I was considering wiring things as in the attached image. The three position switch would switch high volatge 120V AC, and also switch a 5V DC signal.

I say "was considering" this as my project has expanded a bit now and I'll use several buttons hooked to the MCU to change modes now, but originally there were only two modes and I thought this would allow a super simple interface and allow the MCU to know which of the two modes the user selected.

I'm curious and want to learn if this would have been a safe or acceptable way to wire things? Any thoughts or comments appreciated.

 

As long as the switch has the insulation voltage rating of the applied AC then it should be no problem.
It might be advisable to reference the 5vdc common to earth ground as a precaution.
Max.

 

I won't do because of possible EMI/RFI coupling issues with the high-energy 120VAC circuit to the low-energy digital circuit even if the insulation was proper due to the unshielded common-body nature of the switch.

 

I would have no comment yay or nay without seeing the spec for the switch you are considering using.

 

I would have no comment yay or nay without seeing the spec for the switch you are considering using.

Switch is listed as 10A 125V AC

 

Usually the dielectric rating for that type of switch is in excess of 1500vac and they also have a barrier between each side.
Max.

 

I'm sure it could be made safe electrically (but I would worry about the close proximity during circuit testing) but electromagnetically you have a fairly high impedance (10k ohm with the switch open) input of the MCU that could be capacitive coupled to EMI/RFI carried on the power wires in very close proximity due to the switch and external wiring from both circuits.

 

I'm sure it could be made safe electrically (but I would worry about the close proximity during circuit testing) but electromagnetically you have a fairly high impedance (10k ohm with the switch open) input of the MCU that could be capacitive coupled to EMI/RFI carried on the power wires in very close proximity due to the switch and external wiring from both circuits.

Do you mean the EMI/RFI would possibly cause the MCU to read the wrong value on the input? (Or worse, cause damage to the data pin?)

Since the MCU would be power cycled each time the switch is changed from one position to the other, it would take a second to startup, and if the reading of the input was delayed say 500ms after startup would that always allow for a reliable reading?

 

...
It might be advisable to reference the 5vdc common to earth ground as a precaution.
Max.

A ground connection like this Max?

 

Do you mean the EMI/RFI would possibly cause the MCU to read the wrong value on the input? (Or worse, cause damage to the data pin?)

Since the MCU would be power cycled each time the switch is changed from one position to the other, it would take a second to startup, and if the reading of the input was delayed say 500ms after startup would that always allow for a reliable reading?

It could cause the wrong value, lockup the MCU or damage it in an extreme case. If the MCU would power-fail reboot from a fast switch cycle depends on how long the MCU power bus remains good during switch changes if there is no direct AC power fail input signal or trigger.

 

A ground connection like this Max?

Yes.
Max.

 

I may be wrong but the switch is wired so the power supply stays on. Why not just skip the switch and power the supply from line and switch only the low voltage? Unless I misread the schematic.

 

I may be wrong but the switch is wired so the power supply stays on. Why not just skip the switch and power the supply from line and switch only the low voltage? Unless I misread the schematic.

It's a On - Off - On three position switch.

 

My bad, I didn't think of that until well after I posted.

 

Thank you @MaxHeadRoom and @nsaspook for your feedback and ideas. I appreciate it.