I found in this teardown analysis of the Nissan Leaf drive motor that it runs up to about 10,000 RPM but is an 8-pole motor. My understanding is that the inverter cost goes up with frequency. The Leaf motor at max speed conditions requires the drive to go up to 600Hz, whereas a 2-pole motor at 10,000 RPM should only need a drive frequency of 167Hz.

Why would the design engineers use an 8-pole motor in this case?

  • \$\begingroup\$ There are other characteristics that are affected by the number of poles. The most obvious is the torque. You can think of the number of the poles as kind of gearing. More poles reduce the speed but increase the torque for the same input. \$\endgroup\$
    – Eugene Sh.
    Feb 21 '19 at 19:06
  • \$\begingroup\$ In particular, for better torque at the low end of the RPM range. \$\endgroup\$
    – Dave Tweed
    Feb 21 '19 at 19:09
  • 1
    \$\begingroup\$ Switching frequency != drive frequency. \$\endgroup\$ Feb 21 '19 at 19:09

The motor magnetic core size for inductance are inverse with rising magnetic field frequency before eddy current losses start to rise with higher order effects.

  • 8 pole motor has 3 phases and 2 slots per phase per pole seems to optimize the weight, torque, power over a wide speed range for their load using rotating rare earth magnets.
  • Torque current is inverse to winding resistance so optimization is done to minimize L/Rs=T for commutation cycle 1/f while minimizing RsDCR for no load and full load.

  • Commutation losses are reduced significantly with commutation frequency with dead-time-shoot thru fixed at max current losses proportional to f switch.

The Nissan LEAF motor specs are;

Brushless Interior Permanent Magnet
Slots 48
Poles 8
Phases 4
Litz eqjuiv 20 x AWG20
max RPM 10,390 = 173Hz
Max power 80 kW
Goal: constant power & high Efficiency (Torque*RPM) from 2000 to 8000 RPM is almost achieved

Goal to minimize core loss and copper loss are tradeoffs with hot spot rise @ 7,000 RPM, torque vs Hp in this design yet 90% of losses are choice of low core loss yet copper peak losses are 1.5x core loss for short durations during acceleration.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.