Understanding Mosfet specifications

[tawandapro]

Good afternoon everyone

I am having some issues understanding a few things about mosfets when it comes to current delivery. My motor controller states that it can handle a maximum of 50A Peak and continuous current of about 25-30A with a heatsink. The data sheet for the Mosfets used state that the maximum continuous current at 25 degrees  is 160A and 113 at 100 degrees. Why is there a large margin in current capability 30A vs 160A? especially if the board has 8 of these, 4 to control each channel. Is the 30A max and 50A peak rated just for what the board can handle?  here is the controller:
http://www.elechouse.com/elechouse/index.php?main_page=product_info&cPath=100_146&products_id=2179

here is the data sheet:
http://www.irf.com/product-info/datasheets/data/irlr7843.pdf
Thanks any links to recommended tutorials/books will be greatly appreciated..

[fungus]

It's not just the MOSFET.

What about the tracks on the PCB? The screw terminals? How much current can they handle?

[GoForSmoke]

I don't see heatsinks on the board or temperature in the spec.
Should you run more than 20-25A through it anyway? It should last longer the less you push it.

[Grumpy_Mike]

Why is there a large margin in current capability 30A vs 160A?

Because the board is real and the FET data is idileised.
While not actually lying the data sheet gives big headline figures that are unrealistic. For example the simple fact of saying at 25 degrees takes you into the realms of fantasy. That is not the temprature of the room it is the temprature of the junction inside the package. To acheave that in practice you have to use liquid nitrogen to cool the device so that the junction temprature with that ammount if current is 25 degrees.
So factors like heat dissipation limit the real maximum current.

[dc42]

For example the simple fact of saying at 25 degrees takes you into the realms of fantasy. That is not the temprature of the room it is the temprature of the junction inside the package. To acheave that in practice you have to use liquid nitrogen to cool the device so that the junction temprature with that ammount if current is 25 degrees.

I think you will find that rating is quoted at 25C case temperature, not 25C junction temperature. But I agree with your sentiment. 25C case temperature is in practice unachievable when the device is dissipating a lot of power, even with a huge heatsink - except possibly outdoors in Siberia.

[oneoutofmany]

In the data sheet (page 1) it has a small 4 next to the continuous current statement. The four indicates a note somewhere else concerning that figure. If you look at the bottom of page 1 it says that notes 1-5 are on page 11. If you go to page 11 note 4 states:

"Calculated continuous current based on maximum allowable junction temperature. Package limitation current is 30A."

[retrolefty]

It all falls under that old adage that the marketing department creates page one of the datasheet and the engineers write all the remaining pages to cover their rear ends. 

[oneoutofmany]

It all falls under that old adage that the marketing department creates page one of the datasheet and the engineers write all the remaining pages to cover their rear ends. 

Bingo!

[tawandapro]

Because the board is real and the FET data is idileised.
While not actually lying the data sheet gives big headline figures that are unrealistic. For example the simple fact of saying at 25 degrees takes you into the realms of fantasy. That is not the temprature of the room it is the temprature of the junction inside the package. To acheave that in practice you have to use liquid nitrogen to cool the device so that the junction temprature with that ammount if current is 25 degrees.
So factors like heat dissipation limit the real maximum current.

i re-read the datasheet and at the end it says that the package itself is a limitation to how much current it can  handle. the to-252 package in that series is limited to a maximum of 30A

[MarkT]

And the package limitation is probably mostly the gold bond-wires between the
package's copper pins and the chip pads...  Make nice little fuses in fact.