago.

Iam most curious as to how these
will perform under testing.
The 2SC1969 has the same physical pin out as the 2SC2075. So I will be
searching my stocks for a "GOOD OLD 2SC1969 " PROBABLY A PULL FROM
AN OLD RADIO SHACK CB OR SOMETHING. -- I want to find two good old
ones or maybe more. Then I will try the new ones I bought from overseas
to see how they compare against a known entity. Stay tuned for more
Transistor Testing --73 Earl VE3AB
click here to go to PAGE TWO of this presentation.. 24 December 2013..
---------------------->>>>>>>>>>>
which shows the results of me testing the 2sc1969 transistors

(they all tested GOOD FULL OUTPUT..except for one that had a defect)

When I sell parts like these..I always test them first and I would have
caught this defect with my testing procedures. 73 Earl VE3AB
The circuit schematic above. It is about the simplest circuit there is..that will work. It works as a
class c amplifier. RF drive comes in from the exciter. In this case..I use a Ten Tec Argonaut 505 with
low power output of about 100 or 200 mw or so and the output is about 1 watt or so ..with good
transistors. When a transistor is bad or flaky ..the output is zero or maybe 1/2 to 1/3 watt or so.
Coupling capacitors are .01 uf ceramic or epoxy. The transistor socket shown ..I probably got some
at a flea market. They work well for TO 220 style transistors.
I was looking through a local Ontario, Canada on line swap shop for hams - KWARC swap shop (very popular) and saw this posted today
(13 May-2014);

''quote:

Thank you for all your help and advice. Some advice to you all....
I purchased RF transistors from CHINA on 2 occasions. Both times the parts were "re-packaged" TO-220 parts with the part number stamped on
them. The first set of 4 were in fact FETS. Not RF transistors at all. The second set were not even a transistor at all. I broke one apart and there
was ABSOLUTLEY NOTHING INSIDE !!! And I am not kidding.
---------->>> STAY AWAY FROM E-BAY AND CHINESE RF PARTS------>>> -- unquote"


Myself: I wouldn't go so far as to say - stay away from ebay and Chinese parts completely"-- but be sure to test them if you can..and
MY PREFERENCE..is to buy old surplus parts if I can --

I BUY OLD COLLECTIONS OF PARTS..SO IF YOU HAVE OLD SURPLUS PARTS..PLEASE EMAIL ME. I KEEP MY STOCKS SEGREGATED AND LABELLED
AS TO WHERE I buy items. I do buy from China and ebay but I test and guarantee the parts I sell. my email is
earlinclayton@yahoo.com