I'm doing my first WiFi project. It's a mailbox notifier using the ESP8266 Wemos D1 Mini. The power circuit (see pic) uses a magnetic reed switch to ground the gate of a P-channel mosfet when the door is opened, then the processor brings GPIO pin D2 high to turn on an NPN transistor, which continues to ground the mosfet gate after the door is closed. Then when the notification has been sent, the Mini switches its power off.
During transmit there are pretty large current spikes - well over 100mA. I know the regulator in the D1 Mini will have trouble supplying those spikes, so per a video by Andreas Spiess on powering the ESP8266, I've added a 1000µF cap to the 3.3V pin.
But in testing out the circuit, I connected my battery powered scope across the switching mosfet, and found that the voltage across it reaches as high as 0.4V during the spikes. The LM0701 I'm using has a couple ohms RDSon, which I think is clearly too much for this situation. So I'm looking for a replacement with much better RDS values. I've been unable to find anything in a TO-92, but in surface mount I've found the IRLML6401. This is the 12V version of the 20V IRLML6402 that seems to be popular, with a bit better gate threshhold and RDS numbers than the 6402. Power is coming from either an 18650 or three AAs, so 12V should be plenty.
Would this be a good choice? Does a better one come to mind? The only thing that's unusual about this project is that it will be subject to temperature extremes. The circuit will turn on for only about 20 seconds twice a day, and the power will be off the rest of the time. But, you know, it could be pretty hot, or cold, in the mailbox.
The other issue is how to deal with the brick-enclosed mailbox as far as WiFi is concerned. But I probably should ask about that in another thread.