198 mm Stator OD and 205 mm Rotor ID performances

Dear Member.
I have several questions regarding this motor's sizing that corresponds to PMBLDC Direct Drive chinese production by Mxus, QS motor and Others.
As you know the range of power is 350-5000W but the Pedelec's regulation imposes to limit battery voltage up to 48 VDC.

1. one of the available configuration, fine known as I saw discussed in this forum, has stator jacket with 51 slots and rotor annular ferromagnetic steel with thickness 6.5 mm with 46 glued magnets. This is used up to 1000 W nominal power with flat magnets 13.64-30-3 mm /51 Windings (4 turns-15 Parallel-0,53 wire) and for this I have dyno test for unloaded speed of 651.6 rpm with very interesting efficiency. So that I'm interested to improve firstly this configuration or like this with the goal of high efficiency and high temporary torque resistance.

2. the second configuration called V2 and V3 has the same diameters but uses different material and reduced poles to 32 magnets of more sizing.

I saw here several simulations and good competence so my question are:

A. What is the maximum theoretical unloaded speed we can achieve with configuration similar to 1. also acting by controller?
B. what can be suggested to improve efficiency?

Please condider that to increase torque we will use this motor assembled to a reducer having 3:1

Thanks for your suggestions

Most common hubmotors use 0.50mm thick laminations in the stator. the current trend for higher-performance hubmotors is to upgrade to a thinner lamination. 0.35mm is the next step, and recently some companies have even gone to 0.27mm. The thinner lamination reduces eddy-current waste-heat (heat produced that is not a result of adding more power to the motor).

Thinner laminations also reduce cogging in direct drive hubmotors, (geared hubmotors freewheels, so it isn't an issue there).

Some of the V2 and V3 designated hubmotors have an aluminum stator support that acts as a "heat sponge", so an aluminum core dramatically raises the temporary peak amps that a motor can survive, however...hubmotors have a poor heat-shedding path. If you add ferro-fluid (Google endless-sphere.com) then the heat-shedding path from the hot stator to the aluminum side-plates is dramatically improved. By adding an aluminum stator support and Ferro fluid, you can spec a smaller motor than you normally would in order to get the desired power goal.

MXUS has a 28mm wide stator motor with Aluminum core, and Edge has a 35mm wide stator motor with aluminum core. MXUS is most well-known for their 45mm wide stator motor (the 3000W labeled hub).

If you plan to run to higher RPM's than normal, a lower pole-count is desirable. This reduces eddy currents due to the reduced pole-switching. Lower pole-count motors can also run to a higher RPM without exceeding the E-RPM limits on certain controllers. If your controller has an electrical switching frequency upper limit, the RPM and pole-count must be carefully considered. Low pole-count motors have a wider variety of controllers to choose from. I only mention this because you stated this motor of yours will be a non-hub mount, using a 3:1 chain ratio, so motor spins 3 times for every wheel revolution.

"Wheel diameter to speed/RPM chart"
https://endless-sphere.com/forums/viewtopic.php?f=28&t=16114

Thanks for the reply. Question is which can be the achievable max speed with described Mxus configuration 46x30 mm magnets,51 windings, steel jacket, 0.5 mm lamination. This is a new project and we need of about 800 rpm. Is it possible with this standard configuration or it must be changed?. How the max. speed is calculable? We are looking for a specialized guy interested to collaborate as engineer with our start up company. Motor is anyway an hub motor.
Thanks.

Thanks for the answers Spinningmagnets. Today in Mxus we have tested a standard 198 mm stator OD with 3 turns of windings instead of the previous 4 turns with motor has achieved about 650 rpm as already written. This new sample has achieved about 800 rpm but dino-test give no data below 600 rpm because of the motor stop. Have you an explaination for this?
Is it impossibile with 46 magnets and 51 slots to achieve more than 650 rpm with good curves at 48 V due to back fem or what?
Thanks

Sorry for the mistake...dino test stops under 697 rpm as for the attached...opinions?
Thanks