Assuming this is a single phase 110v (120v) or 220v (230v, 240v) design, yes, the motor will run 20% faster with 20% reduced tq. It will run a little warmer. But, not so much to damage the insulation. The real question is why? With the availability of 110v/120v 60hz and/or 220v/230v/240v 50hz motors, use the motor as it was designed. If this is a single phase 110v/220v motor, dynamics change considerably. Cap start? Cap run? PSC or Split? Measured variable inductors without any magnetic core from few uH up to 10 mH. Ten mH inductors are larger but still bendable. Inductance depends on number of turns. Adding magnetic core adds cost. We produced laminated, 3% Si steel for 60Hz and ferrite core suitable to 13.56 MHz. For 50Hz frequency converter, we could apply 25 microns thin amorphous steel core. Too many types of magnetic materials can be used. We produced composite magnetic core consisting of both ferrite plus amorphous steel. I prefer no magnetic core: easier, simpler, cheaper. Frequency converter designers could specify custom inductors and then trim/tune to exact value for resonance. Near field radiation pattern needs to be characterized. I designed different shielding. An application maybe using inductors to shield against conducted and possibly radiated EM. my suspicion applies to both of these and this is that you have used some form of normal ferromagnetic (mild) steel for the rotor shaft (and part of the inner diameter of the rotor). The rotor flux is penetrating this solid steel and causing high eddy-current losses. They are high at start because of the increased frequency in the rotor, but they are also significant under steady-state loaf because the skin depth into steel is a lot deeper at 1 to 3 Hz than at 50 Hz or 60 Hz - depending on your supply frequency.