shaft's total run-out changes after being mounted onto the assembly

[B89]

hi everybody,
it's a shaft supported on bearings both ends.

http://files.engineering.com/getfil...-359d-4edf-9c35-4459e65bc6e7&file=Capture.JPG

when tested by dial indicator outside the assembly frame, total run out is within tolerance, i.e. 0.1mm. but when inserted into the frame, it changes to 0.3mm. I guess it is the frame that pushes the bearings and their housings to a deflection which can cause a change in run out, but 0.3mm (0.2 difference) is too much. I don't believe that little force can change the run out to this level regarding the shaft dimensions:
shaft diameter = 40 (shoulders) & 70 mm (barrel)
shaft length (span) = 450 mm
what shall I do? any idea?

 

[dgallup]

If the frame is out of alignment enough to distort the shaft then you have to fix the frame.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[BrianE22]

How did you measure the shaft run-out before installation? In the lathe? On v-blocks? The end restraints in the frame are probably different then the pre-installation restraints when the run-out was measured.

If you think the shaft is bent you could try to straighten it while mounted in the frame.

 

[BillPSU]

This is a prime example of why machining of co-axial bearing journals are machine in the assembled state. When you assembled the bearings to the frame there were unusual conditions present such as the bearing mounts were skew to the theorectical centerline. It imparted a bending of the shaft which you saw as increased TIR.You could flip the shafte and bearing journal end for end and see if it improves or is worse. If the bearing housing are symetrical you could turn the bearing housing 180 degrees and see if it helps or hurts the TIR.
You could if all else fails determine the direction of the skew and using a hone stone carefully refit the bearing mount to the frame. Or as the old timers would say scrape it in.

Bill

 

[B89]

I measured the run-out before installation by putting the bearing housings on a flat surface.
so you could tell that the shaft is bending, I will examine this as Billpsu said,