As long as you haven't bypassed the 540 with any other ties to ground, I think this should work.

The clock isn't grounded at the camera is it?

 

So I assume the clock is connected to the MCU via I2C, and determines when to turn on the camera power?

Does it also determine when to take the shot? how is that done? How is it interfaced to the system?

Forgive me I'm just trying to get a complete picture.

 

I think I found my answer:
"When the PIC output is LOW, the transistor is OFF and the gate of the P MOSFET is HIGH (12V). This means the P MOSFET is OFF."

Yup, you got it.

 

The clock is grounded at the MCU. Yes, the clock is connected to the MCU via I2C. The clock's alarm time is set for the next photo, and this wakes up the MCU, which then turns on the 12V power. The MCU signals an optoisolator, which in turn closes the camera's remote shutter circuit.

I don't think I've "bypassed the MOSFET with any other ties to ground", but I'm not exactly sure what that means. I think all of the grounds have been accounted for in our discussion.

 

I'm comparing the circuit in the first post to the schematic in post 4 and they do not match. The schematic is showing high side switching with the wrong type transistor.

It also shows a common ground connection at the converter output.

Are you actually using that schematic?

 

Yes, I am sorry I posted that schematic in Post 4. It turns out that the PCB design did not exactly follow my breadboard design. The breadboard design (in my first post) is what I want to fix--and what the recent discussion has been following. I hope I didn't lead people too far astray with that Post 4 schematic; I didn't realize the difference until the discussion had moved on, and I didn't want to complicate things.... Now that I realize I can edit that; I've put in a note.

 

I went through the breadboard wires and checked everything, and replaced the converter with an led. It stayed on all the time just like the converter had. I tried switching the source and drain wires; same result. Then I switched everything back again, and then led started working as it should. Then I put back the converter and that worked too, and it switches the camera power correctly. A possibility is that I had a wire loose somewhere on the breadboard and fixed that when I fiddled around. I am going to do some more experimenting to see how robust the setup is--maybe the design just doesn't work consistently. If not, I now have another method to try--with the P-channel. Thank you all for your help.