This is a very tricky circuit to get your head wrapped around and was discussed some in an earlier topic at https://pinside.com/pinball/forum/topic/world-fair-wheel-only-spinning-half-a-turn. It took a few coaches talking through it and some good live video during the clinic to suss out what the problem was. I'll try to summarize that discussion.

To get to the root cause of a complex issue it helps to understand what should happen before poking at a bunch of suspected things. With that in mind the discussion started with the schematic and an annotated Motor Sequence Chart I had prepared ahead of time (and revised after the discussion). Note that the World Fair schematic does not include a Motor Sequence Chart so I had to borrow one from another game that includes the Motor 1E switch. What's shown below is two charts stuck together because the circuit requires two 120 degree motor turns.

At a high level what happens is that when the ball drains the "Rollover Spotting Unit" (aka Roto Target aka spinny thing in the middle of the playfield) needs to turn starting in the first Score Motor 120 degree turn and the 30 points need to be added at the end of the second 120 degree turn.

With that in mind let's walk through the schematic and chart:
World Fair JK relays (resized).jpg
World Fair Ball Drain timing diagram (resized).jpg
Things start off in the bottom left corner of the chart. When the ball drains a switch in the Out Hole fires the J relay which in turn fires up the Score Motor. The J relay becomes active as shown by the upper horizontal red bar. With the J relay active, the K relay fires when the Motor 1B switch closes at about 80 degrees into the first cycle as shown by the middle horizontal red bar. Once the J and K relays are both active, power is sent to the "Rollover Spotting Unit" or Roto Target solenoid to start it spinning as shown by the lower horizontal red bar. So far so good.

There was some initial confusion on my part about when the J relay and Roto Target solenoid should relax since there are Motor 1E and 3B switches in the J relay lock in circuit. TimMe clarified things when he pointed out that the 1E switch only closes once in three motor cycles because it is activated by a taller post mounted to the upper motor cam. The result is that one time in three the J relay and Roto Target relax when 1E opens at the beginning of the 2nd motor cycle and the other two times of three they persist until the 3B motor opens. Note that both 1E and 3B switches are bypassed or ignored during the 1st motor cycle because the K relay which hasn't yet fired provides a shortcut around them.

TimMe went on to clarify that the reason Gottlieb used two different motor switches is that they found that Roto Units would tend to land on the same few targets if always given the same pulse duration. By varying the pulse duration given to the Roto Target the targets would appear more random.

So it's the overlap of the J and K relays as shown by the red bars that determines how long the Roto Target turns. The K relay also needs to linger to the end of the 2nd motor cycle to add 30 points (by firing the M relay 3 times). The 1D and 1C switches in the M relay circuit block the first two Motor 1A pulses allowing only the last three pulses to get to the M relay.

After watching the J and K relays closely through many iterations we discovered that the issue phillyfan64 was having was that the Motor 1E switch was releasing the J relay prematurely. That alone might only affect the duration of the pulse sent to the Roto Unit. But there was also a problem of unreliable 30 point scoring which was due to K releasing early too.

The K relay lock in circuit has a J relay switch in parallel with a Motor 2B switch. Ordinarily the 2B switch would relax the K relay at the end of the first motor cycle. The parallel J relay switch is intended to keep K active into the 2nd motor cycle by overriding the 2B switch at the end of the first cycle. So J should remain active until after the 2B switch opens and closes. In this case however the 1E switch was probably occasionally opening just before 2B which would relax J and force K to relax the first time 2B opens. When K relaxes at the end of the 1st motor cycle the 30 points wouldn't be added.

Upon closer examination it looked like the Motor 1E switch started to activate as the tall cam post rubbed up along the long straight section of the switch leaf rather than at the 45 degree bend at the end of the 1E switch. It also looked like 1E opened pretty early when activated at all and without much wipe between the switch blades. We asked phillyfan64 to turn the motor so the pin fully activated the 1E switch and then readjust the leaves for an appropriate gap. Things seemed to behave after that.

It was a tricky circuit and a slight misadjustment. Diagnosing it would have been much harder without the live discussion and video.

BTW, I'd recommend renaming this topic to better describe the problem.

/Mark