NDE of welds between cast parts

[trainguy]

All,

For a rail vehicle truck (or bogie) application, what is the typical method of ensuring integrity of welds between cast steel parts?

Is radiographic inspection absolutely necessary, or is ultrasonic testing appropriate?

Are there drawbacks to either approach?

Thanks in advance.

tg

 

[metengr]

trainguy;
Here is some advice. IF you have no specific guidance from another applicable specification or standard, I always look at the function of the weld and what failure mechanisms could initiate from weld defects in service.

With that in mind, I believe surface nondestructive testing, like wet fluorescent magnetic particle or liquid penetrant is a minimum requirement. If the weld is to ensure load transfer or to carry equal strength between members (partial or full penetration welds), volumetric examination should be specified.

Methods of volumetric examination are either radiographic testing (RT) or ultrasonic examination (UT). Again, looking at what orientation of harmful weld defects or defects in the weld region could result in failure, this would dictate the choice of volumetric examination. UT is typically a more sensitive examination method. But more sensitive is not the correct approach.

You would need to provide more information regarding the weld detail and function in service before I could recommend a particular volumetric examination.

 

[dhengr]

Trainguy:
Where and what are you welding? Except for some repair welding aren’t std. side frames and bolsters cast in one piece these days. I agree with Metengr we would have to know more about the what and where and the stresses in the region of the welds to have much of a meaningful discussion. I would be more concerned about welds in tension regions of the member than any welds in compression or simple shear. Why don’t you ask this same question of your truck suppliers, or any other casting suppliers for that matter? They are doing this every day, and they are guaranteeing their product under a pretty severe environment, so they sure ought to know. Also, for the most part, the welding they do is at the same detail, on the same component, thousands of times per year, and usually in a less sensitive area on the component. I don’t know this for a fact, but some castings may be too rough on the surface to get good coupling when using UT inspection methods. I can’t recall that I every used UT on a sand casting. Welding of wear components sometimes does crack, due to the hardness of the wear plate material, etc. This is generally not to serious as long as the cracked weld holds the wear piece captive in service. The only thing I remember a lot of welding on was span bolsters and other very large castings which couldn’t be cast in one piece. However, I have designed and fabricated some components to be used in place of castings.

 

[MintJulep]

Typically a process control approach is taken.

Qualified procedure for the weld, supported by destructive tests.

RT or UT on the first couple of pieces produced.

The occasional destructive test of sample welds of representative test coupons to keep the welders honest.

RT or UT on a sampling basis for continued production.

 

[racookpe1978]

Look at your penalty for failure: ANY crash is bad, but a liquid phosphorous container is "more dangerous" than a carload of scrap metal or a carload of new Chevy's .....

 

[Duwe6]

If it was me, I would MT 100% of the welds, and initially perform about 10% volumetric exam [UT is my preference, due to its ability to find Lack-of-Fusion] As my confidence cincreased on the welds, I would decrease the UT percentage to about 2-5%.

 

[dbooker630]

If loss of vehicle control is the end result if your weld fractures, then I would 100% UT in addition to MT.

 

[kingnero]

Have you got access to EN 15085 ?