Silicon Carbide chips

[Mike Page]

Tantalising ... chips running at 650 degrees C:

http://news.bbc.co.uk/1/hi/sci/tech/3598836.stm

 

[John Larkin]

Tantalising ... chips running at 650 degrees C:

http://news.bbc.co.uk/1/hi/sci/tech/3598836.stm

We've used some SiC high-voltage schottky diodes, and they work fine.

http://www.infineon.com/cmc_upload/documents/098/950/SDP_D_T04S60_2.1.pdf


Microsemi claims to have some, too, but Microsemi must be some sort of
front organization... they never seem to answer emails or actually
ship parts.

Cree and some other people are working on SiC mesfets, but I haven't
got my paws on any yet. They should behave like high-voltage gaasfets;
the data I've seen shows outrageous pinchoff voltages (like -12),
which makes them behave more like toobs than transistors!

John

 

[Walter Harley]

Cree and some other people are working on SiC mesfets, but I haven't
got my paws on any yet. They should behave like high-voltage gaasfets;
the data I've seen shows outrageous pinchoff voltages (like -12),
which makes them behave more like toobs than transistors!

And at 650C, they should be glowing at least a bit. Quick, somebody find me
an audiophile!

 

[Jonathan Kirwan]

which makes them behave more like toobs than transistors!

.... not back to tweaking grid leak resistors again ...

Jon

 

[Boris Mohar]

We've used some SiC high-voltage schottky diodes, and they work fine.

http://www.infineon.com/cmc_upload/documents/098/950/SDP_D_T04S60_2.1.pdf

Do they have any SRD like properties?

 

[legg]

Microsemi claims to have some, too, but Microsemi must be some sort of
front organization... they never seem to answer emails or actually
ship parts.

There seems to be a problem getting SiC parts, generally, out of US
sources (don't tell me .... homeland security.....? ).

IMC, ABB, KTH and Infineon all produce SiC devices in Europe.

Nippon Steel, Matsushita Electric Industrial, Ion Engineering
Institute and the Mitsubishi Electric Corporation all have active
product development as well.

Siemens/Infinion, with it's ties to Matsushita, may even have the
option of shipping Asian wafers, and vice-versa.

No manufacturer can presently ship parts with rated operating
temperatures anywhere near the SiC process limits, simply due to
conventional packaging constraints.

RL

 

[John Larkin]

Do they have any SRD like properties?

I don't think so. They're schottkies, so shouldn't store charge except
as pure capacitance. The application we used them was to gate a 1200v,
25 amp pulse (a couple in series), and we would have noticed if there
was any charge storage.

John

 

[John Larkin]

No manufacturer can presently ship parts with rated operating
temperatures anywhere near the SiC process limits, simply due to
conventional packaging constraints.

I don't follow that. Researchers have been growing 1400 v SiC diodes
for years, and you can certainly buy plastic-packaged 1400 volt fets
and bipolar tansistors, but you can't buy a 1400 volt SiC schottky. I
thought it was a materials defect/yield problem. As far as temperature
goes, there are ceramic TO-220 packages around, but still no 1KV SiC
parts.

John

 

[Ken Smith]

We've used some SiC high-voltage schottky diodes, and they work fine.

http://www.infineon.com/cmc_upload/documents/098/950/SDP_D_T04S60_2.1.pdf [..]
Microsemi claims to have some, too, but Microsemi must be some sort of
front organization... they never seem to answer emails or actually
ship parts.

I got some Microsemi SiC schottky diodes. They seemed to work up to about
150V. I'm about to get back to that part of the project. It will be a
lot of fun to see if they really meet the 600V breakdown spec.

 

[legg]

I don't follow that. Researchers have been growing 1400 v SiC diodes
for years, and you can certainly buy plastic-packaged 1400 volt fets
and bipolar tansistors, but you can't buy a 1400 volt SiC schottky. I
thought it was a materials defect/yield problem. As far as temperature
goes, there are ceramic TO-220 packages around, but still no 1KV SiC
parts.

The restraints are thermal.

Temperature was the main interest of the original post.

RL

 

[John Larkin]

We've used some SiC high-voltage schottky diodes, and they work fine.

http://www.infineon.com/cmc_upload/documents/098/950/SDP_D_T04S60_2.1.pdf [..]
Microsemi claims to have some, too, but Microsemi must be some sort of
front organization... they never seem to answer emails or actually
ship parts.

I got some Microsemi SiC schottky diodes. They seemed to work up to about
150V. I'm about to get back to that part of the project. It will be a
lot of fun to see if they really meet the 600V breakdown spec.

The Infineon parts meet specs. In fact, they can be pushed to higher
reverse voltage for short pulses. Like most regular schottkies, they
just leak worse and worse as back bias increases, till finally the
dissipation kills it, but there's apparently not a hard breakdown past
the rated prv.

Let us know if you find anything interesting on the Microsemi parts.
They introduced and un-introduced them a couple of times... maybe they
got the problems fixed. I think Cree may make the wafers.

John

 

[colin]

Tantalising ... chips running at 650 degrees C:

http://news.bbc.co.uk/1/hi/sci/tech/3598836.stm

If silicon carbide is one of the hardest abrasives how will they slice the
wafers up ?

Colin =^.^=

 

[Keith Williams]

If silicon carbide is one of the hardest abrasives how will they slice the
wafers up ?

One would assume the standard diamond saw used on silicon wafers.
Silicon carbide is hard, but not as hard as diamond.

The polishing process should also be interesting.

 

[John Larkin]

If silicon carbide is one of the hardest abrasives how will they slice the
wafers up ?

Probably diamond grit wires, same as silicon.

John