Khaled:
The 254mm table, 90:1 ratio is a much better choice. However I would still use a reduction from stepping motor to worm screw of table. Since the table worm shaft is much higher from the base, a 60t pulley or even an 80T woulld be better. If you used 20t to 80t timing belt reduction, the steps per would be exactly 200. This would make 1 turn of the motor = 1 degree of rotation. This will have much better torque, and the motor will be operating in its better range. The the cutting action will not be fast anyhow, as a 10mm ball cutter at full depth in the finish pass cannot be pushed fast at all. It also will require a quite LOW spindle RPM, as a great deal of cutter surface is in contact; it must be a slow, steady cutting action with coolant. So, the greater reduction all works for the positive.
As far as backlash, there is usually a means of adjusting the worm screw to ring gear to account for general wear. In addition to getting less backlash, if you put some kind of slight drag on the table with a friction material like a piece of leather, it will stay tight against one side of the ring gear; and the cutting force is a constant load in one direction only. You are not making a cut in a forward/reverse situation, like an X/Y axis anyhow. If you get it apart, you will probably find it needs lubrication if it has sat for a long time. The backlash will not be a problem if all things are setup correct. In rotary axis machining, having everything rigid is very important.
Testing the cutting action with a standard cutter of smaller diameter as I suggested, you can test how the cutting action works best by programming the cut from left to right, and then try cutting from right to left X axis motion. One direction may prove better due to the cutter forces, and backlash. When starting the cut, you can also do an MDI, moving the table and X axis backwards from the start position about 10mm, and then back to the start point, and all backlash will be out of both the X axis, and the rotary table. You can get all the correct settings, feedrate, best cutting direction on a test blank.
Once your mechanical motor setup is done, spend time getting the A axis motor tuning working smooth. There are some good trial and error tutorials in the forum how to get the best performance with stepping motors. The Velocity and Accel settings for the Rotory table will be quite a bit higher than the X or Y axis settings. Just keep increasing the numbers, and listen to the motor sound, and smoothness. It will eventually stall, or sound really bad. Back off the settings about 10 or 15%. You will get a feel after testing, what works best. It is all a trial and error unless you have the high-end drivers that have computerized analysis feedback.
John