The question is really clearly answered by the very small physical size of the die in post 1221: My Transistors, original or copy?
Ok, thanks. I have been offered a partial refund and if the remaining 8 fail testing then a full refund. I am going to ask what tests the seller will be performing. Note to self, never trust eBay.
Real C5200’s will measure about half of those ON Semi parts. About 6 to 8 nF. The ON Semi C3281 clones (and the higher voltage versions) are probably the biggest dies out there for audio BJTs. Bigger than the MJ15 and MJ21 series TO-3’s. If you’re ever in doubt about SOA when replacing Jap outputs you almost can’t go wrong with the ON 3281 or 4281. The 0281 is more in-line with current production Toshiba and most Sanken in terms of die size. It is often recommended as Jap replacement when trying to keep composite fT as high as possible in critical applications (like single digit ppm capable amplifiers). They also tend to match well NPN to PNP. If a 150 watt transistor will cut the mustard, they are a very good choice.
Hell, al I use for a capacitance meter is the capacitance function on my $95 DMM. Works well enough to identify phony transistors, match capacitors and identify some with ambiguous (or worn off) markings. Don’t go thinking you need an expensive one to do this test. It is the BEST way of identifying fakes if you can’t afford to go busting a few of them up. Your results were very reliable.
About “all fakes being the same” - definitely not so. I’ve already posted some pics above, depicting what was inside several MJ15024’s with identical markings. Not even close to the same die. Some were the usual 2N3055-ish parts. One was even a genuine Motorola, something from the 2N5631 family. Some had white goop, some didn’t. But the tops came off all of them way too easy. I also have a batch (big batch, actually) of fake Toshiba C3281’s with *two* 2N3055 size dies inside. They measure about 4 nF base-emitter. Getting close there. They don’t have squat for gain at low current so useless for audio. I don’t trust them to have good SOA because one die will current hog. Useable anywhere a 2N3055 in a flat pack would be useful. I kept them for such usage. I have them clearly marked and segregated into my B-stock.
Hell, al I use for a capacitance meter is the capacitance function on my $95 DMM. Works well enough to identify phony transistors, match capacitors and identify some with ambiguous (or worn off) markings. Don’t go thinking you need an expensive one to do this test. It is the BEST way of identifying fakes if you can’t afford to go busting a few of them up. Your results were very reliable.
About “all fakes being the same” - definitely not so. I’ve already posted some pics above, depicting what was inside several MJ15024’s with identical markings. Not even close to the same die. Some were the usual 2N3055-ish parts. One was even a genuine Motorola, something from the 2N5631 family. Some had white goop, some didn’t. But the tops came off all of them way too easy. I also have a batch (big batch, actually) of fake Toshiba C3281’s with *two* 2N3055 size dies inside. They measure about 4 nF base-emitter. Getting close there. They don’t have squat for gain at low current so useless for audio. I don’t trust them to have good SOA because one die will current hog. Useable anywhere a 2N3055 in a flat pack would be useful. I kept them for such usage. I have them clearly marked and segregated into my B-stock.
When I read this, a question arises that apparently no one has asked here:
How do transistors from currently manufacturing for audio differ from old (NOS) versions within one manufacturer in detail (there are a lot of threads concerning fakes/counterfeits or not) ?
The question arises in order to the fact, that the currently produced quantities of classic Class AB and A versions have become vanishingly small compared to the years 1965 to 1990.
What would also be interesting to know how small professional manufacturers like Pass Labs, Classé, Bryston etc. who still use classic circuits for their home audio products from currently production (i. e. no class-D) perform their choice and do perform their quality check for the later used power BjT/MOSFET power devices for the serial manufacturing.
It may be that the quality that was used in the past is no longer available ?
In my opinion, manufacturers such as Toshiba, SANKEN, Semelab and Motorola earn very little in the meantime from discrete audio semiconductors.
If my assumption is correct, it is best to get amplifiers from electronic scrap as a parts supplier for NOS transistors for the aim of repair.
Last edited:
What has happened in the industry is consolidation. It turns out that very few different types are actually needed for 99.99% of production. So the other .01% has to go scratch. New types replace all predecessors at relativistic speed (especially Mosfets), and a manufacturer just uses what he can get *today*.
I pick up scrapped, pulls, and NOS surplus whenever I can. As long as the price is reasonable. And it has been dwindling over the last 20 or so years. Ive also done lifetime buys of the parts I have designed into my stuff already. Many are already long gone.
I pick up scrapped, pulls, and NOS surplus whenever I can. As long as the price is reasonable. And it has been dwindling over the last 20 or so years. Ive also done lifetime buys of the parts I have designed into my stuff already. Many are already long gone.
They should be giving a full refund. Can you post a link to the ebay item? Some of the fake specialists are very well known.
Also don't forget that older processes are being cancelled, or phased out. The parts that are kept are migrated to the newer processes and sometimes we see an improvement in performance. Then we have newer transistors that are generally superior to what we used to have. It is too bad many signal transistors are no longer available as leaded parts. Adapter PCBs are the answer for this. Expensive, but better than using a fake or copy that isn't as good.
-Chris
-Chris
A friend tell me, that only this two brands produce Audio Power Devices, which are reliable:
1) ONSEMI
Discretes & Drivers
but here I miss a lot - and
2) SANKEN
Bipolar Transistors ｜Sanken Electric
the manufacturer with the largest range. Supplier:
Sanken - Manufacturer Page | Profusion
Under
Toshiba Infrastructure Systems & Solutions Corporation
one don't find any discretes like 2SC5200/TTS5200.
But here is the overview:
Transistors | Products | Toshiba Electronic Devices & Storage Corporation | Asia-English
TTC5200 | Bipolar Transistors | Toshiba Electronic Devices & Storage Corporation | Asia-English
1) ONSEMI
Discretes & Drivers
but here I miss a lot - and
2) SANKEN
Bipolar Transistors ｜Sanken Electric
the manufacturer with the largest range. Supplier:
Sanken - Manufacturer Page | Profusion
Under
Toshiba Infrastructure Systems & Solutions Corporation
one don't find any discretes like 2SC5200/TTS5200.
But here is the overview:
Transistors | Products | Toshiba Electronic Devices & Storage Corporation | Asia-English
TTC5200 | Bipolar Transistors | Toshiba Electronic Devices & Storage Corporation | Asia-English
Last edited:
Well they look like a local seller based in Australia. So I am afraid I have never heard of them since I am not in that area.
When I look at that listing I am concerned that I see that "Toshiba" has been obscured on the photo of the product. I don't consider that a good sign. I would not expect a legitimate seller of genuine Toshiba products to blur out the brand on the photo of the product. Many of the questionable sellers on AliExpress seem to have this practice of removing or blacking out the brand on their product listings.
But I go back to the small die size. That doesn't look right at all. The seller should know that isn't right. If their business is audio kits and they buy 500 Toshiba output transistors at a time surely they know something about the product that they sell.
I suggest an authorized distributor of an appropriate part number that is currently in production.
When I look at that listing I am concerned that I see that "Toshiba" has been obscured on the photo of the product. I don't consider that a good sign. I would not expect a legitimate seller of genuine Toshiba products to blur out the brand on the photo of the product. Many of the questionable sellers on AliExpress seem to have this practice of removing or blacking out the brand on their product listings.
But I go back to the small die size. That doesn't look right at all. The seller should know that isn't right. If their business is audio kits and they buy 500 Toshiba output transistors at a time surely they know something about the product that they sell.
I suggest an authorized distributor of an appropriate part number that is currently in production.
Seller might not know any better, really. And if the seller does buy real parts (from an authorized distributor) the kit will cost real money.
Buyers who run it on a low voltage transformer (because it’s what they have or can get cheap) might not even know something is amiss.
Buyers who run it on a low voltage transformer (because it’s what they have or can get cheap) might not even know something is amiss.
I agree. From observing my failures it appears to be the current that kills these fakes ? They have about 55V on them from the Denon PM-700V.
The right 1943 might be genuine. The rest are fake, the surface is rough from the original marking being abraded away.
Here On Semi 5200 about 100 Rupees each, I am helping a man with a damaged Denon too.
Sanken was taken over by Sanyo, which was split up, the semiconductor business was taken over by On Semi.
I have no idea if Toshiba still makes these.
The perforated emitter design on the On Semi device seems newer, how it affects the sound quality I have no idea.
These are 30 Amp units, try and find a higher rated one, from a genuine source.
Equivalent American and European devices are also available I think. Look for those, a lot were used in TV and amps.
Just a suggestion...
Here On Semi 5200 about 100 Rupees each, I am helping a man with a damaged Denon too.
Sanken was taken over by Sanyo, which was split up, the semiconductor business was taken over by On Semi.
I have no idea if Toshiba still makes these.
The perforated emitter design on the On Semi device seems newer, how it affects the sound quality I have no idea.
These are 30 Amp units, try and find a higher rated one, from a genuine source.
Equivalent American and European devices are also available I think. Look for those, a lot were used in TV and amps.
Just a suggestion...
Better finish...maybe the guy abrading it used finer grit paper!
Stick to known sources, and put slightly higher rated parts, space permitting.
Also, the postage difference will be negligible, and your device will run much within its limits, less chance of failure.
It is like asking a small car and a truck to climb a steep hill every day... the truck will fail a lot later than the small car.
Also, check why it failed.
Stick to known sources, and put slightly higher rated parts, space permitting.
Also, the postage difference will be negligible, and your device will run much within its limits, less chance of failure.
It is like asking a small car and a truck to climb a steep hill every day... the truck will fail a lot later than the small car.
Also, check why it failed.
you don't like links in general in the postings - therefore now no longer added links.
What about the quality of ISC (e. g. 2SC2922) and MOSPEC (large BjT/MOSFET spectrum) ?
Last edited:
Hi tiefbassuebertr,
-Chris
Well, that isn't entirely true. I don't like seeing a lot of links in posts with very little text. I know you are trying to be helpful, but sometimes you just post a series of links. I think some text along with fewer links would be more helpful and people would be more likely to follow the best links that you find.
I'm not sure what you are asking. You have provided specific examples, but they are two different families of device. So I will guess. I prefer using BJT transistors for output devices, but often use Mosfets for power supply regulators and low frequency switching power supplies (<100 KHz). At higher frequencies the drive requirements for a Mosfet transistor become higher and the circuit a lot more complicated. So, you assess each situation and use the more suitable device. Is that what you are asking?
-Chris
I suspect that ISC and Mospec both “make” bipolar and Mosfet transistors. Both make a lot of generic types and my experience is only with Mospec, and all of it just the bipolars. Mospec parts that ie use have been pretty generic - old 2N types, which are not that hard to copy properly. I’ve cracked some open, and the innards look reasonable. I’ve never used anything like an MJ15024 from them, so I don’t know if they put in anything “better” in such types. ISC may also make some good copies of generic types, but their Sanken copies are well documented to be absolute $#**. Their MT200’s do not employ any of the proprietary construction techniques that Sanken uses, and I’d be more inclined to believe they are just using generic epitaxial dies than real LAPTs. The word for that is “fake”.