I'd only 'force it' square using the motors if it's not much out at all. Depends on the stiffness of the frame. If it's out by more then bolt the frame down as Jazz says and if it's still not good use the formulas in mach3 to compensate as that will completely eliminate any error.

It's just occurred to be that you could purposefully have the gantry built a little out of square, force it with the motors/ballnuts to reduce or eliminate backlash. The pressure on the ballnuts from squaring the gantry would preload both X-ballnuts rather like if you have double nuts with a spring. Clearly if the cutting force is greater than that preload you will get backlash, but for light work it would work. Disadvantage is clearly extra load on the X rails and I suppose marginally more torque required, so feedrate will be negligibly less. I wonder if this is why I have practically no backlash in my X-axis...I'm not purposefully forcing the frame, but I guess it wouldn't have to be much out.

Quote Originally Posted by JAZZCNC View Post
Edit: I know the T nuts did but have any of those Jigs turned up yet.?
Yes they have.

If the bearings are not all the same way it wont affect this. However the force rating of the bearings must be different in each direction as the location of the rows of balls is not symmetrical. I doubt it will make much difference, but you might as well have them aligned so it's even.