2SC5200 / 2SA1943 toshiba vs mjl3281 / mjl1302 on

This Cob comparison is apples to oranges, and not really an issue. An examination of the Cob curves shows that these devices are very similar.... I designed an SOA limiter that lets me use the lesser part (hopefully).
hi all, Russel,

each designer have one goal in mind and decline others parameters under his criterias. It can be, performances (fast Tr), intensive use (+SOA), low cost...

Putting a limiter to use low cost transistors, is not my first think. It may alter performances ?

SOA, Cob and others parameters make TTC/2SC5200 attractive for many common designs, aren't they ?
I have a preference for TTC, may I'm wrong ?

But with pairing process, I can understand that buying 100pcs for a project can be very, very unplaisant :tongue:
 
When I say the "lesser part," I mean I'm using the NJW0302/0281 instead of the NJW1302/3281. I think you understand that, but I'm making sure.

I'm putting in an SOA limiter whether I use the higher power part or the lower power part. The cost in sophistication is that I had to put in a 2-slope SOA limiter to use the lower power part. I went to a lot of trouble to design an SOA limiter that does not alter performance with 8 or 4 ohm loads. I can even run 2 ohm resistor loads, without limiting, but the limiter does affect THD there.

The NJW0281/0302 have similar SOA to the 5200/1943 transistors, and I believe they are sonically equivalent, too. In the United States, the ON parts are usually cheaper.
 
You really need two (or even three) slope SOA limiters regardless of which parts you use. Single slope is pretty useless except for the lowest lower levels (20 or 30 watts) or very minimal amounts of protection in larger amps. If you find yourself running on the edge of limiting you need to add another pair.

The Fairchild version of the 5200/1943 may be cheaper. L and N case versions are available as well as the Fairchild house numbers of both. Prices seem to change all the time so what’s cheapest today won’t be in a week.
 
The Guardian. The shortage of chips, which are key elements in any electronics, began last year. Initially, the problem was associated with delays in the supply of chips against the background of factories closed due to the coronavirus pandemic.
Despite the fact that now production has returned to normal operation, the growth in demand for electronic products, caused by the changing habits of consumers in the wake of the pandemic, is now leading to the fact that the shortage of chips is reaching its crisis level. The rise in car electronics, booming sales of TVs and personal computers, the launch of new game consoles and the mobile industry's shift to 5G wireless devices have all caused a surge in demand for semiconductors and exacerbated their shortages.
According to Microsoft forecasts, supply problems will persist until at least the second half of this year.
Recently, the South Korean manufacturer said it could delay the release of one of its flagship smartphone models due to a shortage of semiconductors, even though Samsung itself is the second largest chip maker in the world.

“It's hard to believe that Samsung sells $ 56 billion worth of semiconductors to the industry every year and spends about $ 36 billion on them, but has to delay the release of one of its products,” Kempling said.

According to one of the CEOs of the company Dong-jin Koh (Dong-jin Koh), now in the IT industry there is a serious imbalance between supply and demand in the market for semiconductor components. Despite Samsung's best efforts, the problem is far from resolved.
Analysts believe that the chip shortage could drag on for up to two years. This is how much time may be required to restore a complex semiconductor production. Meanwhile, the chipmakers themselves are going to significantly raise the price of their products for the second time in less than a year.

“So far there are no signs indicating an improvement in the situation or a decrease in demand. At the same time, prices are already rising throughout the supply chain. Ultimately, all of this will affect consumers.
News, sport and opinion from the Guardian's US edition | The Guardian
 
I have heard, that Onsemi's MJ21193/MJ21194 (unfortunately obsolete - go to
https://www.mouser.de/ProductDetail/ON-Semiconductor/MJ21194?qs=HVbQlW5zcXWF%2B7NuGDf/lw==)
together with various power output devices from Sanken - go to
Bipolar Transistors |Sanken Electric
currently the best choice for output power buffers in power amplifiers.
To avoid getting counterfeits order Sanken's power transistors for repair service and new diy projects only here:
Transistors (Buy Online) | Profusion
Sanken - Herstellerseite | Profusion

I have heard, that Toshiba's 2SA1943/2SC5200 also obsolet as genuine part from Toshiba, but still listed under
Transistors | Products | Toshiba Electronic Devices & Storage Corporation | Europe(EMEA)
go to
https://toshiba.semicon-storage.com/info/docget.jsp?did=20427&prodName=2SA1943
https://toshiba.semicon-storage.com/info/docget.jsp?did=20668&prodName=2SC5200

What about the versions from UTC (Unisonic Technologies) ? - go to
https://www.profusionplc.com/parts/2sc5200l-o
http://www.utc-ic.com/uploadfile/2011/1101/20111101104026379.pdf

check out also this thread:
bipolar (bjt) transistor families for audio power output stages
 
Last edited:
IRF640/9640 are older, smaller TO-220 case.
My assumption is that 5 pairs can do up to 100W amp with 50V rails..

I Had seen a commercial Branded amp that uses only a pair of Irf640.9640 tested on actual( FTC Like test) with 8 ohms dummy load resistor and full 1khz sinewave till it clips( almost 100w) and it withstand for almost 15 mins with no smoke.
That time I was intrigue so I took a peek in the inside and shocked to see that it only uses a pair or IRF640/9640 for each channel. This is where i was convinced that mosfet's are also durable for audio.