I'd like to see the biggest transistor. Any contenders that you know of?

This popped into my head while I was painting my porch today and I thought it would be easy to answer. Not so.

I've looked a bit online but couldn't find much. This stupid thing keeps popping up.

There could be more than one answer; physically largest dimensions, highest current, highest power.

Where would you expect to find really big transistors?

 

Hello,

The following transistors might interest you.
The highest power I found is 400 Watts:
http://nl.farnell.com/semelab/mag6332/trans-audio-npn-230v-30a-to-264/dp/1856783
The highest current I found was 120A:
http://nl.farnell.com/stmicroelectronics/esm2012dv/darlington-transistor-sot-227/dp/9802290
The highest voltage I found was 1.05 kV:
http://nl.farnell.com/nxp/buj302a/transistor-npn-4a-1050v-to220/dp/1972410

Bertus

 

The 1000A current supply I used at NIST had six 150A NPN transistors in parallel with something like 0.1mΩ ballast resistors. (We won't talk about where the last 100A came from. ) Each transistor was roughly the size of two decks of playing cards.

 

Thyristors are actually 2 transistors in an NPNP structure. So not exactly transistors. But they do get BIG. Here are some stacks of big ones used to convert HVDC power back to AC. These are used with long-distance under-sea power cables. Big enough?

 

I've used ones like this in the past.

http://www.galco.com/buy/Fuji/1DI200A-120

http://www.galco.com/buy/Fuji/1DI300ZN-120

This type pre-dated IGBT's. Not strictly single transistor - more darlington like array.

 

t n k brings up something I was going to ask, do IGBTs count as transistors? Since they are a npn with a mos "base".

 

Why wouldn't they count as transistors? It's a hybrid of a bipolar transistor and a field-effect transistor, so why couldn't the result be a transistor. And, while not definitive, the T in IGBT is "transistor".

 

These are used with long-distance under-sea power cables. Big enough?

That's a lot more like what I expected to find. 200A is an impressive transistor but I was expecting room-filling behemoths like these. Interesting.

 

4700A IGBT. I'm pretty sure I have run across larger ones before (10kA sounds right), but I can't find them now.

I almost bought a lot of 1200A 1200V IBGTs on Ebay one day. They were about the size and shape of 3" thick phone book.

 

t n k brings up something I was going to ask, do IGBTs count as transistors? Since they are a npn with a mos "base".

It seems to be common useage to use the word transistor to mean only a bipolar-junction-transistor (BJT), but the word applies also to MOSFETs, IGBTs, MESFETs, JFETs, etc. Notice they all have the word transistor in their name, which is a generic term for a three (or more) terminal semiconductor device that can perform linear gain (as opposed to SCRs, etc. that can't).

 

Since it's only a question of idle curiosity, I'm accepting the widest possible definition, basically anything that can be called a transistor. The guts inside are irrelevant.

 

The largest are probably IGBTs - with the maximum Die ( chip) ratings around 4500V and 200-300A - in production ( I am sure some higher V versions have been built for research).

Most of these are used in modules (ref: http://search-ext.abb.com/library/D...LanguageCode=en&DocumentPartId=&Action=Launch ) witch closely couples anti-parallel and Freewheeling diodes - due to these high currents and fast turn off times ( to minimize switching losses) the chips can generate high V when turning off due to parasitic inductance and high dI/dt rates. The higher current of the modules 2000A - are due to paralleleing of the chips in the module, it is not uncommon for on module to have 20+ chips ( IGBTs and Diodes)

The larger systems mentioned are not single devices - but many devices paralleled.

The most common are the 200-300A 1200 & 1700V - this is reverse blocking V - so the application voltages for these are ~800 and 1100V DC Resp. - these are used in modules operating as a single "switch" of up to 3600A - functioning at about 500KW.

 

4700A IGBT. I'm pretty sure I have run across larger ones before (10kA sounds right), but I can't find them now.

Cool! I wonder how high in current we can get. I've found a 1200A darlington.

 

The higher current of the modules 2000A - are due to paralleleing of the chips in the module, it is not uncommon for on module to have 20+ chips ( IGBTs and Diodes)...

Yes, I think the silicon die size is a limiting factor. Of course some huge silicon wafers have been made but maybe not used for making a giant transistor.

 

It seems to be common useage to use the word transistor to mean only a bipolar-junction-transistor (BJT), but the word applies also to MOSFETs, IGBTs, MESFETs, JFETs, etc. Notice they all have the word transistor in their name, which is a generic term for a three (or more) terminal semiconductor device that can perform linear gain (as opposed to SCRs, etc. that can't).

Agreed.

IIRC, the term "transistor" is short for "trans-resistor", basically capturing the concept that the parameters at one port control the effective resistance across another port.

 

Cool! I wonder how high in current we can get. I've found a 1200A darlington.

the 1200A is peak current (still quite impressive) but the Infineon IGBT mentioned earlier is 4700A continuous or 7200A peak

 

Definitely a bad moment to read this thread: I was planning to use today a 2N2222 for some basic tests with a LED.

Feeling bad now.

 

Awwww... Is your one little?

 

have a look at this
http://www.roadsideamerica.com/story/8180

Dieter

 

Definitely a bad moment to read this thread: I was planning to use today a 2N2222 for some basic tests with a LED.

Feeling bad now.

It's not the size of your transistor, but how you use it.