Low voltage, high amps, can it scale > 100kW?

[Arlo1]

As well I've put over 200amps through a TO247 leg for over 10 seconds. The continus rating of 120-160 amps DC with proper cooling is nmo lie!

Stop, put down the Koolaid they have you drinking. You will not be putting 120A continuous (not 10 seconds) through the legs of a TO-247 package continuously without a very extreme cooling method..

I have a video of a real world test at 130 amps but I just knoticed I don't have a video of the 200 amp test.
THIS IS REAL WORLD! I did these tests to find what worked good for heat and electrical conduction from the back of the fets.
200 amps though the legs is doable for over 10 seconds this is no bull shit.
Luke can back me up on this.
The data sheet is not Koolaid if and ONLY if you stay with in all the parameters of the power switch. The tricky part is to stay with in all of them at once!
The desat circuit and temp monitoring as wall as great cooling at all steps to doing so!

 

[zombiess]

Arlo1, are you posting from Jonestown somewhere in South America? They have Koolaid too. If you don't drink it your MOSFETs will if you ask nicely enough :lol:

Seriously, 10 seconds does not in any shape or form equal continuous. If you do not know already, the manufacturers, at least IRF in this case, get the ultimate current rating by measuring those temperatures in an inert nucleated boiling fluid right? I'm familiar with the process in cooling ICE motors as I know someone doing it and it's quite the impressive cooling method.

Either way I would still like you to perform the test ramping up current in 10C increments until failure. It will make a cool video, especially with FLIR. Just make sure the MOSFET is mounted to something good like a CPU heat sink. It would be really neat to do it twice, once with a good electrical insulator and once without (but use compound).

This may seem a bit off topic, but I don't think it is at all. This is the kind of info I wonder about, how far can we push devices safely. How high can we get the power density, what kind of price point can we do it for. I say we meaning this site, the people on here. We have a very unique collection of minds and skill sets. Most want to see a change in the way things are done, so let's do it ourselves.

 

[Arlo1]

Nope 10 seconds does not. But when it is on for 2/3 of the run time in a three phase controller and I mentioned 200 amps for a leg is possible that means their number of 160a is not that unrealistic.
I have done some testing to the limits with both just testing the legs them selfs and the fet flowing drain to source like it would be and I started a thread on it.
[youtube]GlzOFCTPk0A[/youtube] Note I found out after the aluminum will not show the heat with the flir until I paint it with something.

Call it what you will be the numbers in the data sheet are real you just need to keep inside ALL the limits and yes if that means fancy cooling then so be it!

http://endless-sphere.com/forums/viewtopic.php?f=30&t=64527&hilit=+destructive

 

[flathill]

There some custom MOSFETs which output junction temperature. The reason most dont offer this option is because customers demand a standard 3 leg device.

 

[zombiess]

There some custom MOSFETs which output junction temperature. The reason most dont offer this option is because customers demand a standard 3 leg device.

Do you have any links to data sheets or part numbers?

 

[zombiess]

Arlo1, nice video. It is good to see how much abuse the device can take for a very short period. That increases my confidence in doing very short bursts.

A test I would like to see is loading a device similar to what you have done, but using pwm on the gate and a control loop that monitors the MOSFET case temp and varies the pwm to prevent it from going over 60c, 80c and 100c. A log of the current after running the test for ~60mins should provide the continuous current capability. The MOSFET leg temperature should also be logged to get a delta between case and leg temp.