Is this failure by shear or fatigue? 6

[Naruwan]

Dear all,
I had just received a photo of the failed shaft's fractured surface and I was wondering if this fracture surface is caused by shear or fatigue. I had asked our mentor regarding this and he claimed that it was caused by fatigue due to the fact that there were ratchet marks at the left side of the shaft. Seriously, I am not convinced.

 

[Naruwan]

unclesyd, I don't have sufficient info regarding to the state of the rudder when it failed. They just mentioned that the rudder is at the mid-ship position. Thanks for the tip, I think I need to ask them more regarding that!

I had added in another close-up picture of a portion of the fractured surface. I am quite doubtful that that was caused by flame cutting. Ofcourse, I know to confirm it, I need to look at it's microstructure.

 

[unclesyd]

I think you are correct in you assumption. It is still weird as it is 180 degrees from the ratchet marks in the first picture.
A note on the rubber failing in the amidships position. A well designed rudder will go to this position in any failure that frees the rudder. The degree of motion, 90.120.etc, of this rubber might help explain some of the features

Is the line starting from about 11:00 position in the last picture a greasw streak or is it grease emanating from a crack?

Another interesting features the series of parallel lines in the middle of your last picture. I've seen this on some lower strength steel that later turned out to features due to very high stress in will become crack origins. During MT inspection these are normally ignored as they are considered ghost lines.

 

[redpicker]

Naruwan,

I believe you are correct, there is no evidence of flame cutting. What appeared to have been a flame-cut region in the orginal photo is clearly ratchet marks from a crack that initiated approximately 180 degrees away from the major crack. This was probably a result of the increased bending the shaft saw in this area because of the reduced cross sectional area.

It would be nice to see some pictures after the grease has been cleaned. I think beach marks would be clear, then. Also, if you have the chance, a wet magnetic particle inspection of the OD would also be illustrative because I suspect it would reveal many more cracks. The ratchet marks indicate multiple crack initiations that join as the crack fronts propagate towards the center. THe large number of ratchet marks suggests there could be many other cracks that were initiated, but did not join the main group of cracks that eventually joined and propagated towards the center. The number and location of these secondary cracks could reveal more information to the nature of the loading conditions that led to the cracking.

rp

 

[desertfox]

Hi Naruwan

Look at this site it might help:-

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

desertfox

 

[Educonmet]

This is a typical example of a rotating shaft fatigue failure. Form just this picture a failure analyst can tell the origin of the fracture (at one corner of the key hole ?), overload region (diametrically opposite end), and even the direction of the shaft rotation. Very similar examples are available elsewhere in text books and failure analysis courses. The beach marks are clearly visible and the rotation of the shaft makes them appear the way they are. Good post.

 

[Naruwan]

Hi Guys! I'm back with new photos regarding to this case that I had started earlier on. I had cleaned up the fractured surface with acetone and hey presto! I found beach marks! As you can see, there are ratchet marks laying on both sides of the shaft and they are exactly opposite to each other, just that one is larger than the other one. Is it true that the formation of ratchet marks normally will result in beach marks? In this case, the smaller ratchet marks were not able to form beach marks because the final rupture was to fast for the beach marks to be form? I remembered Redpicker mentioned that this shaft had been subjected to bending and rotational stress. The beachmarks formed like 30% of the whole shaft, I reckon that this shaft had been in high stress with low frequency type of fatigue. I was still quite unsure of the crack initiating point. Was it started at the 6 o'clock position (based on the photo) or was it started where the big & long ratchet marks are pointing (which is roughly at the 8 to 9 o'clock position)?

Thanks!

 

[Naruwan]

I had also attached a side view of the fracture surface. The rest of the "unaffected" part of the shaft was being removed. A crack seen running diagonally could be formed during the final rupture and ripped off a part of the material.

 

[Naruwan]

This is another side view of another side of the shaft. Maybe you are able to notice that a clean and straight fractured end was running along the circumference of the shaft which ended the fracturing. Was it due to the brittleness of the material? (I'm yet to obtained the shaft's mill certificate)

Thanks!

 

[Naruwan]

Lastly, I had attached a closeup photo of the ratchet marks with beach marks. According to Redpicker, a wet magnetic particle test might be able to reveal more cracks, what about PT (penetrant test)? Will it be helpful too?

Thanks again!

 

[desertfox]

Hi Naruwan

Thanks for the photo's.

Well the pics show the Instaneous Zone to be quite large and therefore the shaft was heavily loaded.
The side view showing the crack running diagonal is an indication of principle tensile stresses present in the shaft whilst being loaded in torsion.
I am guessing but I think the crack started at 6 o'clock looking at the first of your latest post because the river lines at that point tend to curve to the outer surface but others might no better, however its clear the crack started somewhere on the left of the shaft and grew across to the right as we look at the picture.
According to the reference I left in my earlier post it says if the dividing line doesn't clearly indicate the crack start point then rotational bending was involved.

desertfox

 

[redpicker]

Naruwan,
Things are a lot clearer after cleaning. There is no doubt about it being fatigue, now.

The mag particle test I mentioned would be using a longitudinal field to detect circumfrential cracks on the OD. I would expect them to be short indications (about the length of a single ratchet mark, or about 1 - 4 mm {maybe 1/32"-3/16"}) so wet flourescent mag particle may be needed. Dye Penetrant testing may reveal them, but it is very messy to deal with (that dye gets everywhere) and its sensitivity is a bit low, so it may not detect them.

I put the origin at around the 8 o'clock position in your first photo. Note that directly opposite this is where the crack comming from the other side is the deepest (the initiation point from the other side). This suggests bending of the shaft from side-to-side, with the stresses on one side of the bend being greater than the other (probably because the support points are at different locations). This would lead me to suspect the bushings/bearings that are supporting the shaft are a bit worn so one side is supported better than the other. That's just a WAG, though, since I don't know anything about how this shaft is supported.

Thanks, Naruwan, this has been interesting.

rp

 

[unclesyd]

Another point to consider as rubber problems were mentioned in the OP is to look at the rudder design. If a rudder doesn't have the proper leasing edge,the portion in front of the ruder shaft, will impose much higher stresses in the rudder shaft. This condition will cause the rudder to seek and shaft problems on hard turns.
I would look at the rudder design.

Sorry for the lack of proper nomenclature, it been too long away from the boats.

 

[Naruwan]

unclesyd,
I had managed to get a copy of the schematic drawing of the stock rudder with the bearing only. In this attachment, the first drawing shows the original design of the stock rudder. They re-designed it and came up with a thicker stock rudder with a bigger bearing to accommodate the new stock rudder.

Hope this helps...

 

[Naruwan]

Hi guys, I managed to find some information regarding the ratchet marks. Attached is a picture (refer to Fig. 4) which described the appearance of the ratchet marks when they are view cross sectionally.

 

[Naruwan]

This is my cross sectioned sample which was cut across the ratchet marks. Comparing to Fig 4. in my previous posting, I really can't tell whether it is caused by (fatigue) with bending or (fatigue) with torsion. Can anyone enlighten me?

Thanks!

 

[Naruwan]

Hi, dear all,
this is a photo micrograph (at 100X) of the microstructure of the broken rudder stock which we are discussing earlier. You might be able to see that there are sulphide stringers embedded in the microstructure. The stringers aligned themselves in a way that they are along the longitudinal direction as the rudder stock. Apparently, what I found out that the material grade is 16Mn which happened to be a Chinese grade. Looking up the net, I found the equivalent grade to it will be ASTM A36. The stringers are everywhere, which meant that if I were to perform a ASTM E45 test on it, this piece of steel will sure fail! I am not quite sure whether the sulphide stringers maybe another factor which contribute to the failure of the rudder stock. The microstructure looked strange in the sense that the pearlite are very angular and sharp in shape. The grain size also seemed to be quite large which having a ASTM grain size number 6.0.

Is anyone able to advise me on this?

Thanks again!