Trouble finding a high powered mosfet

Those gadgets are designed to run from 85-265ac, not <= 80v dc from a solar panel of unknown impedance…



2.5^mm solid copper as used in the UK for house wiring is reckoned ok to 22A

2.5 mm diameter wire at 50 ft would have a 10% voltage loss when trying to transmit 20 amps. To keep high efficiency a 5.19 mm diameter wire at 50 ft would have a voltage loss of 3%.

MarkT I looked at that chip driver you suggested looks like its rated for 80 volts. It would bring down the cost of the build. And increase efficiency. Tried looking for other gate drivers and google doesnt want to cooperate. So i may stick with that one.

Only other question would be the inductor. Found a bunch of shielded inductors but i have not a clue yet as to what values to start at. Shielded would prevent interference on the board and components but would still have interference generated by large moving current.

But i will take your advice on using an oscilloscope.

The impedance of the solar panel will be much higher than the wiring .... but go ahead - have fun.. and look for loads of magic smoke!



If it was me, I would wire the panels for 48volt.
Still resonably safe to work with when "live".
I would probably use an off the shelf MPPT charger (ebay), going to a 24volt deep-cycle bank.
More panels could be cheaper and easier than a tracking system.
Calculate the battery requirements for heat lamps.
You might be in for an unpleasant surprise.

OP if you've no idea what the inductor should be you're in for a LOT of magic smoke!

Look at eg LT's website for loads of excellent guidance on designing switchmode psu's



ps - and yes - I have designed big switchmode psu's professionally. - it isn't trivial.

Most off shelf mppt charge controllers ive seen they want upwards of $200.00 to be able to manage a max of 4 panels. That means 4 charge controllers at $800.00 to be able to control 16 panels. Im sure if i only blow up 20 fets and manage to get it to work correctly while only loosing 20 fets that i still come ahead cheaper.

Ive looked at a few of the cheaper ones and they do not appear to be actual buck converters but pwm charge controllers. Id prefer the higher efficiency of a buck converter.

MarkT whats your thought of the ltc4444 for my application. The fet driver you mentioned is a full bridge. The ltc is a half bridge it says its a syncronous control but does not provide a wave form output graph to verify.

The professional devices are expensive because it isn't easy.

If (you can get away with a less efficient but cheap reliable and simple solution see my post #5.
I'll modify the values for panels in parallel if you like - that would be my preferred solution.

else (pay the money || sniff the magic smoke!);



How big are your heat lamps. 250watt?
Are you planning to use batteries to bridge the night.
One lamp will drain a $250 deep-cycle battery in a few hours (can't drain them more than 50-60%).
Calculate about $1000 worth of batteries per heat lamp, just to get you through the night.
And they need replacing after three years.
Batteries alone might already be more expensive than the power bill you have now.
Do the maths.

This spring the wife was rocking 7 (250watt) heat lamps. After opening the power bill she said shell never do that again. Ive repeatedly told her a reptile ceramic heating element produces just as much heat but far less consumption and she wont use them. She has agreed to rock no more than the required 2 heat lamps. 1 has to stay in the pump house the other for poultry raising. Second heat lamp only really needs to run for about 1 month.