Cosmo,
You are correct in stating that this is "odd". But, there is certainly a definite reason for the change, obviously. Pinpointing the reason is the trouble.
I agree with Anole that the 84E dish is quite a load for the SG2100. I wonder if you are not experiencing a problem with the gear mesh or backlash?
If you grab the dish and gently try to move it by hand, can you see the motor stem move any more than it should or more than it used to?
This could make sense, in a way, if your near true south sat is the only one that still gets reception. Try motoring the dish from the east to this sat
and check your signal level, then come from the west to the same sat and same TP/channel and see if you detect a difference in the signal quality. If
there is a great deal of slack in the gear train, it might be discernible by the change in signal quality of the nearest true south or dead center satellite.
The sats in the further reaches of the arc may be so far off that they cannot be picked up at all because gravity is pulling the dish too far off the tracking.
This could be easily explained by way of age on the SG2100 motor and the heavy load that it has been carrying for the past few years.
At the top of the arc, the weight of the dish and its center of gravity might just be counter acting the slop in the motor gear train enough to keep lock on
the nearest due south sat, but when the motor is driven off to the east or west of the center of the arc, gravity and the weight of the dish might pull the
motor position off far enough to lose lock on the satellite.
I've been rolling this around in my head since you first posted this and I cannot really cite any other defect or error in alignment that could allow this if all the
other alignment angles have remained unchanged.
The backlash would be what I would scrutinize first.
RADAR