>Are you sure that the motors could be operated under DC? (I simply don't know).

They are series-wound, compensated (meaning interpoles) motors with the addition of special windings to counteract the currents induced between the commutator and armature windings (called "compensating windings") by the 25 hz trolley power. They would run just fine on DC. This was the point of 25 cycles (25 hz to you youngsters) in the PRR electrification: 25 cycles was AC in the transformers and DC in the traction motors.

>Considering that the motors would have to be rewired anyway (also update the insulation),

That's my whole point. They wouldn't because it's so easy to design and arrange for the replacement power supply's output to match the original motors' voltage and current requirements (frequency too, if you've a mind to do so).

>it might be worthwhile to find out whether a comparably rated 3-phase AC motor with the same mechanical connectors would be much more expensive (I have a hunch that there might be light rail motors off the shelf could be used. (12 x 200 kW or so, would give 2400 kW, definitely sufficient).

Complete, total, and utter waste of time and money.

>The example I am referring here is an EMU from the mid-1940s, which was equipped with a slider transformer (similar to those lab transformers which allow to get any secondary voltage). This got replaced with a fixed secondary voltage transformer and a single-phase AC thyristor control (kind of comparable to what is used in the AEM-7).

What is a "slider transformer"? Is it a step-down transformer with variable voltage taps or an auto-transformer with [variable voltage] taps?

>That sounds more like "not so state of the art" in Memphis. I wonder who made the control system of those vehicles.

BTH (British Thompson-Houston Ltd.), or some other licensee of GE. It is a tried and true system, provided you use all the requisite components.

>motors which could be run under straight DC but also rectified (and smoothed) AC

Meaning? The problem is that series-wound motors larger than fractional horsepower ones won't commutate properly without these special inductive compensating windings. There is also a practical upper limit to the frequency (IOW, 60 hz can't be made to work)

>Really decent tracking without pilot trucks became available only in about 1950.

Don't tell the J. G Brill Co this; their 27-A truck predates 1950 by several decades -- nor the North Shore Line (nor the Pullman Company for that matter).