I'm trying to make some LED lighting systems for my pantry and linen closet; I've got some nice cheap strings of 12V LED modules from MPJA and a 12V/3A power brick.

To make the lights turn on when the door opens and off when it closes, I figured a magnetic switch would be good because they're already designed to be used on doors. I had to shop a bit to find a NO/NC type, but found one that will work; however, the LED strings are rated at 1.6A and the reed switch is only rated for 0.75A.

My first idea is to use a MOSFET to switch the LEDs... is this a good plan? If so, I'm figuring the NC contact (circuit opens when magnet is brought near) of the reed switch can drive the gate of the MOSFET to switch the LEDs. I wanted to use a MOSFET instead of a relay to avoid the clicking sound, and also to minimize moving parts.

Is this schematic sufficient?

 

Add a 10K resistor from gate to source- (ground)

The circuit as drawn has no path to remove charge from the gate.
The 10K provides a path for the gate charge to rapidly drain away when the switch opens.

 

Since it's a SPDT reed switch, could I eliminate the pulldown by reversing the switch and have it switch the gate between +12 and GND so it actively pulls the gate to GND when the magnet is near?

The point of using door-switched LED lights is energy efficiency (LED > incandescent/fluorescent; avoids accidentally leaving them on); are there any ways I could make this more energy efficient? Would a higher resistor on the switch help minimize current draw at the gate? If a pulldown is the only way to go, could I get away with something higher, like a 47k or even 100k to minimize leakage? If reversing the switch would eliminate the need for a pulldown, is there any leakage there to worry about? I know we're probably talking microamperes here or even fractions thereof, but I'm curious (also, I'm considering adding a battery backup eventually, and any sort of "always-on" leakage could be a drain on that during an extended outage such as a hurricane).

 

Ok, Use a 1Meg resistor, then it's consuming micro-amps - only when the thing is on.

 

The circuit in post #3 will work as drawn. I would increase the resistor to 1K.

A MOSFET gate structure is a small glass capacitor. A power MOSFET gate is tens or hundreds of small glass capacitors in parallel. For normal parts (as opposed to high voltage or very high current parts), the static current is a fraction of a uA.

ak