>>2945527
>How do they even achieve that?

The modern method is to use accelerometers or force sensors with a test jig that spins the rotor. It's trivially easy to get something to within arbitrary levels of balance this way.

https://www.crystalinstruments.com/rotor-balancing

>How does it then operate with DC?

The rotor and the stator are both being fed from the same signal (power source). Any change in that signal is reflected on both parts of the assembly equally. In other words, if you reverse the polarity the stator sees, you also reverse the polarity the rotor sees

You can imagine that, if you flip the magnetic fields one of them generates, they're attracted/repelled by the magnetic fields the other generates in the opposite way, and the motor reverses direction. However, if you then flip the fields of the other, the first is then reversed again, causing the assembly to spin the way it was spinning originally. Since, in a universal motor, you can only change the polarity of both the rotor and stator at the same time, you can never actually reverse its direction without physically rearranging its connections.

>Although wouldnt a commutator still be technically a mechanical rectifier

I don't think so, because commutators don't really have electrical "outputs". Rectifiers, by definition, are electrical concepts, but motors are almost always modeled as their own, complete component. The internals are handwaved and its reduced to something akin to a weird inductor(s). The only time the internals are ever explicitly broken down into into individual components are when doing an in-depth analysis of the motor itself. Even then, it would be more accurate to think of it as a sort of multiplexer than a rectifier, since what it's really doing is activating specific current paths based on an input (the relative phase of the rotor/stator).