I went through quite a few threads on madmodder and came up with a system that should work with my new 4 in. Soba RT.
First - the RT:
I was torn between buying a 4 in. and a 6 in. but the price was good on this one and I figured the 6 in. would be a bit big for my small taig mill. I just got it the other day and found a few design problems. Set screws hold the graduated collar and the handle in place. Both are hardened and of course gall the shaft. Small brass slugs were put between the set screw and shaft (ala John Bogs). I tried to centre drill one on the lathe first but being hardened it just dulled my centre drill - should have tested it first of course.

The setscrew for the graduated collar can be seen in this photo. The setscrew for back lash adjustment is just below the 310 mark on the edge of the RT.
The set screw which I assume adjusts backlash did not seem to work. It presses against the worm gear shaft housing and if that is free to move pushes the gears together, thus reducing backlash. The only way I could figure to allow this to properly adjust was to slightly increase the hole diameters for the 4 bolts holding the shaft in place. I snugged the three bolts (one won't fit if it is adjusted properly) up gently then adjusted the backlash set screw inwards till backlash was minimized and then tightened the three bolts. That seemed to do the trick and there is very little backlash now.

Three of the four bolts holding the worm gear shaft in place can be seen in the photo. The knurled brass screw on the left side is the locking knob - it would seen that the designer/builders know that brass against steel will prevent galling - why not do something similar for the set screws??
The inner face of the graduated collar and the outer face of the adjoining holder were also skimmed a slight amount on the lathe and this seemed to smooth things out. I may eventually lap these surfaces a bit as there is still a slight amount of wobble between the two.
The RT centre does not have a taper but has a cylindrical hole below the bottom of the T slots.
I have a small 4 jaw taig chuck and it seemed logical to install it on the RT. It accepts a male 3/4 in. x 16 tpi shaft from the rear.

Taig 4 jaw chuck with RT and finished aluminium chuck holding plate.
I had several pieces of round aluminium about 4.5 in. x 7/8 in. thick in my treasure box and decided to dedicate two of them to the RT. One as a dedicated chuck mount and one as a tooling plate.

Cleaning up the edges and surface of the aluminium round. I did a slight cut at about the 4.1 in. mark as a guide to ensure that I didn't get too carried away. Other than facing each side I didn't reduce the thickness much. That can always be done later if need be.
Once the aluminium plate was cleaned up the centre was drilled and tapped (in the lathe) with a 3/4 x 16 tap. I used the tailstock live centre to keep the tap in line and it gave a good result.

Shows the brass threaded portion at the bottom of the aluminium plate. The portion below the plate just fits into the beginnings of the T slots of the RT. The steel portion fits down into the centre of the RT.
The next part of the project was a brass 3/4 in. x 16 tpi post that screws into the aluminium plate and extends above the plate by about 5/8 in. This extension will allow the chuck to be screwed on the the top of the plate. It is drilled through with a 3/8 in. hole.

3/8 in. steel post turned down to fit into centre of RT.
The final part of the project is a steel post turned down to fit into the hole in the rotary table and left at 3/8 in. diameter to fit into the brass threaded post. This post allows the chuck and aluminium plate to be easily centred in the RT. I may locktite this in place but have not done so yet.
The RT was set up in the mill and four mounting holes spotted near the edge. These will be used to bolt the aluminium plate down onto the RT using T slot nuts.
Another 3/8 diameter post may be placed into the top of the brass threaded sleeve to allow centring the RT in the mill.
As you can see in the first photo I tried to use the side of an end mill to do a final clean up of the edges of the plate (turning the RT to do so). Lots of chatter ensued. Giving this some thought I believe it is due to not locking the X and Y axis of the mill before milling the edge. I will try this next to see if it cures the problem.
Hope this is of some help to someone else with one of these tables. Thanks to others for posting their solutions for chuck mounting an a RT.
John.