Moglice Problem - Fixed?

I am proposing a method of fixing the Epoxy Acme Nut problem.
Following the manufacturers instructions and using their product will
probably get you good results if you have a brand new screw, unworn and
perfect along it's length. What about those of us with piles of surplus
worn out stuff?
My proposal:

Cast the epoxy nut on the most worn section of your screw, but only
cast HALF at a time. The casting is done in a prefabricated split
housing with undercuts to hold the epoxy and prevent separation from the
housing.
After each half nut is cast, machine the faces flat and reassemble
the halves. Use shim stock to adjust the "split" between the two halves
until you get smooth travel over the length of the leadscrew.
An even better method is to take the same two halves and clamp them
together over the screw with a spring of sufficient force to hold, but
that allows outward movement if the nut encounters binding. I know this
lateral movement introduces error in lead, but the error will be only
appx .0002" per thou of "split" deflection for normal 29 deg ACME
thread. (- AND will self compensate along the length of the screw.
I.E. you may have individual .0002" errors but they won't accumulate.-)
This is the Poor Man's zero backlash solution for old machines. I
am going to try and inplement this on the cross slide of my Atlas 12".
I will let you know how it goes.
I have my "artist's concept" picture as a .jpg I will send to anyone
since I don't know how to upload stuff to the home page.

P.S. If this is not a new Idea, please stop me before I mess
something up.