VESC: Approximating Power Losses and Heat Production?

Hi, I am new to VESC and can’t find how to calculate or approximate the losses and heat produced by the controller.

I am looking at options like the Trampa 100/250 or Tronic X12 to power an e-bike hub motor with peak phase of 120A.

There is also the dilemma of choosing between a 100/120/150V MOSFET controller, and how much difference there is in power losses. Should I sacrifice a little bit of top speed and go for a 24s3p build, or should I go for a 27s3p build (9 more cells, better top speed, but more losses in the VESC)?

I would appreciate any information that can help me, including technical details. I’m not afraid to spend time really understanding how a VESC works.

2 Likes

I’m a skateboard nerd, so don’t know much about bikes.

It’s my understanding that losses are more proportional to current, not really voltage. Besides, 9 extra cells is almost certainly going to add more energy then any losses would increase.

If you want to learn more about vesc, there’s a bunch of great threads here on the forum. Search bar in the top right :slight_smile:

2 Likes

Sometimes, but not really

For conduction losses yes that’s how it works, current squared times resistance. But a lot of the losses in ESCs are switching losses, the energy lost while the MOSFET is partially conducting while transitioning from off to on. The energy loss is a function of time taken to switch, current flowing, and voltage to be switched

I don’t have an answer to OP’s questions though

2 Likes

For higher battery voltage, the MOSFETs in the controller must withstand also a higher voltage, which results in a higher Rds(on) value; in other words, they have higher resistance when conducting.

For example, Tronic offers two versions of the X12 controller: a 24s version rated for up to 300 amps continuous and a 30s version rated for up to 250 amps continuous. Both have the same form factor and design, but the 30s version generates more heat per amp.

Tell me if I’m wrong but there’s two switching events happening. One is for making an effective voltage with pwm and the other is the motor electrical frequency and you can reduce the pwm frequency for more efficiency

Yes, I believe the default switching frequency for VESC is 20 kHz. It can be lowered, but doing so results in a lower-quality sinusoidal signal for the motor, which decreases motor efficiency.

How low you can go depends on the motor. For larger hub motors, it might not be as much of a problem as it is for smaller and much faster skateboard motors.

Here’s the answer I got from Tronic. Very fast and good customer service, by the way!

"Hi,

Heat in watts from the controller is about
(200/sqrt2)²*R +switching loss
R is 0.0017/2 for the 24s and 0.0034/2 for the 30s
So 17W plus switching vs 34W plus switching.
Switching is hard to calc but probably similar to resistance losses."

*In this example, the losses are calculated for a phase current of 200A.

2 Likes

So lower voltage with 24s is more efficient than 30s? What’s “plus switching”?

It means that you should add the switching losses to the 17/34 W.

If these are approximately the same, then the total losses would be around 34/64 W.