Good Day All.

Upto now all my projects have used low-side switching.
But now my current project needs me to be high-side switching.

I want to use the 0-5v logic level ouput from a pic chip to switch the 12v high side of my device on and off.

Can you please advice how to go about this.

Thanks for your time.

Best Regards.

 

Please explain more about your circuit application; ie: load current, type of load (resistive, capacitive or inductive like a motor, relay or solenoid).

Are you wanting to use an N-channel power MOSFET on the high side, or can you use a P-channel MOSFET? It is easier to use a P-channel.

 

SgtWookie: Thank you for your quick reply.

I don't mind using N-channel or P-channel.

As for the load I have two projects based about this one, with one pulling very little current at around 500ma, and the other one pulling close to 10amps.

I would like to test on the smaller load to start with.

Thanks for your time.

Best Regards.

 

OK, are you switching these loads on/off slowly, or are you planning on using PWM?

 

Here's a simple driver for low frequency on/off switching:

R1 keeps the MOSFET turned off by default.
R2 limits the maximum current flow through Q2, and helps to snub "ringing" on the gate due to the L of the wiring and the C of the gate.
R3 limits the base current to Q2.
When the I/O pin is high, Q2 will sink current from the gate of Q1, turning it on.
When the I/O pin is low, Q2 turns off, and the gate of Q1 charges back up until it is the same voltage as the source terminal, turning it off.

 

SgtWookie: I would be indeed using pwm.

Thanks for the simple driver design, that will get me started.

Thanks again.

 

Well, that simple driver isn't well-suited for PWM, due to the 1k pull-up resistor discharging the gate. The turn-on time will be OK, but the turn-off time will be very slow.

See the attached. It's more complex, but will give better performance.

If your Vcc is 12v, then you can eliminate Zener D2, and replace R2 with a piece of wire.

Note that you should have bypass capacitors between V+ and ground. Breadboarding probably won't work well with this circuit.

 

I may have missed the point but if you drive the output directly from the pic with a small pull up to a logic level mosfet that should work. I've done it lots with 8051 controllers. The common circuit I've seen has a resistor between the processor and the mosfet, but I can't see why - the processor output is current limited.

 

Well, that simple driver isn't well-suited for PWM, due to the 1k pull-up resistor discharging the gate. The turn-on time will be OK, but the turn-off time will be very slow.

See the attached. It's more complex, but will give better performance.

If your Vcc is 12v, then you can eliminate Zener D2, and replace R2 with a piece of wire.

Note that you should have bypass capacitors between V+ and ground. Breadboarding probably won't work well with this circuit.

I've simulated it with LTSpice and looks very good! may I ask you the principle on how it works?

I'm going to learn at school *MOS and how they works, but I cannot wait to actually try to start playing with some of them the right way! Also, I would see the principle and the limits (if any) of discrete components drive, rather then just using plain IC driver for *MOS

Tnx!

 

I am working on a project that sounds nearly identical. My 12V/10A load happens to be a coil so some form of protection will be needed correct? Also, I'm having a hard time telling if the above diagram will switch the 500mA of the 10A circuit that Mad Professor was inquiring about? Any help would be greatly appreciated!