Ah, how convenient! Lead screw friction was one of my projects this summer at my internship...
I would second the use of a purpose designed lead screw. For one, a threaded rod of any standard size has so many threads per inch that it can be quite slow. Also, UN style threads and ACME lead screw threads are different, UN threads will be prone to binding up as they are suited for use as fasteners.
Now, on to the friction problem. If you look at it the right way it's one of the simplest physics problems around. First, examine what a screw really
is. It's an inclined plane wrapped around an axis. Basically what you have here is the old "
box on ramp" problem from the beginning of every physics class! Your torque becomes a force acting on your "block" (the nut), and you can determine the angle of this "ramp" from the thread data on your screw and a little basic trigonometry.
Set this problem up, and you can determine the maximum coefficient of friction that your system could handle. Lead screw manufacturers almost always have CoF data available on their screw+nut systems (it's easiest to get the nut with the screw). If they don't you can call them, someone there knows it. Find a screw that fits your pitch, diameter, and length needs with an appropriate nut and you're all set!
Any more questions, feel free to ask. If I messed something up, call me out on it!