I think it's possible to generate heat from a motor that is not spinning. It seems like it might be tricky (you don't want it to spin by mistake - I did see your video of the Model S, comments on that below...), but it seems possible to do this if you have individual control of the phases in the poles/motor windings (which they surely do). I would think it would be more effective to use the rear motor to do this (PMSRM) as I thought it was easier to cool than induction motor, since the rotor doesn't heat up much while the stator does (so it follows that it would be easier to extract heat from it too).
Honestly, if the OP numbers are to be believed, I don't have another explanation for the huge amount of draw that he saw. We have quite a bit of circumstantial evidence posted here to indicate that the PTC heater (which does exist,
https://epc.teslamotors.com/?_ga=2....742-1378323594.1533145399#/systemGroups/47381) is about 10-12kW max. I have no idea how the heat from that is exchanged with the cabin (does it use a coolant loop to a heat exchanger or does cabin air go through it directly)? Would be obvious if someone looked in a teardown. The way it looks in the part diagram I guess it's just airflow, but no idea. The part diagrams are lacking in assembly detail...
It makes some sense for the motors/battery & the cabin heat to be separate. That way you're not dealing with excess heat to the motor cooling loop, if someone decides to turn on cabin heat to max.
In any case the OP saw 25kW of drain (25kW*0.75hr / 0.242kWh/mi = 77 miles), so that seems more than you could ever get from just the cabin heater. So it does seem like there's some other energy drawing source. Motor seems like the only other reasonable candidate.
Interesting. There is coolant flowing through there too so I don't know whether you felt spinning or the flowing of coolant or vibration of a stalled motor or what. Even if they aren't using the front motor for heat you might still experience warming since the loops are connected somehow. Or maybe they're also using the front motor...
Seems possible/likely.
It's definitely got ways to convert electrical energy to heat if it's excited in the correct manner. You've got resistive losses in the windings, you can generate eddy currents, generate equal and opposing forces which each generate heat but provide no net torque, etc. And presumably they have complete control over exactly how they excite each of the motor poles. But no idea whether they actually do this. Based on the OP, I would guess they do.
I looked at 2:18 in the video for a while. It's actually hard to see whether the gears are meshed or not (motor and selector)...if you look at where the shafts for those gears go into the housing later in the video, they are fixed position, so I don't understand how they could go from not engaged to engaged... I'm not saying I don't believe what your guy says - I just can't tell whether the gears in question are meshed, and I can't tell how they would move (to engage/disengage), even if they aren't meshed. And normally this is not how gears are meshed with one another...usually gear lash and engagement has to be super precise so adding a moving engage/disengage seems difficult. Gearboxes have synchros and dog teeth engaging the side of the gears for this reason, I thought...the actual engagement points that transfer power are always engaged...on all the gears.
Anyway, I would think you'd hear the motor spinning quite audibly if it were actually spinning and somehow disengaged...