Max here; yep, I'm using a series controller.
And YEAH, David Roden, the Curtis wasn't designed for this application.
The one SPST switch I had lying around and used for F/R has a center-off
position. If you have it in the off position (and thus have no field
current) and hit the "throttle," you can push big current through the
armature winding unopposed but for what small resistance those fat
windings have (and maybe a little inductance, though the Curtis ramps
the current). Not good. Your suggestion is right on; first I should have
a two-position SPST, and second, I should make it so the first TWITCH of
the pedal switches on the field current relay, THEN any further pedal
displacement engages the main contactors and then current from the
Curtis. (Then, of course, the shut-off will happen in the reverse
order... first PWM current goes away, then the contactors open, then the
field winding relay opens.)