Single row angular contact ball bearing

[alagaraj]

Hi,
I have bought two single row angular contact ball bearing from SKF which are of most high precision class(7008-CDP4ADGA.SKF).I am working on a X-ray imaging system which produces contrast 30 times higher than the normally available X-ray imaging systems.This system is sensitive to microradian oscillations.It is a simply supported beam kind of a system in which i am going to have these bearings at both ends of the shaft.I came to know from the SKF manual that i should preload these bearings with 60N force.My system only has radial load which comes from the load hanging on the shaft and there is no axial load.During Imaging, we will be rotating the spindle in some tens of nanoradians.I thought of using wave springs to preload them,but i am not sure how i will know about the force i am applying to them.

Now my question is, is there any other better way to precisely preload these bearings and what kind of cage i can use to support these bearings.

Yours suggestions are welcome.

Thanks,
Alagaraj

 

[israelkk]

The correct way was to buy the bearings as duplex pair then when you assemble them you get the correct preload without the need for the spring. In this arrangement you will have a duplex pair at one end of the shaft and another single bearing at the other side of the shaft. The 60N preload seems low what is the weight of the hanging load? Anyway you should consult SKF for the correct bearings and preload for you system.

 

[alagaraj]

Hi,
I have consulted SKF,they only told me to buy this single row angular contact bearing.( I shud probably talk to them again).
The weight of the hanging load is nearly 150N.

Could anyone give me where i can find some info about preloading the bearing.

Thanks for the reply,
Alagaraj

 

[israelkk]

The catalog have all you need including examples how to preload using springs. The spring manufacturer catalog gives the spring rate and maximum deflection so you can calculate the deflection to apply the 60N load.

 

[alagaraj]

Thanks for the reply.Which catalog do you mean?

 

[israelkk]

SKF catalog/s

 

[Philrock]

Axial preloading springs may not workin in this application, due to the extreme precision demands. For a spring to be effective in applying preload, the bearing it is preloading must be free to slide on the shaft (or in the housing, if the spring is acting on the outer race). This means at least a microscopic amount of radial play, which may reduce the accuracy of the system.

I agree with israelkk's 1st suggestion of a duplex preloaded pair. If SKF doesn't seem to understand the virtue of this arrangement, they are not paying attention. The ideal system would have the duplex preloaded pair at one end of the shaft, and a cylindrical roller bearing with slight radial preload at the other end (this part is not easy to accomplish).

Another approach would be to use 2 single row bearings as you have already purchased, and design the mounting structure for the bearing at one end of the shaft radially ridgid, but with some axial flexibility (springiness). Then mount the bearing with the structure slightly deflected, so when it tries to go back to its free position, it applies a load to the bearings.

SKF has the ability to give you all the help you need with project - keep after them.

 

[Tmoose]

DGA is SKF's code for a light preload, universally ground pair of bearings. If you were mounting a pair of bearings at each end of the shaft >ALL< you have to do is clamp the pair axially in either back-to-back or face-to-face configuration and light preload will automatically be attained. If I correctly understand that your pair is separated then some form of "spring" preloading makes some sense.

The bearing seats' (on shaft and in housing) tolerances are not to be taken lightly. Bearings take on the form of their housings and shafts, so running accuracy and speed capability can be chopped by seat errors. Cylindricity, squareness, runout, etc of 0.0001 inch or better are the norm. Manufacturing parts to these tolerances is tough.
Inspecting can be even tougher. To manufacture two bearing bores in a single housing concentric to 0.0006/12 inches is mighty tough. Pretty small Thermal differences with burn up that tolerance easily. Making/assembling Individual housing housings to that level of concentricity takes magic.