More power = less ESC heat?

Using the basic ESC inside a Pelican case, along with a 13.4V LiFePO4 battery, I noticed that the ESC generates more heat at 60% throttle than it does at 100%. I put a Govee bluetooth temp recorder inside the case and then ran an M200 motor with a weedless propeller in a water tank for 1 hour. Temps rose from 75F to 99F. I then turned the throttle to 100% for half an hour and temps dropped 4 degrees. During a previous test I saw similar results but thought it was due to a change in solar load on the battery box.

Pointing this out because I find it interesting and it may help someone else out.

Also, sharing the idea of using a temp recorder in an enclosure to know what’s going on but I assume others have already done this.

FWIW, my project is a kayak motor.

1 Like

Hi @Knotty

This is an interesting test and an interesting observation, but I guess I don’t find it all that surprising.

The output transistors (MOSFETs) on an ESC perform two separate roles: they commutate the phases of the 3-phase brushless motor that is the thruster, but they also PWM the output signal to serve as a throttle; in EE terms the ESC acts as what’s called a buck converter.

The process of PWM’ing the mosfets is not without loss. There are at least two different loss mechanisms here, one is that you’re sloshing charge in and out of the gate of the mosfet every time you turn it on or off, the second is that, every time you switch a mosfet, it spends some amount of time in what’s called the “ohmic region” where it’s dissipating a lot of power.

When an ESC is running at 100% power there’s no PWM action going on- the only time the mosfets switch is when the motor phases need to be commutated. So while there is a lot of power dissipation going on in the channel resistance of the mosfets, the two loss mechanisms noted above go to near-zero.

Designing a good ESC requires picking parts to optimally reduce power dissipation at the intended operating point of the ESC. It’s an interesting challenge!

Hope this helps.

-W

1 Like

Thanks for the explanation Walt. I knew it was the MOSFETs but my degree was in mechanical engineering, so I only have enough EE knowledge to just get by (or get in trouble).