Boiler tube crack 1

[Gregg B]

Hi, the company I work for recently purchased a hot water boiler. Water flows through formed metal tubes while a burner provides heat.
The boiler installer filled it with water, performed a pressure test, and identified a leak before turning on the burner or main pump. One of the tubes has a crack around a flange. I don’t know how the flanges are formed but I have included pictures of a new tube end. The end is tapered and has a flange that is used to install the tubes. The tapered end is driven into a hole in the header of the boiler. Then a plate is bolted down and presses on the flange. I am told that this plate is only a mandated safety measure. Driving the tapered tube into the hole is sufficient to form a pressure-tight seal. I don't know the force of the plate on the flange.

The boiler company claims they pressure tested the boiler to 240 psi. We only tried to pressure test to 40 psi, but couldn’t achieve that pressure due to the leak.
The tube material as written on the tubes is “WEBCO A/SA 178-A/214 ERW 1.500 OD X .095 MW HT# GC77119”
Here is the Webco and ASTM specification:

http://www.webcotube.com/solutions-resources/logistics/spec-listings

https://www.astm.org/Standards/A178.htm

Here is a video of the boiler manufacturer’s tube installation process:

https://we.tl/yZOM86J6Dh

There are indentations on the flange of every tube, presumably from the installation tool.
What do you think caused the crack? Obviously this is a highly stressed area due to the forming of the flange and the hammering during installation. Why would a crack develop spontaneously when no pressure is inside the tube? (i.e. I’m assuming the crack developed during shipping) I’m interested to hear your thoughts. Thanks!


New tube end

New tube end

New tube end

New tube end

New tube end

Wide angle of cracked tube in boiler (cracked tube circled in red)

Leaking tube

Crack

Crack

 

[mrfailure]

I hope this is under warranty. Based on the last picture, it looks like a fabrication error associated with welding of the tubes to the flange. That should have been identified when the manufacture conducted the pressure test.

 

[Gregg B]

Yes, the tube is going to be replaced under warranty. There is no welding, though. The flange is formed from the tube material.

Yes, the cracked tube should have been identified during the manufacturer's pressure test. They test it the same way we do. They fill it with water and would have seen the water pouring out. I have a pressure test results page signed by two inspectors showing that this unit was pressure tested. The employees were interviewed and remember this particular unit being tested. This means that the crack formed after they tested it. That's the part I don't understand. Why would it have cracked when there was no pressure applied?

 

[EdStainless]

I know these tubes and this forming. They messed up the form.
There must have been a small crack at forming that then propagated in transit.

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

 

[metengr]

I don't buy the cracking in transient based on configuration. The pressure test was either missed or QC did not do their job. I can tell you this we had a batch of heat exchanger tubes supplied by a mill that contained defects. The supplier determined that their great QC flagged the tubes and instead sent us the boxes. Trust but verify.

Getting back to your original question. You need to review the forming process to better understand if something changed or one or more defective tubes slipped through. This could be a one and done deal or other tubes may contain defects which have not opened as of yet. The pressure test is not a guarantee of future performance only there are no gross defects. As a minimum I would have several of these tube ends removed and checked before replacement of the one tube leak.

 

[RolMec]

Fully agree with metengr.
To achieve external and internal collar radii that small in the forming process the pipes had surely to be heated (they're blued due to this, correct?), and clamped very strongly. I'd presume a manacle like device with a kind of spikes on the inside to bite into the pipe material. But even so, perhaps, the pipe slips a bit during application of forming force, thus the "hammering marks" on the non-finished side near to the bend.
But however clamped or not or hammered or not, I'd assume rather important forming forces being applied. Now, you name welded pipes. This makes an elongated imperfection in longitudinal direction, overlaid with a strongly deformed zone across the section. So, lots of possible root causes (deformation vs. wall thickness + forming temperature appropriate to material, local weld imperfection, hardened region around weld vs. required formability) but if looking for the start of the fracture i'd look here:

imo the crack was there before pressure was applied, the manufacturing process (& of course QC procedure ;-) is to be reviewed.
Regards



Roland Heilmann
Lpz FRG

 

[SMF1964]

Could this be a strain-age embrittlement from the forming process of the flange if the forming process was done cold?

 

[metengr]

You would have to perform a proper failure analysis to confirm this. I can think of several possibilities if you're guessing.

 

[Guest102023]

Agree with metengr, but it sure looks like the forming was serially messed up.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"