question about mosfet ratings

What's the problem? :)

The leads are a bit overstresssed by that much current, as compared with the ratings for wires., but I assume the silicon is going to be dissipating considerably more heat than the wires, and it's mounted on a heat-sink. Plus the leads are good up to the melting point of solder, so their exact temperature isnt of that much concern.

If I were using them though, I'd keep the leads as short as possible and transition to more capable thickness of copper wire asap.

No. Look at the IRF1404 device. It says it is rated at 202 amps. However, note 6 says that the TO220 package itself is only rated for 75 amps. Fairchild doesn't play this game. For instance, their ISL9N302AS3ST has even lower Rds but they rate the device at only 75 amps.

So, it appears that about 75 amps is the TO220 package limitation.

Cheers, John

John - KD5YI a écrit :

Nah. Just add the obviously missing invisible small print footnotes stating that's with an infinite HS immersed in LN.

--
Thanks,
Fred.

News==----

Newsgroups

IRF2804??

I note that the nominal lead cross-section for that part is 0.050" wide by 0.020" thick, or 0.001 square inches. I compare that with the "fusing current" (the current that can melt a wire in a nominal room temperature environment) for copper wires: 16AWG at 0.051" diameter,

0.002 square inches, is rated at 117 amps. This tells me you'd better get the heat out of those leads by conduction to something cool nearby if you really want to run 280 amps through them! They CANNOT be very long, or you won't get the heat out of them without immersing them in something that will take away the heat (e.g. fast-moving air or some coolant fluid).

Another way to look at it: the 0.001 square inch cross section will give you about 0.67 milliohms per inch at room temperature (and higher at higer temp: over 0.8 milliohms per inch at 70C). At 280 amps, that's 52 watts per inch of lead, at room temp; over 60 watts per inch at 70C.

Maybe you can operate it in a liquid CO2 bath. ;-)

Cheers, Tom

A bit overstressed? The fusing current of 11 AWG wire is 280 amps.

From

Just from looking at the TO220 I think we are under 11 AWG. I compared it to some 14 AWG magnet wire and that looks about right

Or looking at the specs for TO220 It should have a cross section of about .000675 square inches and a #11 wire should have a cross sectional area of .00650 square inches.

So I'm having some trouble buying 280 amps of actual sustained current through the part.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Thanks John; that makes sense.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Can a heatsinked TO3 package radiate away more heat?

Michael

It helps if we differentiate between the current-carrying capacity of the silicon and the leads, and between continuous versus pulsed operation.

The 280 amps is likely to be the current-carrying capacity of the silicon for say a few milliseconds. That tells us the silicon isnt going to blow up under that amount of current.

But in the long-run, there's going to be some limit as to how much power the package and leads can handle. That's probably a lot smaller than 280 amps. 75 amps sounds like a reasonable number.

But we can probably still push 280 amps thru there,but not for long, and not for more than about 23% of the time.

ITYM 7% of the time.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com

Hello Fred,

Like that super-duper audio power amp I saw at Walmart: 120 watts PMPO or whatever, fed by a little plug-in transformer the size of a small fist ...

--
Regards, Joerg

http://www.analogconsultants.com

No, it's the die rating

t News=3D=3D----

#1 Newsgroup Service in the Wor= ld! 120,000+ Newsgroups

=3D----

That's how the keep the price down, by eliminating the truth.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.

The 280 is sustained current not pulsed. Pulsed is 1080 amps. Part is an IRF2804

John's got the reasonable answer.

I might still use it since the last time I had a problem a 15 amp fuse didn't protect a 30 amp mosfet.

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Those are Chinese watts = 1/1000 watt. Everything Chinese is smaller.

It has a larger contact area so it should transfer more heat to a heatsink efficiently. But the last one of those I used had steel leads.

They claim this part, IRF2804, will work up to 175 degrees Celsius; are there TO3 parts that can get above 125?

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Maybe you mean liquid N2 - CO2 can't exist as a liquid at atmospheric pressure. But, you could put some dry ice in some kind of freon - in grade school science, we would put dry ice in acetone to freeze stuff. :-) FWIW, the grocery store where I shop has a deep freeze with dry ice for sale.

You could look at some of those "overclocking" web sites to get some ideas for some really aggressive cooling.

Cheers! Rich

These days, I wouldn't be surprised to see 120 watts from a fist-sized switcher. I've seen them about twice the size of a matchbox that will do

5 or 10 watts.

Cheers! Rich

OK, ignoramus time here - 75 is 23% of 280, so why do you say 7%? Is it because of thermal inertia effects or so?

Thanks, Rich

Maybe I'm just a chicken, but I can't see putting that much current through a single device in the first place, unless it's, like, a hockey puck thyristor or something. ?:-/

Thanks, Rich