Thought I would give an update on this issue in case it helps anyone else out.
I mentioned previously that I had calibrated the motors, and trained the drives and it did not cure the described issue.
The other night after reading on line advice about training drives I had another go. This time I peformed the task accurately, first calibrating the drives and then training the drives using a 9.7mm EP and Polaris as my point of reference; I also set a slow slew speed (no2) each time to centre the star. This completely cured the rubberbanding, no drift at all when goto'ing an object and no drift after centering. Great!!! Very pleased about this, the reason why it failed the first time, I think, is 1. The terrestrial object was too close 2. I did not select a slow slew speed when centering 3. I calibrated the motors after the train and not before (this was probably the main cause)
So all good from a rubberbanding perspective. However the 'jerk' still persists. After further checking I found it does not happen if I centre first on the dec axis and then on the RA. I mentioned this to a Meade service engineer and he suspected that it could be a handset issue so suggested I reset the handset which I did, and at first in max slew during a goto the jerk did not take place but after I put it into quiet slew, the jerk returned, which is a shame. Not a big problem, just an inconvenience. The next thing to look at is the worm/drive gear in the base on the scope, but I doubt very much there will be a problem there. If I ever get to the bottom of the issue I will let you know!