PDA

View Full Version : Ideas for fitting leadscrews



irving2008
27-09-2008, 11:08 PM
Been playing around with ideas for fitting the 10 x 2 leadscrews to my DIY CNC mill. Conventional wisdom machines the end of the lead screws. I dont want to do that but I still want to make use of skate bearings for support and some form of thrust control. So I have come up witha ferrule arrangement as per the diagram. This supports a circlip/thrust washer arrangement at the stepper moter end and a threaded element with thrust washer and back-lash adjusting nut (and ideally a lock-nut as well) at the far end. The ferrules are retained on the leadscrew by a 4mm hole drilled through the ferrule and the screw. This will be pinned or the ferrule could be tapped for a 4mm grub screw. The skate bearings are held in bearing blocks machined out of some suitable material (Nylon or poss MDF) set so they are opposing each other.

The other pics show a lash up of one ferrule I made today on my lathe out of some scrap. I need to get some 15mm rod to do the real ones, and grind up some sharp machine tools. The test piece is too short and the finish and accuracy arent great because I dont have any sharp machine tools and there is far too much back-lash in the cross-slide on my lathe, but it fits. Its the first thing I have ever made on my lathe though :)

m_c
27-09-2008, 11:27 PM
My cross slide has over 10thou of backlash just now, but it doesn't cause me many problems. You've just got to compensate for backlash, and realise what you can and can't do.

Looking at the diagram you've got, it looks as though you're planning on controlling endfloat by using a bearing at each end, and then preloading the entire ball screw. If you are, it's not a good way to do it, as due to expansion/contraction of the various components, things can get deformed.
You should control endfloat by having one end of the ballscrew fixed (ie. a couple double angular contact bearings to set endfloat) and the other floating (ie. either a bearing that supports the screw but allows it to slide through it, or a sliding coupling if you're using the stepper motor to support the shaft).

irving2008
28-09-2008, 12:02 AM
My cross slide has over 10thou of backlash just now, but it doesn't cause me many problems. You've just got to compensate for backlash, and realise what you can and can't do.

Looking at the diagram you've got, it looks as though you're planning on controlling endfloat by using a bearing at each end, and then preloading the entire ball screw. If you are, it's not a good way to do it, as due to expansion/contraction of the various components, things can get deformed.
You should control endfloat by having one end of the ballscrew fixed (ie. a couple double angular contact bearings to set endfloat) and the other floating (ie. either a bearing that supports the screw but allows it to slide through it, or a sliding coupling if you're using the stepper motor to support the shaft).My cross-slide has >1mm of end-float and about 0.5mm of backlash, basically the threads are shot! Its usable, just, as long as you keep pressure on it... but then it is 85y old!

Re the thrust arrangement, yes I know thats the right way, but its also the expensive way and I did look for a cheap tapered roller bearing as an option but there weren't any to be found and I was looking for a cheap and easy solution that would work for me most of the time. :) I was working on the assumption that the leadscrew will experience the same temperature variation as the linear rails, which ARE locked solidly at both ends, so while there may be some distortion it cant be any worse than they would experience. Another way to look at it is the leadscrew has a coefficient of 13um per metre per deg C, so for a 400mm screw it will change length 13 * .4 * 10uM or 0.052mm for a 10 degC variation (which is what I would expect in its current location). I'll be happy if the endfloat is less than 0.1mm!

irving2008
29-09-2008, 12:04 AM
Revised arrangement using back to back skate bearings and thrust nut at far end and sliding bearing at stepper end...

irving2008
09-11-2008, 09:40 PM
The previous arrangement put an axial load on the skate bearings which they don't like and as a result the friction is much higher than I liked. Part 4 in the diagram is a acetal thrust bearing from BNL, supplied by RS at £0.73p which addresses the issue.

The only critical dimension is the thrust washer, part 5. The ID of this part must be large enough not to foul the shaft and the OD large enough to butt to the bearing black and not slip inside to foul the skate bearing whch is recessed 0.5 - 1mm to stop this happening.

Robin Hewitt
11-11-2008, 09:44 PM
A belleville washer could save you hours of fiddling about setting that up, but just how do you plan to drill through the screw and where have the bleedin' smilies gone when I need them?

irving2008
11-11-2008, 09:51 PM
what do you mean "drill through the screw"?:confused:

and the smilies are here... :rolleyes:

Robin Hewitt
11-11-2008, 10:02 PM
what do you mean "drill through the screw"?:confused:

and the smilies are here... :rolleyes:

"The ferrules are retained on the leadscrew by a 4mm hole drilled through the ferrule and the screw".

There used to be a handy panel of smilies next to this here reply box, now it's gorn http://www.robinhewitt.net/eek.gif

Lee Roberts
11-11-2008, 10:05 PM
Not if your using the quick reply !