Rail crack

[JHWC]

Hi Experts,

Would like to seek your opinion on the rail crack (see attached). Please advise.

1. Is that consider a beach mark around the circular patch? To me, it seem more like an oxidation rather than beach mark.

2. Consider thermite welding was performed to this rail 7 years ago, how will it affect the surrounding material? I noticed inclusion in the middle of the rail. Such failure could be due to the process but I am surprised it actually survived for 7 years w/o failing.

3. What are some of the possible defects that I have missed out? I saw ratchet marks.

Thanks all.

 

[btrueblood]

Is the thermite weld actually the thicker section behind the crack face shown? Is the crack thus in the HAZ from the weld?

 

[JHWC]

Please see attached. You can see the thermite welding at B side. Thanks.

 

The round dark patch is both the extent of cracking (up to catastrophic failure) and oxidation.
The shape is highly suggestive of a fatigue crack radiating out from the surface (or a discontinuity near the surface). The centre of the semi-circle will be your crack origin.
The surface looks degraded, so ratchet marks may have been weathered away - the dark colour was caused by exposure to the environment. Microscopic striations will be long gone.
Yes an internal inclusion or weld discontinuity could have been the crack initiation point. Ratchet marks are a hallmark of fatigue, although they are not necessarily always present.
I think you have pretty much already solved this one.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[JHWC]

Haha. Thanks!

 

I fired my response before you posted your second photo, which confirms the weld was the culprit. Serious stress raisers are visible.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[kingnero]

How long after the rail break was this picture taken?
After having seen hundred's of rail breaks, the top half of the rail looks very familiar (fatigue cracking, followed by brittle fracture). Picture is not clear enough to comment about the bottom half.
Initiation point indicates failure due to geometry. Is it possible to get a side view, looking at the right side of that picture?

Rail breaks due to thermite welds are very seldom due to the rather good process control (talking about Europe). If it is linked to the weld, most likely culprit is too high hardness, caused by not enough (or incorrect) preheating. Processes like SoW-5 and PLA small gap welds require > 600°C preheat in the joint. Older processes use air-fuel gas preheating, more recent (after year 2000) all used oxy-fuel gas for this reason.

Inclusions can come from the mould: when using excess oxygen in the preheating phase the moulds deteriorate and spall. Particles can ambed in the liquid metal. Other possibility is from the luting sand (looking at the picture, I believe sand instead of paste was used). Incorrect closing of the gap between the moulds and the rail (perhaps due to different vertical rail wear) can also cause inclusions.

 

[JHWC]

I believe several combination may cause the rail crack. Welding is definitely one of them. But i am surprised that it can last for 7 years. Wow. Amazing.

 

[kingnero]

Different posts were added while I was typing, but can you confirm both rails have different height? It does look so on the second picture you posted.

 

[JHWC]

B side. Photo 1

 

[kingnero]

I still see rail breaks of welds performed late 1990's today. Due to the markings (this is now required by the EN 14730-2 standard), we are able to identify the weld type (including preheating). The tonnage that these welds have seen is astonishing.

 

[JHWC]

B side - 2

 

[JHWC]

If you look closer, there is a weld dip at the top.

 

[kingnero]

Always, due to the lower hardness in the HAZ.
If you look close enough, you'll actually see two dips, both sides of the weld.

 

[JHWC]

Hi Kingnero,

Could you explain a bit further? What do you mean by always? And where are the 2 dips that you are referring? Thankss..

 

[kingnero]

All thermite welds have rather large HAZ's (up to 20 mm) both sides of the weld. The thermite charge (= filler metal) is designed to have a similar hardness as the rail grade (R260, R350, ... meaning at least 260, 350, ... HB), however in the HAZ you'll lose about 30-50 points HB. On these locations you have accelerated vertical wear due to lower hardness. This causes the dips you'll always have (after some time in track).

 

[JHWC]

Thanks!

 

[kingnero]

Attached a screenshot from one of my presentations about thermite welds.
This shows the values on a R260 grade rail (meaning hardness tolerance is 260 - 300 HB, with almost always >= 280-285 HB as delivered).
You'll clearly see that the filler is as hard as the base material, with two dips in hardness in the HAZ.

 

[mrfailure]

First, I would look closely under a microscope at the origin at the head-web corner on both sides to make sure you are seeing fracture as opposed to lack of fusion. Then I would prepare a longitudinal cross-section through both sides and look at structure as-etched to see where the crack started (i.e. in the weld, in the HAZ, or at the weld-rail interface). You may well see the crack starting at an inclusion in the unetched condition, but remember fractures have to start at some area of weakness. General microcleanliness in the unetched condition is more of an indicator with regard to this. Finally, as also suggest, take a microhardness profile across both sides of the fracture. Good luck!

 

[Compositepro]

Is it common that thermite welds add so much thickness? The added thickness alone creates a significant stress riser due to the change of stiffness of the rail.