20mm to 16mm ball screw

[gstprecision]

I am in the process of doing some maintenance to my Novakon NM-135 mill. X Axis backlash is 0.006". I took it apart to take some measurement. The current X Axis ball screw is 20mm. It is a block ball screw, not sure if I explain that correctly. It is a 1 piece block 36mm and the ball screw cartridge is mounted inside of that block, cannot take it out.

Anyway, the current 20mm ball screw cartridge/flange are much bigger than this old one. There is no way I can mount it in, even if I machine the casting.

I can get 16mm ball screw easily that would fit with minimal machining. It is a pretty small machine, is there any risk of going to a smaller ball screw? As an example the Y axis is 14mm??? I don't think you can get 14mm ball screw anymore.

Thanks
GST

[Sparky_NY]

If you could post a pic of the ballscrew that would help a lot.

Meanwhile.... how sure are you the ballscrew is causing the backlash? I ask because on inexpensive machines it is not uncommon for the ballscrew support bearings to contribute a pretty fair amount to total backlash. To a lesser degree, gib adjustment also effects backlash.

[gstprecision]

Here is the Y axis screw. Did not take picture of the X and it is now back together. They are the same exect for size.

I put everything back together. there is a bit of slop in the ball screw. The angular contact bearings did not look the best. I adjusted the preload best I coul, and the gibs seem tight but not too tight.

Thanks
GST

[Sparky_NY]

Can't see much of that nut block on the ballscrew. Are the return tubes external? On ballscrews, its the balls that wear the first and often repacking with slightly oversize balls gives new life and far less backlash to a screw. I have repacked several with excellent results.

[gstprecision]

What you see there is the ball nut. There is no return channel or any mean to replace the balls. It appears to be a cheap disposable ball nut design.

There is a plastic sleeve inside of that block and it appears to be press fitted in.

I can get 16mm ball screw to replace that one easily. I do not think it should be a big deal going down to 16mm as it is a very small machine, but this is just my opinion, there might be more to it I do not know.

Thanks
GST

[Sparky_NY]

You are probably right, 16mm would probably be ok. Only issue is it would not be as stiff and more prone to whipping at high speeds. There are online calculators for critical speed of a ballscrew that are simple to use. I would run the numbers and see how it changes your max speed available. Other than that, the 16mm one would wear faster than a 20mm one (assuming same quality etc) Load ratings of ballscrews are pretty amazing for their size, life being changed as diameter changes.

Here is a link to a critical speed calculator (step 4) http://www.cncroutersource.com/lead-screws.html#

[gstprecision]

Thanks for the calculator. From the best of my knowledge I filled the blanks and it gave me minimum size 0.24”. I should be fine with 16mm.

I normally take very small cuts anyway as the motor is very weak. I have to keep it very conservative.

Thanks
GST

[Sparky_NY]

Glad to help. Critical speed is a interesting thing to see in action, you hit a certain speed and it REFUSES to go any faster. Its a resonance that causes the screw to start to oscillate in a whipping motion. Its also a often overlooked thing in homemade routers and machines. Having support bearings on both ends of the screw makes a huge difference opposed to only one bearing.
I am still curious how that original screw was loaded with balls when it was made, there has to be a return on it, usually a external tube but internal returns are also common but generally have a "plug" at one end so the balls can be loaded. The balls circulate in the nut, thus the need for a return path. I suppose its possible they have a special insert for the nut allowing them to load the balls, then screwing in the main screw after.

Good luck ! Don't discount the importance of those support bearings ! When the time comes, search for the method to set the gibs using backlash measurement, not hard but important also.

[gstprecision]

I ordered a new set of bearings for the X-Axis. I double checked thins morning and it appears that the angular contact bearings are shut. I put dial indicator on the table and another one on the screw end itself. The screw moves about 0.004-0.006" before the table moves.

If that does not fix it I will rebuild the x and y axis in the very near future.

[Baz]

If you use a 16mm screw would you use the same pitch? I think maybe with the same pitch the ramp angle of the thread being steeper will require more torque which could be a problem if you motor is weak as you mentioned. I'm not sure about this though as the smaller diameter might be providing a compensating mechanical advantage. Perhaps someone can put me right on this.

[gstprecision]

Hey Baz. I see you point. I was talking about the spindle motor being very weak. The axis motor are actually overkill for this machine, kinda weird combination.

[gstprecision]

Got the new angular contact bearings in the x axis and the backlash is down to just shy of 0.002”. I was not expecting better from an old machine ao I am happy for now.

I will gather the parts over the next few months and change the ball screw next winter.