Building zero backlash ballscrew nuts.

[dcprecision]

I am in the process of deciding upon a design for building zero backlash ball screw nuts. Most methods employ disc springs for backlash takeup and I have seen some using compression springs. Has anyone ever tried using Polyurethane flat disk spring washers.

[900steve]

i haven't seen any using the polyurethane washers, but i imagine that since a similar method is used in spider couplings, that it may work. just be sure that the polyurethane washers won't have too much compression set. it may be tough to find the right equivalent spring rate for similar behavior to compression springs.

what about using wave springs instead? there's more consistency, less temperature dependence, lots of spring constants available, and more travel for a given stack height.

-steve

[dcprecision]

Is the purpose of the backlash take up springs to allow for minute variations in screw thread profile and pitch.??? I believe this allows the nut assembly to move axially slightly and rotate freely without binding when encountering such variations. If so, I would think that the highest practical preload with the minimum backlash travel possible is the most desirable situation.

[900steve]

the springs are there to take up the clearance between the nut and the threadform. the variations in pitch are not made better by an anti-backlash nut since it's these variations in pitch which will determine the lead accuracy.

so an anti-backlash nut helps you with repeatability (precision), but if you have variations in pitch, you will not have accuracy.

you can probably use the PU washers, but like i said it's a matter of hitting the right spring rate, which may endup with a larger stack of washers. it's just easier to tune and hit the right spot with a wave washer. the extra travel of the wave washer doesnt necessarily mean more backlash travel, but it gives you a larger preload range to use when adjusting the nut.

-steve

[dcprecision]

Thanks for the info,

The polyurethane discs I was looking at had a compression load of about 700lbs at .025 deflection so they are pretty stiff. These are for a 5/8 od shaft and are .250 thick. The reason I am curious about this is if it is possible to preload at about 100 lbs with these, you would have a pretty rigid assembly and it would behave pretty much the same as disc springs. I dont think any particular advantage exists with using them, just exploring the possibilities.

[900steve]

give it a shot, it'll be a good experiment! but be sure to keep posting and let us know how it works out.

i think the one other thing to be cautious about when using a PU washer is that it's going to have some amount of compression set. so if it's under compression, or even tension for the sake of arguement, the 100 pounds at where you set it to originally will relax to a lower value. i've seen compression set for a PU to be roughly anywhere from 10%-20%. so your 100 pounds over time may relax to say 80 pounds.

-steve

[S_J_H]

here is a decent link - http://www.rockfordballscrew.com/techdata.htm#preloaded

They recommend 10%, and up to 30% preload( maximum) of the screws dynamic load rating. They use 85lbs on the typical hardened 5/8" screws a lot of us use.

I figure a finish pass brings everything back into spec if you blow past the spring preload during roughing on most jobs. Can always tighten them up a bit or even lock them in place if a finish pass is not an option right?

On mine I do have the ability to lock them in place or use the spring.
I sort of doubt with my smallish X3 mill that I'll need more than 70lbs of preload. But we'll see.
Steve

[dcprecision]

This has got me thinking again. The optimum zero lash nut would have the maximum practical preload that allows free rotation, but would only have about .001 of travel beyond that point. This would be enough allow for screw variations between the nut spans but never allow more than .001 deflection. How to do it ???
Take a polyurethane washer and drill 4 - 1/4" dia. holes in a radial pattern. Now calculate the amount of compression necessary for preload. Take that compressed thickness and machine 4 dowel pins to that length less .001". Install the pins into the washer, Assemble and presto, you have it. The washer takes up the small thread variations, the dowels stop compression movement greater than .001"

patent pending

No patent infringements please.

DCO
DC Precision Tools, Inc.

[S_J_H]

I don't see any problems with limiting preload movement to .001".
I think you could do the same with a spacer and coil spring but I sort of like the poly spring idea as you can easily make your own springs.
I think you have an idea worth trying.

Steve