Ballscrew support, fixed end.

[RP Designs]

I've read through all the reads I could find and couldn't find a definitive answer. Is there a difference in the fixed end support? For instance there are a ton of C-7 quality fixed end supports on ebay fairly reasonable but is there any difference between them and a support for say a C-3 ballscrew. They both do the same job, preload the ballscrew so why make two different ones. Anyone know a definitive answer. I got a good deal on a new C-3 ballscrew and need a fixed end support, but don't want to spend a ton on one either.

[louieatienza]

The difference is the accuracy. A matched pair of C3 anglular contact bearings probably cost about 10 times what a C7 set cost. I'm sure you could use a C7 fixed end wih your C3 ballscrew, but then you effectively make your C3 ballscrew a C7 ballscrew,,,

The bearings can also be arranged face-to face, or back to back. I believe the face to face method provides more preload ability and accuracy, but don't quote me on that.

BTW, you're not preloading the ballscrew, you're preloading the bearings, which creates the "fixed" end.

[RP Designs]

The difference is the accuracy. A matched pair of C3 anglular contact bearings probably cost about 10 times what a C7 set cost. I'm sure you could use a C7 fixed end wih your C3 ballscrew, but then you effectively make your C3 ballscrew a C7 ballscrew,,,

The bearings can also be arranged face-to face, or back to back. I believe the face to face method provides more preload ability and accuracy, but don't quote me on that.

BTW, you're not preloading the ballscrew, you're preloading the bearings, which creates the "fixed" end.

That is what I meant to say, preloading the bearings. I have been doing more research and it seems that the fixed end is the important side, so I will probably just suck it up. I don't really need C3 but I got C3 for less than C7 prices in the size I needed, so...

[wildwestpat]

Just take a deep breath and think about the differences between C7 and C3 specifications for the bearings. The increased cost relates to the smaller tolerances permitted in the C3 specification. You do not say how big the lead screw is or how long or how fast it is to rotate. These are the factors that determine the design of the fixed bearing and the need / type of bearing support at the non driven end.

For small lead screws it is often OK to use flanged deep row ball bearings in a very simple plain bore housing. Consideration of the thrust load on the fixed bearing will determine the type of bearing that is required along with any margin required to obtain satisfactory life. In practice there is not a lot to chose between angular contact and normal deep row ball races when used as a pre-loaded pair. Just do not use filled races unless precautions are taken to ensure the pre load is not going to force the balls into the filling slot rembering the filling slot can take different forms. That said races with filling slots are not common items. For moderate thrust load there are double row deep row ball races that are pre-loaded during manufacture and are suitable for most lead screw supports. From your description you have a better tolerance lead screw and that can only be good in terms of location precision along the axis. The use of C3 bearings in the fixed brearing block will do little if anything to improve the accuracy over that obtained from a C7 spec bearings used back to back and suitably pre-loaded as the rotational speed is low. If you want definative answers look at the data sheets on ball screws by NSK.

Regards

Pat

[mxtras]

Somewhere on this site there was a retired bearing engineer that probably forgot more about high tolerance bearings than all of us combined currently know. I can't recall his screen name as it has been probably at least 3 years since I have been to this site but I will see if I can track down his posts on this very topic and report back. Or you can search for bearing preload or angular contact bearing - you might stumble across his posts. It would be worth your time.

Scott

[RP Designs]

FWIW, I looked at NSK's brochure for the ballscrew I have as well as their rolled ballscrew and they specify the same fixed end support for both. Fortunately I found one on ebay (NIB) relatively cheap and will use that.

[wildwestpat]

As the Japanese NSK site appears to be very slow I have attached the document which is part of their Motion Control series. This shows the use of a pair of single row flanged ball races for the fixed end support of a small ball screw. I have used this design in the compound slide of a small lathe.

Regards

Pat

[wildwestpat]

The need to support the far end from the drive for the ballscrew depends upon the diameter and unsuported length of the screw plus the maximum rotational speed. Here is a link to a paper that sets out the design criteria with some useful graphs to save doing the maths!

http://www.jena-tec.co.uk/JT%20Serie...gn%20Guide.pdf

Similar guides are available but this is the easiest to use.

Hope this helps debunk some of the bearing mist -
Regards Pat

[RICHARD ZASTROW]

I think the bearing engineer you refer to is NC Cams. Unfortunately, our good friend has passed away. MHRIP

I've been reading your posts. I may have forgotten the part on C3, C5 & C7 on angular contact bearings but I think that refers to ball-screw accuracy. It could also refer to internal clearance of radial ball bearings like deep groove ball bearings.

Angular contact ball bearing accuracy grades here in the USA are ABEC 1, 3, 5, 7, & 9.
ABEC1 being the lowest, ABEC9 about as high as they get generally. The higher the number, the higher the accuracy.

When mounted face to face, AC bearings are a little more forgiving as to alignment. Mounted back to back, they are more rigid.

Ball screw support bearings are mostly axial (thrust) bearings, usually 60 degree contact angle. In lower force applications, a lesser angle is probably OK. Check the bearing specs to be sure.

The ball screw support bearings are usually made with fairly high preload so they do not like sustained high speed.

Lots of information about bearings out there. Read the tech sections of the bearing catalogs, good stuff.

Too bad we don't have Dennis anymore.

Dick Z

[RP Designs]

I think the bearing engineer you refer to is NC Cams. Unfortunately, our good friend has passed away. MHRIP

I've been reading your posts. I may have forgotten the part on C3, C5 & C7 on angular contact bearings but I think that refers to ball-screw accuracy. It could also refer to internal clearance of radial ball bearings like deep groove ball bearings.

Angular contact ball bearing accuracy grades here in the USA are ABEC 1, 3, 5, 7, & 9.
ABEC1 being the lowest, ABEC9 about as high as they get generally. The higher the number, the higher the accuracy.

When mounted face to face, AC bearings are a little more forgiving as to alignment. Mounted back to back, they are more rigid.

Ball screw support bearings are mostly axial (thrust) bearings, usually 60 degree contact angle. In lower force applications, a lesser angle is probably OK. Check the bearing specs to be sure.

The ball screw support bearings are usually made with fairly high preload so they do not like sustained high speed.

Lots of information about bearings out there. Read the tech sections of the bearing catalogs, good stuff.

Too bad we don't have Dennis anymore.

Dick Z

That is a shame about NC, I truly enjoyed reading his posts.

For me, I think that it is fair to say that what I know about bearings pales compare to what I don't know about bearings. When I originally posed this question I made a basic assumption that the support bearings, being angular contact bearings which would need to be preloaded to remove axial free play would not necessarily need to be anything special. From what I have read on ABEC classifications, I am sure someone will correct me if I am wrong, they are simply dimensional tolerances for the bearing itself, but don't specify things like axial or radial run-out so in essence that is left up to the manufacturer. The end support I wound up buying was recommended in the catalog for the Class 3 ballscrew I purchased. Incidentally, the same support was also listed for the rolled ballscrews from NSK. The MSRP suggests it could be loaded with matched bearings since it was close to $500, fortunately I didn't pay anything near that or it wouldn't have happened. But the basic question is: If I am going to support a Class 3 ballscrew and am going to use a pair of angular contact bearings which I adjust the preload so that the axial play in the bearings is essentially zero anyway does it really matter (for most applications) if I use a bearing that someone is calling a class 3 bearing mount or a class 7 mount as long as I mount and adjust them properly?

[rowbare]

The system of bearings, mounting, screw, nut, coupler all come together to define the accuracy of the system. The various components affect different aspects of the overall accuracy.

The difference between a C3 and a C7 screw is lead accuracy i.e. all other things being equal if you turn a C3 screw a certain number of degrees you can expect that the distance the nut has moved is closer to what would be expected than if you turned a C7 screw of the same pitch the same amount.

So by using a C3 screw, you are improving the lead accuracy portion of your system. That is something you can't tweak out of a system so you are ahead of the game. In addition, one could reasonably expect that the screw/nut portion of backlash will be less in a C3 screw than a C7 since the more accurate lead means that you don't need quite as much clearance to prevent binding.

So I think you are ahead of the game. Use the components you can afford and do a really good job of assembling them and if need be massage them a bit to make them better and you are off to a great start.

bob

[michaelthomas]

I don't think it matters.....but I'm one of those guys who has used deep groove bearings in place of angular contact bearings. I had no detectable play in the axial plane, and they were still as good as new after a year of medium duty use on a 26 x 52 router.

The deep groove bearings aren't nearly as impressive as the AC's when documenting a build log......but they have performed well for me.