OT: Need a Mosfet

[Bob Reeves]

I have an idea for a project that will require switching around 100 volts at 30 amps to an LED array. I have never worked with mosfets and discovered there are a few different types in what little research I have done. I know some of you guys are designing motor timers using mosfets and wondering if you might point me in the right direction for a simple fast N channel mosfet that can switch a 500 microsecond pulse.

Thanks

[John Rist]

It's been a while sense I worked with Mosfet.   However Mouser is a good source for them.   They have a selection guide  - however it requires some understanding of mosfets to find anything.

http://www.mouser.com/Search/Refine.aspx?N=18338127

PS I usually try and pick a part that has a 2x greater rating then I need.  Not always possible but especially in the voltage categories it is a good practice.

[Tim Wescott]

Only a few?  You need to do more research!  DigiKey lists over 15,000.  Even allowing for duplicates, that's about 5000 different types.

http://www.digikey.com/product-search/en?FV=fff40015%2Cfff8007d&mnonly=0&newproducts=0&ColumnSort=0&page=1&stock=1&pbfree=1&rohs=1&quantity=10&ptm=0&fid=0&pageSize=25

I prefer DigiKey's search engine to Mouser's -- I find that if I have a specific task in mind, I can narrow my choices down quicker with DigiKey's search engine than with Mouser's.  That may just be me, though.

100 volts and 30 amps isn't trivial either.  You're asking for a pretty herky device.  You'll run into issues with intentionally optimistic specifications, making sure your gate drive circuit is good enough (the thing has to turn on and off fast so it won't fry in the transition interval), choosing a part that will work over your desired temperature range, and the thermal design of your board and/or packaging.

At the very least you need to be able to look at data sheets and work out whether any given device is going to do the job for you given (mostly) the channel on resistance, the thermal resistance from junction to case, and the maximum die temperature.  It's not rocket science, but it's not trivial, either.

Contact me at my email address and I'll give you some pointers to places where you can learn more.  Or outline what you're really trying to do, and I'll give you an estimate of what it'd cost for me to design it for you.

[Bob Reeves]

Thanks for the offer, think I have it pretty well figured out.. I also like the Digi-Key search better than Mouser just wished the had the ability to do greater>than or less<than when searching for a particular value.

Came up with this 100V 68A mosfet which should do the job.
http://www.digikey.com/product-detail/en/PSMN013-100PS,127/568-4976-5-ND/2206344

This is the schematic I drew of what I am playing with. Basically it's a high power photo flash using LEDs rather than a flash tube. The plan is to over drive a 100 watt LED array with a short pulse. From what I was able to find it should put out around 40,000 lumens and as long as I keep the duty cycle under 5% it shouldn't be too hard on the LEDs. Using a cap to fire the LEDs to keep the power supply current requirement down to a reasonable value. Didn't want to have to build a 35volt 30 amp power supply.

 

[John Rist]

Looks like it will work.   Add a photo cell and a feed back loop to the reset pin on the 555 timer and you will have an auto strobe.  Ps I also love DigiKey.

[Tim Wescott]

That MOSFET has much more honest specs than the first few that popped up in DigiKey.  It should work for you even when it's at boiling water temperatures, which it shouldn't get to unless you start running the flash at video speeds.

It may work, but I suspect you have some tweaking to do.  Your voltages and currents have changed -- you mention 100V at the start, but I see a 35V supply, and that 1.5 ohm drain resistor implies no more than 23 amps through the LEDs.  Also, you are aware that if you do draw 30A, your capacitor voltage will drop by 15V in your 500us?  You may want to invest in a bigger cap there.  It'd be a good idea to double-check capacitor ESR, too -- you could end up dropping a substantial voltage in the cap with a high ESR.

For all my snarking about details it certainly looks like you have a good starting point.

[Bob Reeves]

Thanks for the encouragement, came up with the above starting point after doing a little math using safe numbers. I ordered five 1000uf 100v caps and several various low value high wattage resistors from Digi-Key plus a 48V 4.2A power supply from MPJA. Should be able to blow up at least one LED array and maybe a mosfet or two. First step is to see how much light I can get out of the LEDs then move on to getting the cycle time I will need. Can't go much further till the parts start arriving.

[Howard Rush]

Bob does such cool stuff.

[Bob Reeves]

Ya, sometimes it's fun stuff and sometimes it's work. Really wished I could find some younger person that knows Windows networking, electronics and enough about programming in C++ to get by. I would almost give my business to him.

Right now I have five projects all going on at the same time.
1.The strobe that started this thread, something I just want to see if I can do...
2. Photo project for a McDonald's in Australia..
3. Microcontroller coin operated gate opener.., I will make about $5.00 an hour on this but it's fun.
4. Rebuilding cables for a coaster in Florida.
5. Transfering video files from a GoPro via wireless network from a Skycoaster to the sales booth.

This isn't counting the Photo Booth I have sitting in my shop I need to sell or put someplace where it will make a buck. Or marketing the software I wrote for it. Also want to build a new stunter over the winter.

I'm working too durn hard but in reality enjoying every minute....

[Brad Smith]

i was a kid when i meet BOB he was allways building and designing cool stuff  there is probably more stuff in his head then he will ever build lol or rebuild. i watched him design and build FM transmitters in pens back in the 70s.

[Tim Wescott]

Ya, sometimes it's fun stuff and sometimes it's work. Really wished I could find some younger person that knows Windows networking, electronics and enough about programming in C++ to get by. I would almost give my business to him.

They're out there.  Lucky for you, there are a lot of good candidates that get rejected in the HR screening process, because most corporate HR departments have turned into classic bureaucracies, where no one ever gets fired for saying "no", but you can get a black eye in a minute from saying "yes" at the wrong time.

Put out an employment ad.  In the requirements, put "must have started a profitable business before receiving your BS degree".

[Bob Reeves]

It works..

Finally managed to receive everything I ordered and put a prototype together. First I tested with a safe 35 volt power supply. This is whiten the LED's continuous ratings so I was pretty sure I wouldn't blow anything up. Set up my flash meter and fired away.. Only had the flash meter about a foot away from the LED and it read F5.6 at 400 ASA. Then I replaced the 35 volt supply with Big Bertha, an 80 volt 10 amp supply I built inside the case of a 1000 watt 220 to 110 step down transformer. Fired it several times and nothing smoked.

The flash meter now read F11 at 400 ASA the same distance as in in the first tests. Gained two F stops which if I'm right is 4 times the amount of light which is exactly what I was expecting. If the specs for the LED are accurate this should be giving me about 40,000 lumens. Next step is to add a lens and reflector to direct all 40,000 lumens in one direction and compare the light output with a studio strobe.

After that I will connect the trigger input to a pulse generator. Actually the parallel port on a computer running software I wrote that simulates the trigger signals from a coaster. I will do a little experimenting and fire it about 10 times as fast as it will fire, wait a minute or two and hit it again. This will simulate a 10 car coaster running about 50MPH (I hope)... I will let it run for 24 hours just to see how well everything holds up.