Hi,

one worm gear shaft broken one week ago, see pictures attached.
Can anyone give me a clue what might be the cauae from the damaged crosss section? Fatigue, overload or something else. Why?

Kevin

XE03-MDO_GB_-1st_shaft-broke.pdf (664 Kb, 153 downloads) Pictures

Kevin,

This looks like a rotating bending fatigue failure. The origin is probably in the corner of the shoulder where the inner race seats. There seems to be some spalling on the rollers as well, could have been some vibration going on which would contribute to the failure. I would investigate the alignment of the pump bearings to the shaft and take a close look at that shoulder. It could be a sharp corner which would be a stress riser. A better design would be an undercut radius. If you want a good reference book on failures pick up "Understanding How Components Fail" bu Donald J. Wulpi. This failure looks like the picture in figure 23 on page 150.

Good Luck,
Charles

I agree with CGTS about the rotating bending fatigue. It has the appearance of low stress with high stress concentration.
The high total stress at the exterior of the shaft at the exterior of the shaft causes failure to begin at many locations. (info from Practical Plant Failure Analysis from Neville Sachs) From the photo it appears that the shaft shoulder has a razor sharp radius, it looks like it just stepped down.
Some questions:
Was this a "home" made shaft? and how long this shaft has been running?

We have 4 gearboxes running at lines and share two spare GB from gear vendor. They had been rebuild several times by changing bearings but this is first time we have a shaft broken and need shaft replacement. They have been in service since 1992 but we can't tell how long it has been running for each gearbox(or shafts).
Any further suggestion if more detailed inspection need and how to prevent it from happening again.
Thanks, Kevin

Kevin, I've just seen this one, I couldn't get the photo to open before.
I agree with above, there doesn't seem to be a radius at the shaft shoulder, although I would undercut the face with a radius tool rather than undercut the shaft.
Best regards,
Joe Mc Cormack

I personally can only guess. There is some info that would help me (or anyone for that matter) make a better judgement. What is the speed of this gear box ? Have vibration analysis been taken on the gearbox before and what did they look like ? I also noticed that is very clean, what type of lube was used and its condition ? What I can demiss from the photos is that the bearings look good and so do the surfaces of the gear. The break is fairly clean, so twisting is not visable. That would indicate something locked up or mechanical stress somewhere in the drive system. From a metallugical point of view the shaft does not look to be flawed. There is some slight discoloration but hard to tell what that is from in the photos. The business end of the shaft, where you might have had a coupling, looks questionable. How was this gearbox aligned when installed. Was this gearbox rebuilt before ? what was the back lash set at ? This being said, and the fact the break is right up against the bearing, I would have to say there was definitely a stress on the shaft itself, not the bearings or gear. The shaft may not be true or you might of had an alignment or back lash issue.