Part failure - need help figuring out why - Stress Concentration

[DWLetourneau]

Hello everyone I have a problem that I need some help with.

At my place of work as a co-op we had a part fail. This part is a shaft thats dia. 6" and steps down to a dia. 4.125". There is a 3/8" fillet to help reduce stress. The material is AISI C-1015HF, I believe. This part snapped in half at the step down between the two diameters.

I have two questions...

1)I was under the impression that for ductile materials, stress concentration factors should can be ignored. This part, I beleive, is considered a ductile material, but it has been heat treated for hardness. So at what point does a part go from ductile to brittle, and when should stress concentrations become a factor again (In the case of considering the stress at the fillet of these to diameter features)??

2)Also I have another question about static loading. If you have a shaft on the ground and lift it up by the ends and held it say 3' off the ground.When you pick the shaft up a moment created, correct?. Is this still considered a static load even though there is a moment created in lifting the object?

If you have any questions let me know. I can draw a quick sketch.

Thanks.

 

[Gymmeh]

Thank you for clearifying.

 

[DWLetourneau]

Hey fellas thanks for the replies. Sorry I have been out of the office (I'm actually still in school and only work on Tue/Thur part time). I have attached a sketch of the part that I have been reffering to. I will try to get a picture of the actual failed part this afternoon.

Attached is the end of the roll. It broke at the very end diameter to the next step (at the 3/8" fillet). The overall dimension of the shaft is 137.125" and acutally steps up a few more times in diameter. The part is symmetrical as well.

We have used these many many times. The difference this time was there shafts were regrinded for smoothness. The blocks that were welded on at the end where to mark which side of the roll for the dolly to pick the shaft up on.

The way they pick these shafts up are in two. They use a coupling to attach to shafts by end and lift by the middle (both side balance out). When they lifted this one it sheared right off.

Ill get picutres of the carnage today. I'm not to the point yet in my education career to know failure analysis (i'm a co-op) so I'm hoping you might be able to learn me

Thanks again guys.

 

[DWLetourneau]

Here are the pictures of the part

http://files.engineering.com/getfil...2-4e4b-8f7f-1f73baea99c9&file=Picture_001.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_002.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_003.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_004.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_005.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_006.jpg

http://files.engineering.com/getfil...8-4664-98ab-90160cbe5f18&file=Picture_007.jpg

 

[rb1957]

only pic 6 came across ...
you say they couple two of these 138" ingots of steel (7" dia) together and lift ... does the coupling reach onto the full diameter of the bar, or only the reduced rod end ?

if it's just the rod end, then there's 130" x 50in2 of steel ... 6500in3 *.3 = 1 ton
moment something like 2200*65 = 143000 inlbs
sounds like s lot (but then its a big rod end).

if your place does this often, then something went wrong with this one. maybe it was the grinding (unlikely IMHO), maybe it was the lift (sudden loading ?) ... maybe things were unbalanced ???

 

[NickE]

I only can see files 1 & 2... There are features on the fracture surface that appear to be common to ductile failure... however it doesnt appear that the part bent before breaking, at all.... Weird...

 

[hydtools]

At first look, the fracture runs close to the end of one of the welds. Nasty welding. In picture 001, there appears to be a heat affected zone that is cresent shaped and may have been the start of the fracture.

I would not expect welding would be allowed on a heattreated part. That would create a discontinuity or crack in the case. Were the blocks welded on differently than had been done in the past?

Ted

 

[DWLetourneau]

Yes these blocks were welded to let the dolly drivers know what side to pick the shafts up on. There may have been other reasons that are unknown to me.

It looked like to me as well that part of the fracture happened right around the penetration of the weld. I thought the heat from the weld might have caused a problem in the material (being that it was already heat treated).

 

[Pud]

If the 3/8" fillet is produced by grinding, it may have had grinding cracks produced in it, although this would be evident as an early fatigue failure.


Cheers


Harry

 

[rb1957]

how does the coupling pick up the two ingots ?

 

[stodge45]

If you examine the broken faces you can usually tell if a crack already existed. The "old" crack usually appears darker than the area that has finally broken. If the shaft was long and picked up with the original crack at the bottom it may have broken under its own weight.

 

[asishonline2000]

i guess your load condition is the torque type.The normal failure in toque is due to section variation.anyways if you are using any kind of stress analysis software failure should be detectible.

 

[esqrkim]

Either the two pictures of the broken end piece are not the same piece, or there are two welded-on blocks on both sides of the broken end piece. I thought only one block was welded onto the end piece.

In any case, I think the end piece broke due to a hair line crack. The crack most likely occured during the heat treating process. Probably caused by thermo shock due to improper temperature control. I do not think the welding had anything to do with it. The heat from the weld is at most only local. Yes, the welding job looks terrible, but it's not an extensive weld.

Eugene Kim

http://www.eeksolutions.com

 

[minerk]

Looking at picture 6, which is the only picture that clearly shows the fracture face, it's clear that this is a fatigue failure. The "beach" marks evident in picture 6 leave no doubt that this part failed in operation.

The end of the shaft finally breaking off during handling is largely coincidence, not the root cause of failure. I'd say it's a possibility that the weld in pic 1 may have been about the only thing holding the shaft together.

Without more good pictures of the parts it is difficult to provide much information. It does appear that there is also a stress concentration issue, but from what I can see in the pictures provided it looks like rotational bending was a larger issue.