As the title says - I can see a-lot of people doing this on forums etc.

This being: the omission of galvanic separation.

I need to PWM a 24V DC fan fan from my micro-controller .

I have already sorted the power supply for the fan, but i was digging through my box and found a suitable MOSFET that will be more than open enough for my use at 4V.

Though, if I hook the PWM output up to the gate terminal with a suitable series resistance and resistor to source - then i will need to connect the source of the MOSFET controlling the fan to VEE of my micro-controller board.
This means that one of the terminals of my power supply will be connected directly to my micro-controller, which has me questioning weather or not i will need an opto-coupler.

The schematic below shows my circuit configuration. (two zener diodes are 12V to give 24V)


If i were to use an opto-isolator, then I would need to generate another DC level at 5V for gate driving of the logic level FET. Which is of course MORE work, though i would prefer that to a blown Micro-controller.
A voltage divider across the 24V DC would suffice for providing 5V at <1mA that the FET needs, this has just occurred to me.

Any insight on the matter would be more than helpful.
Thanks!

 

It seems to me you have sufficient isolation, just the the DC fan supply should be shown as a separate supply (24v)?
Although you could use two supplies with commons connected?
Max.

 

@urb-nurd
You are way overthinking the problem. It is no problem to add the fan to the same common (0V) as the micro. Just add a reverse-biased diode across the fan motor terminals.
Add a nice 7805 for the 5V supply to the micro and PWM to your heart's content.

 

The omission of galvanic separation is the rule, not the exception.

 

It seems to me you have sufficient isolation, just the the DC fan supply should be shown as a separate supply (24v)?
Although you could use two supplies with commons connected?
Max.

Why would i use more than one supply if i can derive all the power levels required from one? (genuine question)

@urb-nurd
You are way overthinking the problem. It is no problem to add the fan to the same common (0V) as the micro. Just add a reverse-biased diode across the fan motor terminals.
Add a nice 7805 for the 5V supply to the micro and PWM to your heart's content.

A diode is already present on the schematic, unless you are talking about adding a second one.
The Microcontroller is actually a ready for PIC board from MIKROELEKTRONIKA, so it already has a supply regulator.
I am currently powering the MCU board via USB as i am sending variables and measurements over serial for graphing my PID performance.
Thanks for the info!

The omission of galvanic separation is the rule, not the exception.

I see - wise words as always.
Thanks all!

 

Why would i use more than one supply if i can derive all the power levels required from one? (genuine question)
Thanks all!

Fine, but whether you use opto isolators or not, the commons will still be the same.
This is often done with AC & DC loads and much higher voltages.
Max.