The sensor can only switch AC so we have to nix the idea of the DC power supply. My fault. the devil is in the details.
Here goes for part 1. Mount the sensor inside the microwave. You then have to bring out two wires or a cable and satisfy class II wiring. Those two wires are just the contact closure from the sensor.
I'll, sort of refer you to here:
https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&cad=rja&uact=8&ved=0CDMQFjAD&url=https://escventura.com/manuals/sola_introNECclass2_rg.pdf&ei=-S9oVcfxAoG2sQXNjYHgAg&usg=AFQjCNFja1NF30SY4L1OtMRn9A44MzVp4w&sig2=BEGTWneLd_wwXmmH3RVY1g&bvm=bv.93990622,d.b2w
==
The NEC code can be read online for free with registration. You can't copy/paste or search for info excpet by section #'s if I remember right.
==
So, the point is to get around the don't modify the appliance, so connecting it to the input power source is an option.
So, a slight modification would destroy it's UL listing, so that gives you two choices:
1) In line before it even gets to the microwave
2) Current to the fan. the neutral is used because the voltage to ground is low or near zero and therefore breakdown is less likely.
You want to make the change totally reverseable to keep the microwave warranty which is unnderstandable.
Common sense says stranded wires and your basically wiring a doorbell. it might get a little warm and it might have some moist air nearby. Rubber grommets don't like heat, they crack.
You want to stay out of plenum air spaces and use wire insulation that won;t burn. Teflon (PTFE) insulated wire is actually used for fire alarm signaling, so your kinda looking at alarm and security cables, not "hook-up" wire.
See
https://www.generalcable.com/NR/rdo...1_Carol_Alarm_Sec_Brochure.pdfec_Brochure.pdf
Ideally, you'd like a connector mounted to the microwave.
The whole point is to use wire for the intended application and 12 v automotive wire won't work, there's hook-up wire, appliance wire, building wire (THHN, THWN), security wire, fire alarm wire etc.
So, I'm thinking you can get away with two twisted (using a drill) Teflon insulated wires with an underwriters knot with a silicon grommet. For further protection, you could put it in a fiberglass sleave. Hand made has these as well.
terminate in a connector of some sort. I don't know how much room you have.
Teflon insulated wires in a fiberglass sleave, I THINK, would satisfy the code guys. Will they like you modifying the microwave - don't know, but I would have pictures handy.
So, <24 VAC/VDC power limited has been deemed safe to run without rigid protection.
Now, I don't know if your microwave has a plug a plug on it or it;s hardwired, but the point is to get it out safely and easily for service. 16 to 18 AWG is the smallest wire size I would use. The other sensor manufacturer recommends two conductor shielded cable with the shield connected at the receiving end.
It's common to connect shields at one end only.
Does it make sense so far?
I have a hunch that using a sensor like this
https://www.workaci.com/content/amcs-0 one, you could get away with not modifying the microwave at all. It;s fully adjustable from 0.32 A to 150 Amps. With two turns you could lower that sensitivity by a factor of 2. Be sure to secure the wires. I could give you a small length of Kapton tape which is like $20 a roll. two turns is about 19 W and 3 turns is about 13 W.
So with the adustibility and 10% hysteresis, I think you can dispense with modifying the microwave and make everyone happy. the sensor is 2x what you picked out. i suppose that using the microwave as a timer would not turn on the fan. if the exhaust fan of the microwave runs continuously and has a detectable draw, you might be "In like flint".
Let this digest for a while. This is why I like to brainstorm with lots of ideas.
Unfortunately, without doing a test you can't know. The cost is 2x your sensor price and if it doesn't work as expected, you can always use it in plan B.
Your thoughts?