Heat exchangers U-tube - Tube failure

[Eduardo1982]

Dear guys.
Does anyone have experience in tube failure in heat exchangers type "U"?
Actually we have had a failure in a heat exchanger and we foudn that 1 tube was broken.

We are recovering all data from previous to event, however I dont lnow if you guys could recognize the failure with picture attached.

 

[r6155]

More information,please: material, diameter / thickness, design pressure & temperature,.......etc.

Regards

 

[EdStainless]

Where in the bundle is this? We need it located in e dimensions to be of any help.
Along with all of the other info that is missing.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Eduardo1982]

Tube material
Material: SS 347H.
Diameter: 25.44 mm.
Thickness: 2.032 mm.
Pressure (shell side):25 BAR.
Tube side: 155 BAR.

Temperature (shell side): 565 °C.
Tube side: 370 °C.

Fabricated under ASME VIII DIV.2

 

[springrise]

This looks like a "fish mouth failure" which is often attributed to the local overheating of the pipe. However, there are other possible reasons that might have caused this type of damage, e.g. local thinning due to erosion-corrosion and subsequent ripping of the pipe.

 

[r6155]

If there is only one tube, I'm guessing it was a defect that went undetected during manufacturing.

Regards

 

[GD2]

What row of tube it happened. My take will be one of the inside rows. Go back and check the purchase order condition for the tube? Was heat treatment required after the u-bend, specially for inside rows?

Even though 347H is stabilized SS, the operating temperature of 370C still make it to run in the sensitized zone where some Chromium can still precipitate making it susceptible to corrosion.

The second level of defense would have been to heat treat the tubes, atleast the inside rows, where there is high possibility to induce embrittlement and susceptibility to stress corrosion due to cold work of U-tube forming.

By the way what are the tube and shell side processes? Is this just one tube that failed? Is there any way you can check the thickness around the bend in the nearby tubes?

GDD
Canada

 

[Eduardo1982]

Well.. the tube broken was found in almost the middle of exchanger.

Broken part was in straight side (not curve part).
"U" tubes were COLD BEND (was indicated in mill test certificate).
In a note (of mill test certificate) it is indicated that all U tubes are solution annealing at the U bend portion plus 300 mm of straight leg at min. 1110°C and 1235°C for 10 sec. And cooling by force air.

 

Have a failure analyst with some experience with process equipment perform a proper investigation.

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

 

[GD2]

Eduardo,
Surprising. My comment will be as below:
If its a one tube failure, just forget it by considering it was a bad tube and one off incident.
If on multiple tubes - is the rupture on the transition zone of the heat treatment?

GDD
Canada

 

[Eduardo1982]

GD2.. No.. it was almost in the middle of tube length.
with the shape of failure (this "fish mouth" shape) that a kind guy said before, it is not possible to determine a possible source? I was searching into literature and short overheating seems like it had happened.

 

[EdStainless]

Manufacturing defect is my guess as well.
I don't care if it is welded or seamless, I have seen this in both.
Plug it and move on.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[chicopee]

A metallurgist will tell you more about the problem if you send that person samples of the failed sections for analysis otherwise it is just a guessing game. You may also want to review past literature from the National Board as it maintains good documentation of past failures related to NB/ASME pressure vessels assuming that your pressure vessel has a documentation submitted to the National board and has a NB/ASME number.

 

[SnTMan]

Likely impossible to get a sample of that tube without removing adjacent (outer row) tubes to gain access.

I don't care if it is welded or seamless, I have seen this in both.

I've seen a seamless, stainless HX tube blow up on a 300 psi hydro. It happens...

Regards,

Mike




The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[weldstan]

As Ed stated, "Plug it and move on."

 

[dik]

CTRL-F located three instances of 'guess' or 'guessing' on this page.

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

 

[Christine74]

LOL, you get what you pay for.


-Christine

 

[EdStainless]

Until they go in with a deep cutter and slice that tube off just beyond the rupture and then pull the stub out and take it to a lab, guesses are all that they have.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[dik]

Four.

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