Biggest Complementary Transistors

Status
Not open for further replies.
Jump to Latest
E

EWorkshop1708

Member
Joined 2004
#1
I've been searching, online, but not much luck on real good info.

What are the biggest NPN/PNP bipolar transistor pair made? Don't matter what package.

I've seen here TO-3 devices that are 250W 16A, but I imagine there's much bigger.

Huge Complementary transistors can be used in more than just amplifiers. However, a pair of really big transistors would make a darn good amp. Could come in handy for all the big amp builders here too.

What are the biggest complementary transistors available today?
 
#3
I'm not aware of any 250W+ parts at all. That's easy to understand as the biggest standard package widely used is the old TO-3 and the plastic TO-3P - both limit maximum dissipation to the values you mentioned (250W which is by the way only a theoretical number as it needs to be derated and SOARed).

You need larger packages for more and of course uncommon packages are very costly, thus uneconomical.

Have fun, Hannes
 
#8
EWorkshop1708 said:
Huge Complementary transistors can be used in more than just amplifiers. However, a pair of really big transistors would make a darn good amp. Could come in handy for all the big amp builders here too.
Click to expand...

I would respectfully disagree.
Really big transistors require a great deal of compromises which leave them less than ideal for amps.
It's better IMO to parallel smaller transistors.
 
#9
Be carfull you're not looking at switching transistors. They typically have a higher current capacity, yet have limited SOA at larger Vce. Also linearity is more desirable than SOA, IMHO. I always wondered if one could use a device like that in a cascode arangement with error correction.:dodgy: One pair cranking out huge currents.😀


Unless you're really bored it would be a wast of time I'm afraid.
 
#10
Re: Re: Biggest Complementary Transistors

myhrrhleine said:


I would respectfully disagree.
Really big transistors require a great deal of compromises which leave them less than ideal for amps.
It's better IMO to parallel smaller transistors.
Click to expand...


CBS240 said:
Be carfull you're not looking at switching transistors. They typically have a higher current capacity, yet have limited SOA at larger Vce. Also linearity is more desirable than SOA, IMHO. I always wondered if one could use a device like that in a cascode arangement with error correction.:dodgy: One pair cranking out huge currents.😀


Unless you're really bored it would be a wast of time I'm afraid.
Click to expand...


I agree. Seemingly ideal and powerful transistors leave a lot to be desired when it comes to audio output stages. Often these transistors are slow, have low Hfe and low Vce. This is why parallel units of more suitable but lower powered transistors are used.

In terms of power handling a transistor rated at 250 watts might only be good for around 20 watts at Vce = 100v. In this case you would be far better off deploying a mosfet solution if you wanted fewer output devices.

Cheers
Quasi
 
#12
Basically, for high power handling levels it all boils down to high thermal transfer speed and consequently large thermal contact surfaces, no ?
Both MOSFET and LAPT devices adequately expose the misconception of minimal number output stages.

But the funny part is that both pro designers and diy folks still have the tendency of rather choosing large+few instead of small+many.
For example : Up to power levels of ~250W/8, the lower P-rating Toshiba 2SA1301/2SC3280 are superior to the popular 2SA1302/2SC3281.
As are the 150W Sanken 2SA1215/2SC2921, compared to the bigger brotha 2SA1216/2SC2922.
3 parallel pairs of the smaller duos have better SOA and thermal transfer capability than 2 pairs of the biggies.
Still, while the 2SA1301/2SC3280 could still be relatively easily found, and for modest prices, the mob continued chasing expensive 2SA1302/2SC3281 for many years to end up with fakes. :clown:

And then there's the bridged power amplifier concept: some folks rather bridge a couple of 150W/4 Leach amps for a total of 6 pairs of originale 2SA1301/2SC3280 that cost them $0.75 each than building one of them Double Barrel thingies.
 
#15
Greetings...

Mr Didden, i too would like to thank you for indicating this handy circuit and calculator.

I'd just like to ask, from the more experienced readers of this thread, a bit of a "sanity-check" of the spreadsheet. I did my best to adapt it for the MJL4281/4302 pair (as used in Mike Bittner's Symasym). I couldn't find the DC SOA graph in the datasheet, so i assumed some lower Ic values than the ones in the 1sec SOA.

Did i mess anything up? :smash:

(I'm attaching the "personalized" spreadsheet)
 

Attachments

#17
Khron,

I didn't check the SOA curves, but I did try with the 4 ohms reactive load and then you can see the SOA is too small. So, this amp would be OK for 8 ohms resistive and complex loads and 4 ohms resistive. For a 4 ohms reactive load it would be safer to use two pairs of output devices.

I did change some of the breakpoints to get more output power in reactive loads by crossing over into the 100mS SOA for reactive outputs

To check for two output pairs, you don't need to change the SOA. You can just used 8 ohms reactive load, knowing that with two pairs you can actually drive 4 ohms reactive.

Jan Didden
 
#18
Thank you for the speedy reply, Mr Didden.

I intend on powering a pair of Seas L15RLY/P (8ohm) mid-woofers, each with a pair of bridged Symasym's, at +/-35V or so. I'll be using OnSemi MJL4281/4302 power trannies.

The crossovers are active, so the amps will be driving the Seas units directly. Their minimum impedance is around 6ohms around 300Hz.

I'm attaching an "updated" spreadsheet.
 

Attachments

Status
Not open for further replies.
Top Bottom