Video Suggestion on power mosfet's

[seulater]

Something I never could understand concerning surface mount power mosfet's is their insane current ratings.
Take the MCACLS290N04YHE3-TP    N-Channel 40 V 290A for example. It claims a Continuous Drain of 290A. 
I just can't believe that at all.  They want me to believe that a few SM pins on the drain and source pins are going to take that ?

Hope Dave could do a video explaining the data sheet with these insane numbers.

[ataradov]

It is not about believing. This package has large exposed copper pads on both sides. The die is sandwiched between those copper plates.  The pins will have no issues carrying 300 A. The current capacity is only limited by the Ohmic loses, which are minimal for the copper like this.

Your biggest issue is to keep Tc and Tj within reason from the Rds loses in the transistor itself. This is where you need to engineer a really good cooling solution. And with sub-optimal cooling you will not get anywhere close to 290 A.

[boB]

And realize that the 290A figure is with the tab case temperature held at 25 degrees C.

That's so that with its specified thermal  junction to case resistance, the die will be no hotter than 175C, the max die temperature rated.

That goes for the maximum device dissipation too. 

We once tested a collector lead of a TO-220 and it held until about 150 amps.  Just in room air.

boB

[RoGeorge]

It claims a Continuous Drain of 290A. 
I just can't believe that at all.

Yeah, that's a lie.  In some datasheet (seen in some MOSFET datasheets from IoR, in the fine print notes under the tables is specified that that's only a calculated current, based on the thermal resistance of the transistor's die and the max allowed junction temperature, while the packaging itself can only withstand 75A or so.

[RoGeorge]

Found the example for such a datasheet, see:
https://www.eevblog.com/forum/projects/transistors-die-pictures/msg5454833/#msg5454833

[ataradov]

Yeah, that's a lie.

It may be a lie. In this case I have no reason to not trust this number. The package is well designed to handle heat, it just needs really good heat sinks, probably mounting on an aluminum  substrate PCB and a huge heat sink above. And may be active cooling.

[RoGeorge]

My guess is, even if the package were to be soldered on top of a copper anvil, the bonding wires will still vaporize like a fusible at 290A DC, long before anything else gets hot.  Would be an interesting test to measure of how much continuous-regime DC-current the transistor can withstand in practice, but I didn't try.  I'll bet for maybe 10-30A, certainly not 300A.

For example, in this AN-1140 from IRF, a different (bigger) packaging is used, yet they only talk about something like 45A instead of 290A, and even those 45A come with a lot of unrealistic conditions to be met in practice:  https://www.infineon.com/dgdl/Infineon-ApplicationNote_MOSFET_Continuous_Current_Rating-AN-v01_00-EN.pdf?fileId=5546d462689a790c0169166671e9454c

[ataradov]

There are no bonding wires in this package. The die is sandwiched directly between the copper contacts, which are exposed on both sides.