One thing about the Citicar - it is a relatively easy platform to experiment with!
While I had experimented with field weakening in the past, I was never able to find how much parallel field resistance should be used to have a noticeable effect. The least amount of resistance I was comfortable using was a 6 ohm, 1500 watt resistor (heater element). When connected in parallel (per various references on field weakening) the heater coil had no effect on speed, so I shelved the idea.
Reading a 100 year-old electric motor book found at Google, I came across a section regarding speed control of series motors. On page 144 there is a reference that infers the weakening resistance would need to be quite close to the field resistance to have an effect.
The field coils in my 3.5hp GE motor have about .015 ohm resistance (according to the manual). Interestingly enough, there was a part sitting in my garage that came with the Citicar, but one that I don't use because I have an electronic controller - the 'low speed' resistor. This 'resistor' (actually a nichrome strip) has a resistance of about .1 ohms - much closer to the field resistance than the 6-ohm heater coil.
I dug out the low speed resistor from the box of old parts, and hooked one end to a field connection at my F/R controller. I connected a wire to the other field connection so I could 'mash' the resistor end with the wire once I was at full, stabilized speed.
So, off I went. Made it to 33 MPH (typical max these days with 50f batteries). Stabilized there for a few seconds, then connected the low speed resistor in parallel with the field. No untoward sparks (good thing) and the speed increased (albeit gently) to a whopping 36 MPH. Not bad for a junk box experiment - almost 10% gain!
To make sure it wasn't just a fluke, I turned around and did the same thing a second time - stable at 33 MPH, connect the low speed resistor in parallel, and... 36 MPH! (I actually passed a Ford Excursion...)
Disconnecting the field weakening resistor made the Citicar slow down, but not dramatically - took a while (about 15 seconds), but it decreased to 34+ MPH and stayed there. The next acceleration cycle (without field weakening) stabilized at 33 MPH like usual.
The book infers that you can't leave the resistor there during acceleration, as the field will be too weak to produce enough starting torque. It is only for 'steady-state' full speed (i.e. accelerator hammered down on a flat road). My next step will be to get a relay can to switch 'on' when I need a speed boost for those 35 MPH roads.
Makes me wonder how it would work with those of you that have other brands of series motors?
Disclaimer... I haven't measured the amperage or temperature of the motor in this configuration, so experiment at your own peril...