> >Early EMD switchers start in full series, and
> then change to series-parallel.
> >"Transition" varies with locomotive design. Early
> EMD switchers start in full series, and then
> change to series-parallel.
>
> EMD had 300v traction motors? If so, never heard
> of it before.

The consideration in EMD's case has nothing to do with the voltage rating of the traction motors. The apparent issue is the (lack of) capacity of the main generator. The typical low horsepower rating of switch engines did not justify the cost/investment in the largest main generators which would rarely even approach their maximum ratings. To connect a pair of EMD traction motor strings across a main generator requires around 1500-2600 amps to work at full capacity (depending on model/era of traction motor design). By connecting the four motors in series instead of series-parallel, a smaller (and cheaper) main generator of 800-1200 amps is more than sufficient. Since the typical switch engine is primarily operated at low speeds, this is not really a drawback under most conditions.

The tractive effort formula for diesel-electric locomotives is traction hp x 375 x efficiency divided by speed, understanding that at low speeds the theoretical tractive effort will exceed available adhesion. In other words, the approximate tractive effort (drawbar pull) halves as speed doubles, and the ammeter will reflect that.

I run and work on the electrical systems of various GE and EMD switchers, and GP units built/rebuilt 1950s-1980s; and can attest that there is a wide variation in the electrical arrangements. I believe all the EMD switchers from at least the SW1/NW2 use the full series to series-parallel arrangement. The SW1000/SW1001/SW1500/MP15DC use a permanent series-parallel arrangement with a D32 main generator, though there was an option of the older/smaller D25 main generator and the older transition arrangement on the SW1000/SW1001. Even though the MP15AC uses an AR10 which will support all four motors in parallel at 4200 amps, the motors are in series-parallel with one step of field shunting with the AR10 limited to 2500 amps(per Service Manual). The AR6 would have been a good option for the MP15AC/GP15AC/GP15T but apparently parts standardization on the AR10 won out. IIRC the SW900 manual shows that in switching (instead of road), it operates in series-parallel instead of series. All these variations were apparently done by EMD to improve the operating characteristics of switch engines.

The SD45 manual shows it has up to ten steps of forward transition: series-parallel with 5 steps of field shunting, and parallel with 3 steps of field shunting. Field shunting does not typically cause the loading to "stutter" compared to motor connection changes. Not sure when an SD45 would normally make series-parallel to parallel transition, but it may actually be fairly high (15-25 mph?) due to the number of field shunt positions and depending on gearing.

While temperature does have an effect on resistance, the primary cemf resistance change in traction motors is caused by the magnetic fields cutting through each other.

GE, on its trolley, smaller and electric locomotive traction motors did have a wide variety of insulation valves and voltage ratings, with some of the same models having more than one voltage rating. (On GE part numbers, the suffix can be nothing more than an improved model or an entirely different gear ratio, voltage rating, interlock arrangement, etc.)