# high side/low side mosfet

#### bluebrakes

Joined Oct 17, 2009
245
I've generally got a good idea how mosfets work.

But i've got a couple of questions you've answered hundreds of times before and are getting sick of it.

What is high side and low side?

Unlike transistors that use current, mosfets are driven by voltage, so why do on a few schematics that i come accross, do they have the gate voltage side listed as 5v and the switch side as a higher voltage, such as 12v?

What do people use to drive the gate, if the only power supply is say 12v and the gate needs a minimum of 2v to close?

#### MikeML

Joined Oct 2, 2009
5,444
The issue with a FET is not what is the minimum voltage that just barely begins turning it on, but how big the gate-source voltage needs to be to fully saturate the FET. Most FETs, even though they have threshold voltage of ~3V, really need a minimum of ~10V to get the advertised drain-source ON resistance, and low self-heating.

High side/Low side usually refers to where the FET is relative to the Load. If one end of the load is grounded, the switch is a HIGH-SIDE switch. If one end of the load is tied to Vcc, then the switch in series with it is a LOW_SIDE switch.

A FET can be configured as a switch, or as a source-follower, regardless of it being high-side or low-side. The gate drive direction and level is a function of the configuration. For example, if a high-side switch is configured as a source-follower, the gate should be driven about 10V more positive than where the drain is connected.

#### bluebrakes

Joined Oct 17, 2009
245
ok I think i understand thanks.

So basically the high side refers to whether the mosfet is controlling the positive side or ground side?

I have another question...

As an experiment, I thought I would see how the mosfet would react at different gate voltages. I've used the IRF3205 to make a h-bridge to power a 360watt motor. I'm basically making a very powerful servo motor using an m51660l and it's working brilliantly. I've got my laptop controlling a PIC based servo controller, which then goes through the m51660l, then optcoupled with the h-bridge to drive the motor. Not being very experienced mosfets (why?! I never really bothered before when I could get away with your average transistor). So my question is, what should i use to drive the mosfets because currently i'm using an isolated power supply to drive the gates and I would like to use the same power supply as the h-bridge (12v).

BTW here were my findings from my experiment...

#### jpanhalt

Joined Jan 18, 2008
7,694
Your data don't make sense to me. The IRF3205 is an N-channel mosfet. If the gate is 10V(+/-) relative to the source, the mosfet should be fully turned on, and the source (if low side) should be near ground.

Can you post the circuit from which you made those measurements?

John

#### bluebrakes

Joined Oct 17, 2009
245
This would be my circuit that i'm doing the test on. The two 10k resistors are to pull down the voltage on the gates, so they are open when they are supposed to be. I'm using a TTi PL330QMD power supply (as shown as batteries) to do the testing with and a fluke multimeter to check the output on the h-bridge.

#### bluebrakes

Joined Oct 17, 2009
245
When I was using the bridge to power the motor as a servo, there didn't seem to be any problem (no heat, no current being used when not in use, etc). Also I have tried removing one of the resistors to see the effects, which were none at all. Although it does cause the mosfet to stay on when it should be off.

#### jpanhalt

Joined Jan 18, 2008
7,694
Look at it like this: When Q1 and Q4 turn on, the voltage drop across Q1 is very small, so its source is at roughly 12V relative to ground. If its gate is also at 12V relative to ground, it will no longer be turned on. The circuit will not work. However, you find that the circuit seems to work. Try putting the motor under full load.

John

#### bluebrakes

Joined Oct 17, 2009
245