MOSFETs and Drivers

http://www.google.co.uk/search?q=+IXFH12N100F+-sIXDD509PI+and+IXDE509PI+MOSFET&sourceid=ie7&rls=com.microsoft:en-gb:IE-SearchBox&ie=&oe=&rlz=1I7GZEU_en&redir_esc=&ei=jeaOTOzDH5bKjAfQ_NzzDA

http://www.google.co.uk/search?q=+IXFH12N100F++and+IXDE509PI+&sourceid=ie7&rls=com.microsoft:en-gb:IE-SearchBox&ie=&oe=&rlz=1I7GZEU_en&redir_esc=&ei=1OaOTNO6LYWUjAeumqGGDQ

You cannot find one specific model number of a MOSFET for a class of students. Well now, I would expect a quality school would teach students suitable techniques to deal with obsoleted parts. Are you sure that you cannot find these parts from anywhere other than this one vendor.

Problem is that for fast MOSFET manufacturers are only few. IXYS and Infineon. For 1kV range only IXYS is manufacturing. These models were selected for lower price range as students can not afford high price.

I have already had classes with Vishay MOSFETs but they are not so fast.

Application is high frequency Ultrasound with several MHz switching rates at 500W to 1kW high power to penetrate atmosphere and to sense Doppler velocities of Molecular layers.

I have little choice.

Well I still think you have a very useful lesson here for your students. I wish you and your students the best in dealing with this problem.

You can try with Pankaj Electronics, New Delhi. We buy from them high voltage, high frequency MOS FETs of IXYS. The company is prompt. Web address is: www.pankajelectronics.com

Good idea. Thanks.

Hello Shyam,

Another source is DigiKey here in the U.S., I don't know what kind of costs you'll incur for foreign shipment but their quantity cost isn't too bad.

http://parts.digikey.com/1/parts/1714030-mosfet-n-ch-1000v-12a-247ad-ixfh12n100f.html

The driver IC requires a request for quote from them:

http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=IXDD509PI

http://search.digikey.com/scripts/DkSearch/dksus.dll?vendor=0&keywords=XDE509PI+

In your case it sounds like a suitable part may be one that distributors stock, as well as one suited for your application. I'm looking at the substantial collection of IXYS mosfet drivers Mouser stocks, and am thinking there has to me a matchup to your needs there someplace. For example, they might have an inverting (I) rather than non-inverting (D) version. And as is so often the case these days, you may have to give up on miniDIP packages.

The ixdd509 driver is an impressive 9A driver, but I'm not sure it's the best choice for high-frequency RF mosfet driving. I tell myself the high-current drivers have lots of internal capacitance, and may not be as well suited as a lower-current part. I like to look for a supply-current vs frequency plot. For the '509 this stops at 2MHz. Other parts, like the ixdd415, have plots going to 25MHz. Now that's more like it!

Yes, the supply current at 10MHz is almost 1A with no capacitive load (and 1.3A with 3nF), but hey, that's why it comes in a low-inductance power-pad package!

Hah, the '415 is a dual 15A driver - so much for my lower-current argument! BTW, Mouser has 35 of these in stock for $17.

Many of the IXYS mosfet drivers are very fast, with 15ns drive speeds, yet their operating-frequency plots stop at 2MHz. How well they are suited for (continuous) higher frequencies depends in part on their power-dissipation capability. For example, the ixd_409 types are available from Mouser in 5-pin TO-220 (CI suffix) and TO-263 (YI) power packages, with θ_JC = 2.5 ºK/W. Extrapolating from 2MHz, a '409 with a 2700pF load has 500mA supply current at 10MHz, for 12V supply. That's workable in a power package, even if it's not shown on the datasheet.

If you are making short RF power pulses, then it's more a matter of the thermal mass of the die and package. In that case you may be able to use one of the more commonly-available SMT parts. Also, the TO-263 package may not be so easy to bolt to a heat sink, but it has more thermal mass than an simple soic.

I posted an answer with respect to mosfet gate drivers for your RF MOSFETs, but looking at the IXFH12N100F datasheet and its assertion for routine use at 13.56MHz, I'm struck by the fact that these must most often be driven by a transformer circuit, in a resonant fashion, taking into careful account the transformer leakage inductance, etc. I doubt the 13.56MHz product engineers are using cmos gate driver ICs.

My applications are pulse with rise time and fall time within set limits and load is capacitive. I am using push-pull scheme to charge and discharge the load capacitor.

pulse width and repeat rates are programmable and that makes a bit difficult to compute average power.

Peak current of 12A for <50ns and much lower current thereafter during pulse width formation is required.

Diver peak current of 2A, 4A, 9A and 14A are tried for large gate capacitance for the MOSFET they drive.