Chinese (Fake) transistor experiment

Electoneman

Member
2016-02-27 7:28 pm
I have a parts unit NAD 2100 power amp that I recently got going again. My experiment was to install any blown transistors with parts from Aliexpress and ebay and see how it would perform.

These were the parts that were ordered and installed:

two 2SA1302 (Toshiba output tranny)
two 2SC3281 (Toshiba output tranny)
two 2SA1011 (Sanyo)
two 2SC2344 (Sanyo)
several 2SA1370 and 2SC3467 (Sanyo)

The parts arrived after several months of travel from mainland China. The big transistors look very similar, but the connection legs are much cheaper looking and feeling than the genuine parts. The TO-92 package parts are virtually indistinguishable from the real thing.

After installing, I fired up the amp on the DBT and there was no smoke and the relays clicked. Both bias trim pots were set at the absolute minimum and centre voltage check was performed. Initial sound tests were also positive - amp sounded normal, transistors remained pretty cool to the touch.

Here is where the fun began. I decided to turn off the amp and adjust idle current to NAD specifications: 15mvDC (+/- 1mv). Bias adjusted normally, but I noticed that the trim pots were turned quite a bit more clockwise than I remember. I could start to feel heat coming from the A1302s and C3281s. I let it idle for a little while at this bias level and heat remained pretty constant. Turned the amp off, hooked up a pair of speakers and audio to the input jacks. Fired it back up and we had sound. This lasted for approximately 15 seconds before the right channel shut off and resistor R444 burst into flames :eek:! Left channel wasn't far behind as resistor R443 was starting to let out the magic smoke, but I shut the amp off before it caught fire too.

So, moral of the story is, these fake transistors will function in a certain capacity, but do not try to bias them. Of course, the true moral is avoid these parts at all cost unless the originals have no substitutes.
 
Not only this happens with chinese products. Texas Instruments Argentina in the 80's and 90's erased the label on TIP31 devices and re-labeled them as TIP41, say, a transistor of double current capability. Thus, when manufacturing (on those old times) 12V converters for use as emergency lightning with 20W fluorescent tubes, the literally blown out in few seconds of powering on the device. I don't know if they continued doing it still later, but here was a very important case. I believe that those frauds were only possible in abominable countries (Argentina) like mine.
 

Ian Finch

Member
Paid Member
2010-04-11 4:22 am
Coffs Harbour, NSW
I seems like TI didn't or couldn't exercise enough control or monitor the practices of the local division. Perhaps the relationship between parent and local companies wasn't a good one and there were a lot more management issues that went unchecked.

To my mind, that problem starts as an integrity issue, right from the top of the organisation. If the boardroom guys aren't committed to business integrity, you could not expect the lower management to fix that from their position, so corruption spreads and becomes endemic. It only takes one rotten apple......as the saying goes.