broken shaft of 280 KW SQIM 4

[genman196]

We had experienced a broken shaft on our 280 KW,SQIM while it was running under load. We dont know what was the real cause,as it was within its nameplate values and the OLR did not trip. Can anyone please give some hints of possible causes of broken shafts for large motors? thanks a lot.

 

[DickDV]

Misalignment if on a coupled application.

 

[electricpete]

A general question deserves a general answer (I'm sure others can add more).

Why could a shaft fail?

Stresses can be produced from torsional load, bending load (less likely axial load). Those loads can be static, oscillating, or infrequent transients.

Failure can occur as ductile failure, brittle failure, fatigue failure.

Material can be weakened by stress concentrations from machining (for example shoulder without a radius), from attachments (interference fit hub without a radius), by surface finish and corrosion. By bad heat treat.

We had a pump shaft failure that we blamed on corrosion damage weakening the shaft which reduced fatigue strength.

Often looking at the fracture surface gives clues about the type of failure and types of stresses that caused it. There is quite a bit of textbook guidance about what to look for (I like "Practical Plant Failure Analysis" by Neville Sachs.). Also Austin Bonnet has some good papers on shaft failure.

If you told us more about the application or the failure appearance we might make a better guess, but still a guess. Probably Mech E general could give better guesses.

From what I gather the most common scenario is bending load (such as from belt) and failure at a stress concentrator.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[waross]

Quick summary.
Belts too tight.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

I'll add to the confusion. Excessive vibration. Poor material choice. Some repair of the shaft in that area by welding.

If you want proper answers, then post more details.

Muthu

http://www.edison.co.in

 

[desertfox]

Hi genman196

A picture or photo of the broken shaft is worth a 1000 words.
Lots more info needed if you want a better answer.
how long as the motor been in service?
What was the torque load?
What was the torque on the shaft?
etc etc.

desertfox

 

[burnt2x]

More confusion: Did it break at the notch of the shaft? Was the break a clean break on the material or rugged one?

 

[genman196]

Thanks for the valuable info and references you all gave.The motor was belt coupled to a clutch driven flywheel, which allows it to run freewheeling with the flywheel when material extrusion is stopped.The shaft was broken right behind the drive end bearing, inside the motor itself and it was a rugged break.

 

[desertfox]

Hi genman196

If the break surfaces are approximately at an angle of 45 degrees (see sketch uploaded) then its possible that the failure is due to fatigue.
If the shaft is a ductile material subject to unidirectional repeated torsion, then this type of failure can occur due to fatigue cracks spreading due to tensile stresses perpendicular to the principle planes.
How long as the shaft been in service? Is it only loaded in one direction?
If you can upload a picture or photograph of the failed shaft and anymore detail it might help getting a better answer.
Have a look at the sketch I have uploaded and let us know if that resembles what your broken shaft looks like.
If the shaft as been in service for a while then fatigue is a good possibility.

desertfox

 

[burnt2x]

Any idea of the type of shaft material used? I once had a shaft failure (980kW motor) due to brittle fracture (low energy) where the fracture surface was shiny and flat. In your case, it could be what desertfox described (metal fatigue) due to cyclic loading on the shaft. There are a lot of posters who are very knowledgeable on "fracture mechanics" at the mechanical forums, I guess.

 

[genman196]

ive located the broken shaft and took pics,it was a clean break rather than a rugged one, sorry for the error. Thanks for the info.

 

[desertfox]

Hi genman196

Thanks for the picture, it looks like the shaft as failed at 90 degrees to its rotational axis and not as I assumed in my previous post.
The failure picture suggests a shear failure across the section, indicating that the shaft material is weaker in shear than tension (a low carbon steel).
What puzzles me though is the shiny smooth area's which if you look closely seem to have classic beach marking usually associated with fatigue.
Also there appear to be a number of cracks at the outer perimeter which is where I would expect the cracks to start
due to fatigue.
What do the fracture surfaces look like at 90 degrees to the picture you have posted?
I'll try and dig some more information up.

regards

desertfox

 

[desertfox]

Hi Genman196

http://proactivefluidpower.com/shaft_failures.htm

Look at this site above this picture looks very like yours and gives a failure mode as bending fatigue due to misalignment of the coupling.

desertfox

 

[electricpete]

High number of ratchet marks and very small instantaneous zone suggest the primary problem was stress concentration, not high stress. Radiusing approach on shaft shoulder is the prime suspect. There are limitations of shaft shoulder since it has to coordinate with bearing radius - but there are various ays to overcome this.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[desertfox]

Hi electricpete

It looks to me that the shaft as broken on the straight part of the shaft, level with the outside of the bearing and not at the shoulder side but I could be wrong in which case I would agree with you.
But the failure mode describe at the link in my above post seems to fit the bill, which would suggest poor alignment of shaft to power source.
Perhaps the OP could confirm if this is a new installation or as been stripped for maintainance recently.

desertfox

 

[electricpete]

It looks to me that the shaft as broken on the straight part of the shaft, level with the outside of the bearing

I assume the photo is looking at the broken off shaft extension, complete with bearing (no longer part of the motor), based on....

The shaft was broken right behind the drive end bearing, inside the motor itself

But the failure mode describe at the link in my above post seems to fit the bill, which would suggest poor alignment of shaft to power source.

It could be. Neville Sachs' book has a large number of pictures/drawings of shaft fracture which helps narrow down the type of stress. However, it also talks about clues of excessive stress. The large number of ratchet marks is one. It means multiple almost simultaneous sites of failure intitiation which is characteristic of stress concentration all around the circumference. It also shows very small instantaneous zone indicates the load on the shaft at time of final failure was low. The latter in itself is not conclusive, but put them both together along with failure location and known stress concentrator (shoulder) and the diagnosis is strong imo.




=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Correction:
"It means multiple almost simultaneous sites of failure intitiation..."
should have been...
"It means multiple almost simultaneous sites of fatigue crack intitiation..."

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Another correction (last one)
"However, it also talks about clues of excessive stress."
should have been:
"However, it also talks about clues of excessive stress concentration."

The point being what is unusual is not the loading but the the stress concentration.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Let me backtrack and say that to improve the situation we'd like to address all factors: limit stress concentrations and limit loading.

There are seemingly-obvious clues I've mentioned that stress concentration is more relevant in this case, but that doesn't have to be the end of the story...

From my limited viewpoint, determining the nature of the stresses that caused the failure is more complicated....I haven't worked through it myself. Will spend some more time when I get a chance. I didn't mean to downplay anyone else's comments.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[desertfox]

Hi Genman196

Re-reading your post on the 28th Dec I realised I missed an important point as you stated the motor was belt coupled.
This belt coupling would create a rotating bending stress on the shaft thus creating tensile stresses needed to propagate fatigue cracks in addition to the the torsional shear and shear generated by bending.
Whilst we haven't any load figures to number crunch to verify this, I would put my money on rotational bending fatigue.
If you can provide load details and dimensions of the shaft we might be able to prove it but in addition we would need the shaft material information also.

desertfox