crack on a105 flanges 1

[gordonmech]

dear friends,
during hydrotest of 6"#600 pipes we found some cracks in the taper area of flanges, the cracks are far from weld seams, what is the main reason of these cracks, welding procedure or material construction (forging of flanges)? what kind of test is necessary to find out the reason?
Thank you.H?

 

[unclesyd]

There are several possibilities for checking for cracks of which there should be none.
If as you posted that they are visible then a quick test would be dye penetrate. The second option would be magnetic particle.

To help with your problem answers to the following questions would be nice.

What is the orientation of the cracks?

Are they visible before or after the hydro?

What is the country of origin for the flanges?

What was the test pressure and temperature?

 

[weeeds]

We experienced similar problems.
The taper was radiographed and found to be full of defects from welding. Instead of receiving a single piece forged flange we received a fabricated piece that was in fact a counterfit, and a very bad one at that.
Do a radiograph and see what is inside, you may be surprised.
Our product originally came from China.

 

[metengr]

Radiograph test (x-ray) the weld seams.

 

[ndeguy]

If the cracks are far from weld seams then the cause should not be related to welding metallurgy, e.g. embrittlment, high HAZ hardness.

As A105 is carbon steel MPI will reveal all surface-breaking cracks. Hardness and PMI (Positive Material Identification) in the vicinity of and distant from the cracks (for comparison) may give some indication of localised metallurgical anomalies.

Radiography should be a last resort for volumetric NDT considering both the hazards involved (radiation, chemicals) and its poor capability for randomly-oriented crack detection .

To perform volumetric NDT use encoded recordable UT (Phased Array and/or TOFD). It has superior detection capabilities than RT for randomly oriented cracks and planar flaws and a small quantity of water is the only consumable. A combination of TOFD and Phased Array will detect both embedded and surface flaws.

 

[stanweld]

One of the possible causes of cracking is forging from continuous cast reforging billets that had not received the proper ammount of hot work. I had two major incidents that were attributed to the condition of the reforging billets.

 

[moseley]

As everyone has stated above there could be many different causes for this cracking. My experience with issues exactly like this has been that the manufacturer's who provide these forgings do not follow proper preheating or weld technique when making forging repairs. As a welding specialist I have the unique opportunity to be able to review these welding procedures prior to use and have found the majority of them do not meet basic ASME requirements because no one has really ever asked them to review. We recently had approximately 200 A105 flanges and valve bodies that exhibited cracking through out. Again did not discover until Hydro-test. Further investigation found that the areas surrounding the cracks had embrittled tremendously and some even had been repaired with the wrong filler metals (CSEF type used). Anyway the real only way to prevent this is to review the manufacturer's WPS' and PQR's prior to permitting them to perform any sort of weld repairs! Hope this help.

 

[ndeguy]

OK, so though as per gordommech states "cracks are far from weld seams" then repairs (taper area) have been made by welding. Then the 1st step if weld repairs are not immediately visible in the region of the cracks would be an etch to reveal any repairs and HAZ and see if the cracks are in the location of the same.

Do ASTM forged flange specs allow for weld repairs?

 

[stanweld]

Section 10 of A-105 governs repair by welding. "Prior aproval of the purchaser is required to repair-weld special forgings made to the purchaser's requirements."

 

[Guest102023]

Hold on a sec.

Not every 'linear indication' is a crack. If remote from the weld it could be a forming defect such as a fold or lap.

In my past experience the 'root cause' of defects in fittings was often the name of the (3rd-world) country of origin. Despite the fact they all had ASME stamps.

 

[Guest102023]

p.s., perform some surface NDE (PT of MT).
You may be able to remove minor surface flaws with light grinding in areas with sufficient meat. You would rather not weld (with possibly mandatory PWHT), but that decision depends on the size of the fitting and the cost implications.

 

[gordonmech]

Hi all,
i performed tensile, hardness, impact, and chemical composition tests in a new flange with the same heat number, the results are enclosed.
as you see, the carbon equivalent is about 0.49.
in your opinion what is the main reason for the cracks.
it is very interesting that after ut test we saw some flaws in two area of the flanges(0 and 180 degree on the holes area).
Thank you.&

 

[salmon2]

Really my 2cents:
1) why the first set of tensile's elongation awfully lower than 2nd set?
2) the "crack" looks awfully like inclusion to me.

 

[gordonmech]

dear salmon2;
the first set of samples made from facing area and secnd one made from neck (taper) area of flange.
thank you

 

[Guest102023]

Polish and macroetch the region. Perhaps they weld repaired (poorly) without declaring it. Would not be the first time I have seen that.

 

[salmon2]

Maybe a dumb question: why face area differ so much from neck? I know the flange was forged to its dimension.

 

[gordonmech]

dear friends,
i don't know, but we found that all flanges with this heat number has a lot of flaws in the raised face area.