Weld issue - high carbon content discovered , supposedly 1018

[Scott.C]

New to the forum here, looks like an incredible source of knowledge in these groups and glad to meet you all.

We have a small bracket that is failing, and in a cross section I noticed what I would describe as a brittle looking crack in the weld. Upon sectioning more parts, and doing an SEM analysis, we are seeing a very high carbon content in the base metals, which are "supposed" to be 1018. We are seeing carbon in the 3 - 4% range! Presently awaiting material certs from the supplier, if they even have them (we haven't required certs). Any thoughts on why we could see THIS high of a carbon content? Maybe they are using a high carbon steel for some reason? It is a flat 1/8 plate, with a ~1/4 pin welded on the back side of a through hole in the plate.

Thanks for any advice
Scott

 

[EdStainless]

Over 2% is cast iron with free graphite in it.
I wouldn't trust EDX to do quantitative analysis unless you have someone who is very well versed.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[AidanMc]

In my experience, energy dispersive spectroscopy is not a reliable method of determining carbon content in steels. EDS isn't usually great for lighter elements. There are some technical challenges to using it for lighter elements and the peaks tend to get conflated with other elements when checking a spectrum.

Aidan McAllister
Metallurgical Engineer

 

[Tmoose]

" a brittle looking crack in the weld."

http://www.lincolnelectric.com/en-us/support/process-and-theory/Documents/weldcracking.pdf

http://www.arcweld.co.nz/Data/Default/Documents/Literature/WC515 Why welds crack.pdf

Got pictures ?

Got drawings ?

2 down, 18 to go.

 

[Scott.C]

Thanks for the great feedback! Steel analysis is not our usual task, so I will take that chemistry lightly.....it's just way too high to make sense.

I will add a couple pictures as well, these are longitudinal, as you will see in a minute......

 

[Scott.C]

Try again, adding photo link

 

[kingnero]

That looks like a geometry that's screaming for fatigue failure, depending on the acting forces, ofcourse.
Send it to a materials lab for full analysis, as this won't really help you, asking a bunch of internet bozos to do a failure analysis on a vague description ("brittle looking crack") and those pictures.

http://www.fusionpoint.be

http://be.linkedin.com/in/fusionpoint

 

[Scott.C]

Agreed, and I already have a new design in process that will be MUCH more robust. But we are just trying to get out of the emergency, as parts are failing nearly right out of the box. Agree with the Fatigue potential but the emergency at the moment is the infant failures. Thanks!

 

[weldstan]

Did you etch the surfaces in the photos and if so what etch did you use. Why the partial penetration weld? And what welding process was used?
Do check the chemistry to verify material.

 

[EdStainless]

If you want to grasp at straws I would try tempering/baking a few of these.
It could be something as simple as hydrogen embrittlement.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[desertfox]

Hi Scott

Can you provide more information on the bracket loading and orientation of the weld, also what is the welding procedure for this component? Is the bracket stress relieved post weld? How long do the brackets last in service? What NDT is performed on the weld.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[Tmoose]

Were these parts successfully welded before ?

Is the weld made in one continuous pass?
I'm guessing TIG
The etch makes it look like something between a fillet and plug weld.
What is the pin made of?

14/6