How to test transition frequency of a transistor

[AndreyS]

I've bought used guitar amp Yamaha G100B-212. And it contained two non-original parts: power transistors 2n6254.
I've bought original 2x Toshiba 2SC1080 from China. It was said, that they were refurbished. As the seller later explained there were pulled out from other electronic devices.
They looks original, but anyway I'd like to put them to the test.
Is there an easy way to test them? And how to test properly a transistor meant for audio purposes?

[AndreyS]

My thoughts on this matter are that transition frequency and linearity are the main points. Is this correct?

[magic]

from China

Well, I would be more worried about SOA than all that "performance" nonsense, but to test it you would need to mount them on a good heatsink and get them hot.

I suppose it's not a bad idea to test beta linearity, fakes are likely to show poor performance and perhaps low overall beta to begin with.

As for fT, I suppose some signal generator and scope will be required. How high are you trying to measure?

[SparkyFX]

..in the range it is meant to be operated in.
Both types do have some differences (pulled from the next best datasheets for these types)

	2n6254	2SC1080
Vceo	85V	100V
hFE	5-70	40-140

There are expensive component testers out there ofc that do the job for you, building one requires some way to plot the results, either measure several points individually and plot them graphically one by one or by automated test equipment and scripting.

[T3sl4co1l]

Audio amp isn't going to care.  Put it in and check if it oscillates on a full amplitude sine wave.

Tim

[magic]

If I buy something from AliBay etc I would rather know what it is for sure before plugging it in

Just because the seller appears to show honesty by telling you that they are recycled parts doesn't mean he isn't lying or his supplier didn't lie to him. I have seen offers of supposedly recycled expensive opamps which I'm quite convinced are fakes, looking at the photos. I'm actually going to order one as a curiosity so hopefully I will have more to say in the future. Call me nuts, I started to collect fake opamps when I see them for cheap

That being said, your transistors may surely be real. I have successfully bought recycled, genuine vintage parts from China. I just don't think it's the best idea to put a perfectly working device at risk by installing parts from suspicious source.

Did the vendor at least have some positive feedback for this particular or other very similar items? And by positive feedback, I mean somebody who actually used the device and reported success, not just "item received, thank you".

[Audioguru again]

Rod Elliot at Elliot Sound Products says that old Toshiba audio output transistors are most certainly counterfeit. I think most transistors bought from ebay or Ali are fakes.

[floobydust]

It seems to have a quasi-complementary output stage, Sziklai pair on the bottom for TR310, TR312.  These can be finicky and oscillate at 10-15MHz which is a PITA. It sounds like a gritty distortion that gets worse with volume and shows up on the scope as a burst of RF for 1/2 cycles, at the speaker output. Doesn't happen at low volume.
I would take care substituting these two transistors for similar hFE and fT as the originals. The others are not critical in the circuit.
Modern audio power transistors are have a much faster fT ~30MHz so I would not use them. This schematic shows 2SC1586 (O,Y).
Your eBay parts are likely counterfeits and you can tell looking under a magnifier/microscope the font is not quite Toshiba, the surface finish and seal might look a bit cheap.
2N6254 is really old and slow, fT=0.8MHz like the old 2N3055's.
MJ21194 is what I would consider, very tough and fT=4MHz.

[magic]

I found some basic information on those 2SC1080, they are supposedly 4MHz too.
Maximum ratings: 110V/12A/100W, SOA unknown. Beta 40-140, as somebody reported before.

[ratatax]

Would they really bother making fake oldschool transistors then making them look used by scratching the surface ?
I don't know, maybe yes if they sell them for $100 each.. but I guess it's not the case here

[magic]

I would say they look genuine, most fakes are lower effort, but who knows. The scratches may come from storing and shipping a dozen of them in a bag.

The thing is that the amplifier works and is likely made of unobtainium, if something blows up it will be a bad day.

[Kleinstein]

The transistors are nothing special - so chances are that these are not fakes, even if from China.
Testing them before use may still be a good idea.

The 4 MHz fT is more or less standard - so even a fake / wrong type may very well be in that ball park. Much to slow or too fast may cause oscillations in a given audio amplifier circuit. This could also happen with different capacitance - so it's not just fT. If the amplifier is not oscillating chances are it would be OK.

If they are fake / wrong type, the main parameter to worry about is the voltage rating / SOA (e.g. if they are actually 2N3055).
So the interesting point is more like how to do an SOA test. No need to record a full curve and find the real limit. It is more like give it a stress test (e.g. 1.5-2 A at 60 V for a few seconds) and see if they survive.
The difficulty may be for the test circuit to survive a blowing / runaway transistor.