need help - direction of rotation fractured shaft 2

[kingnero]

Please see attached file.

Could someone help me out here?
I need to know what fracture image comes from which direction of rotation.

I do not see these things very often, so I once made notes (and photographs) for myself, however now I can't find these documentation...

Any other - clearly - explained reference is also good, however most of them are not very clear.

As per the image attached:
when looking from the right hand side (point of view is marked) along the axis of the shaft,
and the motor is turning clockwise,
will you see image (B) on face (1) ?

Many thanks in advance...

 

[metengr]

kingnero ;
From the information you provided, looks like rotating bending fatigue crack propagation under low nominal stress, clockwise shaft rotation.

 

[desertfox]

Hi Kingnero

Does fan pulley at the end of the shaft have a belt fitted ?
if so the shaft could fail by one way rotational bending in the direction from the motor shaft pulley to the fan its driving due to the belt tension.
If it is a bending fatigue failure then the faces of the shaft will be roughly parallel to each other.
Does the motor rotate in both directions? or only clockwise as you have indicated, I am not sure why your concerned about which way the shaft a rotated, to me and I could be wrong it would look the same on failure, whether it rotated clockwise or anticlockwise.
Have a look at this site about shaft failure:-

http://www.plant-maintenance.com/articles/rcfa.shtml

desertfox

 

[kingnero]

This is a fictional example that I sketched.
However I'm currently encountering a broken shaft, that has roughly the same "face" than the one I drawed.

There is no bending in the shaft (or not apparent by the first look)

I want to exclude the possibility that the pump was driven (by the electric motor) in the wrong direction.
But I don't know how to interpret the failure "face".

From the 12 and 5 o'clock position I can conclude the direction of the motor is correct.


What would it mean if:
1) the area of the final rupture zone is larger than drawn
2) the position of the final rupture is more towards 3 o'clock instead of at 5 o'clock ?


Am I correct in thinking that a small area of the final rupture zone (ag. 10% (or less) of area of shaft) means an overdimensioned shaft?

Am I correct in thinking that a 12 and 4 o'clock position [of initiation and final rupture] means higher rotational stress than 12 and 5 o'clock position?

 

[desertfox]

Kingnero,
A larger final rupture zone would indicate that the shaft was heavily loaded.
For a fatigue failure to occur you need a stress concentration like a keyway for instance, or a slight groove or undercut on the shaft and fluctuating tensile stresses, as fatigue cracks only grow under tensile stresses. The failure you have sketched is a typical one way rotational bending failure.
What is actually on the end of the motor shaft? Is it a pulley belt or is the motor directly onto the pump?
If the shaft has failed under torsion then the failure faces will be approximately 90 degrees to the axis of the shaft. However, if the fracture faces are at 45 degrees to the axis of the shaft that would indicate cracks growing on principal axes due to principal stresses.
The latter type of fracture (45 degrees) can occur under torsional load so for instance, say the pump impeller was jammed, then the driving shaft would obviously continue to try to rotate, thereby generating tensile stresses along the principal axis.
Have a look at the link I left you in the last post and do you have any other information on the failure eg how long has the shaft been in service?
desertfox

 

[EdStainless]

a pump you say, did it have a backflow preventer?
Power off, flow start turning the pump and motor backward, power is restored and motor restarts, shaft snaps.

We had this happen with a fan that the wind was spinning backward when someone hit the start.


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[desertfox]

Hi kingnero

Have a look at this link:-

http://books.google.co.uk/books?id=wG4EXTXm0MoC&pg=PA77&lpg=PA77&dq=

failure+of+unidirectional++rotating+shafts&source=bl&ots=
k3B5pEpyHC&sig=bK2eptIQ9rWa55YGNulQ54oOaf8&hl=
en&ei=b6u6Tf7uKsav8QOej93GBQ&sa=X&oi=
book_result&ct=result&resnum=2&sqi=2&ved=
0CCcQ6AEwAQ#v=onepage&q&f=false

It gives various failure modes but no clue to actual rotation.

see pages 70 to 81

 

[1gibson]

CW rotation with the load locking up would fail the same as CCW rotation with the driver locking up (inertia of load causing the failure.)

 

[redpicker]

CW rotation with the load locking up would fail the same as CCW rotation with the driver locking up (inertia of load causing the failure.)

Excellent point! The fracture marks/fracture orientation cannot tell you the direction of rotation, only the direction of the principle stresses.

rp

 

[metengr]

CW rotation with the load locking up would fail the same as CCW rotation with the driver locking up (inertia of load causing the failure.)

For a one time, overload event, yes, but not progressive high cycle fatigue crack propagation in service for a rotating shaft.

 

[desertfox]

Well I am not convinced that the failure was caused by running the pump the wrong way, if it was a one off failure would we be seeing fatigue marks?
If its a fatigue failure how many times do you think someone would run the pump the wrong way before noticing.
However without further info from the OP we can only speculate.

desertfox

 

[kingnero]

As said before, the pump was not running the wrong direction, I just wanted to make 100% sure nothing happened that no-one was aware off.

The motor drives a hydraulic pump through a rather long shaft (55mm dia, 6 m long) because of the environment.

There are multiple bearing blocks, which have been aligned by a bearing manufacturer with the proper tooling for it.

The failure happened at 1/3 distance between two bearing blocks. No keyway, no other visible starting point.

After the failure, both motor and pump run freely.

Shaft has been over 12 years in use, pump has been revised several times, motor gets new bearings each year.
It has a flexible coupling to the shaft, so minor misalignments are forgiven. The motor does get measured when re-installing.

We sent the shaft off to be examined.

One possibility is that the shaft has been damaged when the motor room was painted. Our two mechanics don't know what could have happened, they haven't been round the motor for some time (confirmed).

I hope of getting the results soon.
Thanks in the meanwhile for your ideas.

desert fox, thanks for the link.