I have a machine that uses 7206 angular contact bearings in a df configuration. The
proper pair is available 1/15/2011 along with a new mounted housing from THK available the
same date.

This is a single direction application, basically a ball screw that is used to press items
together.

If it was a spindle, I have a feeling a heavy preload might be in order since it would
increase rigidity at the cost of bearing life.

The question is how much preload or as I think of it, interference fit if you shim a
bearing? Once you reduce the clearance to zero, you are now compressing the balls.

Anyone here that understands this black magic want to enlighten me? Worse comes to worse,
I have a couple 60 dollar bearings and 0.1 mm shims coming in tomorrow that I'm going to
stick in the bearing housing I'll be switching out tomorrow. I'll just go for no play
plus a shim and see how it goes.

Thanks,

Wes

Wes wrote:
I have a machine that uses 7206 angular contact bearings in a df configuration. The
proper pair is available 1/15/2011 along with a new mounted housing from THK available the
same date.

This is a single direction application, basically a ball screw that is used to press items
together.

If it was a spindle, I have a feeling a heavy preload might be in order since it would
increase rigidity at the cost of bearing life.

The question is how much preload or as I think of it, interference fit if you shim a
bearing? Once you reduce the clearance to zero, you are now compressing the balls.

Anyone here that understands this black magic want to enlighten me? Worse comes to worse,
I have a couple 60 dollar bearings and 0.1 mm shims coming in tomorrow that I'm going to
stick in the bearing housing I'll be switching out tomorrow. I'll just go for no play
plus a shim and see how it goes.

Thanks,

Wes


If you are pressing in one direction only I wonder why you don't set the
bearings in a tandem configuration. I would not worry about preload
since you are pressing is one direction. You just have to have the inner
and outter retainers exactly the same width and that will give you 0
clearance. Preload is necessary only if you have to eliminate all end
slop on the shaft but pressing in only one direction it should not be
necessary for any added preload.

In short go for the no play and if you want you can compute the shim you
need to get the preload that is given for that bearing combination.
There should be a table somewhere for preload vs. shim thickness. I
would leave out the shim.

John

John wrote:

If you are pressing in one direction only I wonder why you don't set the
bearings in a tandem configuration. I would not worry about preload
since you are pressing is one direction. You just have to have the inner
and outter retainers exactly the same width and that will give you 0
clearance. Preload is necessary only if you have to eliminate all end
slop on the shaft but pressing in only one direction it should not be
necessary for any added preload.

I still have to deal with the weight of the press so I can't go tandem. The other end has
a deep groove bearing for radial support.


In short go for the no play and if you want you can compute the shim you
need to get the preload that is given for that bearing combination.
There should be a table somewhere for preload vs. shim thickness. I
would leave out the shim.

When I look at bore diameters, I think they mean the ID rather than the OD. For a 7206 it
looks like +4 to -16 micrometers which is something like .00015" to -.0006" I don't have
shims that fine. I'll have set the bearings up on a granite and determine with a test
indicator what amount of preload may be ground in already tomorrow.

I'm probably putting too much thought into this, it isn't a high speed spindle or a high
accuracy lathe, just a ball screw operated press that resolves to +/- 0.001".

I've been putting in high buck bearings when they crash it, I'm ready to try 62 dollar a
piece unmatched ones in pairs instead of the 350 dollar sets.

Wes

On Mon, 08 Nov 2010 21:34:40 -0500, John wrote:

Wes wrote:
I have a machine that uses 7206 angular contact bearings in a df configuration. The
proper pair is available 1/15/2011 along with a new mounted housing from THK available the
same date.

This is a single direction application, basically a ball screw that is used to press items
together.

If it was a spindle, I have a feeling a heavy preload might be in order since it would
increase rigidity at the cost of bearing life.

The question is how much preload or as I think of it, interference fit if you shim a
bearing? Once you reduce the clearance to zero, you are now compressing the balls.

Anyone here that understands this black magic want to enlighten me? Worse comes to worse,
I have a couple 60 dollar bearings and 0.1 mm shims coming in tomorrow that I'm going to
stick in the bearing housing I'll be switching out tomorrow. I'll just go for no play
plus a shim and see how it goes.

Thanks,

Wes


If you are pressing in one direction only I wonder why you don't set the
bearings in a tandem configuration. I would not worry about preload
since you are pressing is one direction. You just have to have the inner
and outter retainers exactly the same width and that will give you 0
clearance. Preload is necessary only if you have to eliminate all end
slop on the shaft but pressing in only one direction it should not be
necessary for any added preload.

In short go for the no play and if you want you can compute the shim you
need to get the preload that is given for that bearing combination.
There should be a table somewhere for preload vs. shim thickness. I
would leave out the shim.

John


0.1mm (4 thou) is a hell of a lot of preload! Unless you really _must_ have
the preload, mount the (presumably DU or SU) bearings as DF and let it run.


Mark Rand
RTFM

Mark Rand wrote:

0.1mm (4 thou) is a hell of a lot of preload! Unless you really _must_ have
the preload, mount the (presumably DU or SU) bearings as DF and let it run.


Yeah, I figured it out after I found the right table. I've googled all over for bearing
shims but came up empty. I know they exist, I've pulled down housings that had some that
were extremely thin.

Of course my interest in setting preload properly has other future projects in mind where
it might really matter.

Wes