I *think* this is what Bruce is talking about (see image attached.) It stands to reason that current draw would be lower with shunt wiring because the resistance of the armature and field are in parallel, rather than in series. If it works, I would expect the motor to have lower RPM and a more constant speed response to changing loads in this configuration.
Of course the original post is about TWO motors, which yields additional possibilities: The two motors could be wired in parallel to each other (but each motor having a series circuit between its own armature and field.) Or, both Fields could be wired in parallel to the track while the armatures are in series to each other; etc. It might be fun to experiment and see how various configurations affect the performance.
I've often wondered whether modern, hi-tech solutions such as the Dallee e-unit, ERR AC commander, etc., do funny things such as holding the field current constant while varying the voltage to the armature, etc. A solid-state circuit could also impart a pulse, or use half-wave DC for improved starting performance. By adding a few diodes, I would think it's even possible to use a command control board intended for a DC motor. @gunrunnerjohn, @bruce benzie have you ever experimented with this? What further insight can you add?
