Prime Example of Fatigue Failure...
 

Another text book example of a type of failure - fatigue. I took this pic last week with my cell phone. A snapped shaft due to fatigue.

Fatigue is the progressive damage that occurs when a material is subjected to cyclic loading. The maximum stress applied is less than the stress limit of the material yet the material is subjected to a constant or random elongation, bending or compression which will eventually take its toll and produce a failure. Notice the twist patterns as well as the color differences on the below pic:



http://forums.pelicanparts.com/uploa...1274116838.jpg




Typically, a fractured piece will have a dark area which is an indication of slow crack growth, and a bright area which is an indication of sudden fracture. Fatigue begins with the crystalline structure experiencing dislocation. Eventually small micro-fissures develop which begin to compromise the physical integrity of the part, tool or material. Many materials will not normally recover if fatigue is relieved.

So, answer me this students-o-failure, why would the color be dark on the inside and not on the outside? How do you think the stress developed?

You have it back wards. The dark area is the area of sudden rupture while the brightly colored area is or slow crack growth. The crack initiation was at key-way and progresses inward as indicated by the "beach" marks. the dark area is of dimple rupture where the crossectional area was no longer large enough to handle the load and therefore the material Yield and UTS were exceeded.

I'm guessing that the color change ordinarily is a result of fretting and the color is from oxidized (very small) particles of the parent metal. Did this shaft stay together after the break?

The failure looks to have originated at the keyway. Was there a scratch or gouge in the bottom? I think I see one in the picture. I also see cracks starting to form at the corners. Was the key a bad fit in the slot?

JR

A dark area from slow crack growth could be from oxidation on the cracked surface along with the two opposite fracture surfaces fretting together and wearing away the beach marks, but that's not the case here.

I'd say the fatigue crack in this shaft grew pretty fast as evidenced by the relatively wide spacing on the beach marks by the key.

I'm not an engineer (wish I was tho) and I didn't sleep at a Holiday Inn, so my ignorant guess (in euber layman's terminology) is the dark color is from heat as well as a collapse or degradation of the molecular structure of the metal (tearing it self apart and compacting on its self) as it fractures?

The brighter outer area is a faster fracture and experienced less compressed energy when fracturing?

pretty lame right?

Cool pic!

Here is one of my 944 axle after suffering a failure:

http://forums.pelicanparts.com/uploa...1274119121.jpg

Looks like a cat - and why is it smiling at me?

I thought it was a breast implant that had been USED too much.


I have been in OT for way too long.

You can't assume that the material was isotropic when made. (cold-drawn?) And, strain-hardening (outer perimeter) can come from heavy flexing.

So, in engineering terms; Like a Ho-Ho ... hard chocolate outside, and a creamy/cakey filling inside.

I understand.

The highest stresses on a shaft in torsion are at the surface, and the crack would start there if there was no keyway. Except, there is a keyway and it has sharp edges where it was machined into the shaft. The the sharp edges of the root of the keyway caused localized stresses high enough to cause a fatigue crack to initiate. If it had grown slowly the first part of the crack would have darkened by oxidation, but this one grew fast enough for the fretting (rubbing of the sides of the broken crack together) to keep it polished. The dark area on the inside of the shaft is where the remaining material was stressed beyond the point of ultimate failure and came apart. It appears dark because it is a rough surface. This might have been avoided if the corners in the root of the keyway had been radiused.

That's not a torsion failure. . . more like reverse-bending fatigue. That 944 shaft, OTOH...

oh, and keep in mind that the key (in the keyway) likely provided support. ..it's not always so simple to point to corners as stress-risers. YMMV

LOL, I have three pieces just like that on top of my file cabinets in my office.

With just looking at the picture I don't see a stress riser or significant material flaw other than the keyway.

My guess is it endured too much radial load from misalignment causing it to bend at ever revolution which work hardened it until it became too brittle and sheared. Have you run any hardness tests on it yet?

We need to know more about the material. Was it heat treated, and if so, was it a through-hardening steel or was it only surface hardened? It has discoloration of a pretty uniform depth around the diameter, which appears to follow the contour of the keyway as well, indicating it may only be surface hardened.

The keyway itself does not appear to be an issue. The bottom corners are well radiused for the size of the keyway. I only see one little crack propagating from the keyway, from the left side, and it appears as though it way have been a result, not the cause, of the failure.

My guess based only on appearance is is 4140 centerless ground shaft quality. Surface treated only.

hint...

http://forums.pelicanparts.com/uploa...1274154344.jpg

The dark portion of the fatigue fracture surface exhibits classic “beach marks”, “thumbnail marks”, “stop marks” and are clearly present in the darker portion of the fracture surface, the light portion of the fracture surface is where "fast fracture" (final separation) occurred. A common rule is Bright = Brittle fracture. Dark = Fatigue fracture.

Mike... Don't take this the wrong way but.... You wear a Pocket Protector don't you?

He keeps his pocket protector in a pocket protector.

He has a good pocket protector he wears to church.

and a leopard print one I wear when I'm feeling frisky!

Materials was one of the more interesting classes I remember from college, too bad I didn't finish it (I don't think)

Looks like E-3 to me :D

LOL many moons ago I was going back to college for about the gazillionth time. I had requested and recieved a copy of my transcripts and it said I took an introductory and advanced metallurgy class at Fullerton college in 1980 and got A's in both classes. Funny thing was, I don't remember taking either one of those classes. I don't remember drinking that much back then either ;)

Opening graphic
 

This is a bit off topic but I have reccently written a maintenance learing module that covers among other things fatigue cracking. Would it be possible to have copyright permission to use the photo in the opening post please. Many thanks

Martin

Looks to me like there was some inclusion in the base metal. Center of picture, and there appears to be a "path" leading to it from somewhere around the 5:00 position in the photo.

A crack can only propogate if there is a void. Usually they start on the outside, unless there was a flaw in the material such as this example has.

We had a very important shaft fail in my first job. Caused a paper machine roll to break under operation. Similar material. About 12" diameter. Under a LOT of tensile pre-load to counteract bending of the body of the roll. Long story short - lots of pieces of roll body material throughout the rest of the machine. Machine down for about 3 weeks to repair damage. Root cause of failure was a similar inclusion - however, ours was on the outside of the shaft.

send me a PM and we can work sump'n out! ;)

Mike .. could you possibly tell us the dimensions, application, and an opinion of this shafts material? It just seems like quite a large keyway on a relatively small diameter shaft.

I have to wonder whether core strength was compromised in just the short keyway portion of the shaft during the heat treatment process, due to the irregular cross sectional shape. That's kinda what the photo looks like to me .. a large hard zone around the keyway, stealing uniformity from a core that's skewed off-center. Copper coating the keyway prior to heat treatment is a probable solution .. or replacing the key with splines. .. (just a layman rambling) ..

In 1987, I saw guy wearing a pocket protector and thought it was pretty neat! Then I got better...

We ever going to learn all about this failure?

It's been 4 1/2 months and we're not going to get any smarter, so new guesses may not be forthcoming.

JR

Thats an easy one, Lube.

You see its all to do with (doing my best Rowan Atkinson impression here) electro-magnetic orientation of the adjacent strata of atoms in relation to the inclination degree perpendicular to the direction of the force.

In other words, it went tits up

Well.. yes. :D

Opening picture enquiry
 

You can email me at martin.glaeser@openpolytechnic.ac.nz

Cheers

You must really love getting spammed to post your email adres in a public forum.

Using a keyway like that to transmit torque is a big no-no under most circumstances.
An interference fit between the hub and the shaft would be optimum so the keyway would not see any shearing forces.
Also I'm guessing the keyway was cut with a standard end mill, leaving sharp edges. Stress riser city.

Design the hub with a taper lock or standard interference fit of .001" per inch of dia. up to 2", then additional .0005" per inch after.
If it's belt driven, make sure proper tension is maintained. Too loose can do that more than too tight, snapping a string like.
Have em cut the keyway in a way that eliminates the sharp edges.

If it's direct drive tell em to use a good laser alignment system to get the alignment just right to reduce vibration.

Don't have enough info to recommend a material upgrade.

Photo request
 

Hi, any chance of a reply to the request to use the photo? We could probably buy a copy of your book.

Tits up in the ditch!