>
>The setup was designed, as I understand it, for a cylindrical bearing. I
>can't remember seeing a cylindrical bearing on a car of any sort in ages.
>
>With the conical bearings in normal use today, wouldn't you want to do away
>with the spacer? I think you would want to put the load perpendicular to the
>axis of the individual bearing cylinder. When using cylindrical bearings,
>the load is perp when driving straight but off perp while cornering. When
>using conical bearings, the only time you're side loading the bearing would
>be when driving straight ahead. I guess maybe if you fudged the spec
>endfloat, you could get the load compromise evened out, but this is probably
>nitpicking. If not, what would the spec fudge be and how would you
calculate it?
Not so. The correct bearing is conical or tapered (to use the more normal
name). Many people throw away the shims and use the 'tight less one
castellation' rule. The use of shims was designed to place the load
correctly and prevent overtightening. Having seen massive damage done
caused by incorrectly fitted front bearings (bearing welded onto the stub
axle) strongly suggest you stick with the approach MG recommended.
Regards,
Graham McCann Rivett, ACT. Australia
gmccann@pcug.org.au
06 2889055
______________________________________________________________
|