Failure of Air Preheater Tube in CFBC Boiler 4

[AKAK96]

Hello everyone!

I have been assigned the task of Failure Root Cause Analysis of Failure of Air Preheater Tubes of a Coal-based Circulating Fluidised Bed (CFBC Boiler). The details of the failure are as below. Input from the community will be highly appreciated.

- Air Preheater is of tubular type and consists of 3 Tube Banks (called modules). One of them is a single pass heat exchanger for secondary air and Two of these combined as double pass heat exchanger for primary air.
- Air Preheater with cracks leading to tube dislodging occurring in the vicinity of Tube-to-Tubesheet joint (opposite to the tube end just beside the tubesheet on fireside). Photos attached for clarity.
- The failure is brittle fracture with no sign of thickness loss in vicinity of failure location.
- Failure percentage is much higher for roller-expanded plus welded tubes in comparison with expanded-only tubes.
- Numerous case histories handbooks of boiler tube failures have been searched but none of the failures match our case.
- Sample for lab testing is not available (as boiler is currently operational), yet the failure root cause analysis has to be performed based on literature with supporting calculations (if required).
- Superheated steam soot blowing is used to clean fly ash from air-preheater tubes, which may be a cause contributing to failure.
- A possible cause of failure may be thermal expansion and contraction of tubes during soot blowing or during start-ups & shutdowns leading to Thermal Fatigue Failure at Tube-to-Tubesheet Joints.
- Thermal Stress Analysis for the APH has yet to be performed on ANSYS to confirm the above failure.
- Stress Corrosion Cracking (SCC) may also be a probable cause but we are unable to identify the corrosive media which may be assisting SCC.

Guidance is required based on above details on what might be the actual cause of failure which might be proved without lab testing of failure samples. Also, possible solution to the problem has to be decided based on the proved failure root cause.

Please do not hesitate to give your input whatever it might be.
Thanks!!

 

[FacEngrPE]

Please add some material information. I can infer from the presence of welded and non welded tubes that this air heater has been re-tubed previously. might be more than one tube material. Where exactly are the welds? Tube to tube sheet or somewhere else?

 

[SWComposites]

Sounds impossible without lab tests of the failed tubes.

Are the failed tubes still in use?

 

[AKAK96]

@FacEngrPE
The plant is relatively new. Installed 5 years ago and the possibility of reworked tubes is slim. The Expanded only and Expanded plus welded tubes are part of the OEM's design. Please view the drawings and material details in the file linked below for further clarity. Location marked with red arrows on joint detail is the failure location.

@SWComposites
Those tubes which are completely dislodged are not repaired, instead plugged. Tubes which have cracks only are repair welded and kept in service. Yes it seems almost impossible without lab tests but management wants it done this way (about which I cannot do anything).

 

[EdStainless]

You can make a list of possible modes and tell them what would be tested and how that would indicate the true RC.
And from there a fix.
Without testing it is all guessing.
If they want to just pick one off of your list and proceed, then that is their business.
Don't risk your reputation on making guesses for them.
That said, vibration and thermal fatigue would be my first two thoughts.
I am a bit incredulous that they can heavily expand a tube into a 0.125" plate and not expand any unsupported portion of the tube.
I'll wager that they over expanded metal beyond the plate.
One way to put these back into service is to install thin liners.
Get some steel tube that will just slide inside the heater tubes, 1/4-1/3 of the wall thickness.
Wire brush the insides clean.
Install a 12-16" long piece.
Start at the far end and lightly roll expand it just to contact.
Work you way to the tubesheet.
Since the liner will loaded in shear it doesn't need to be full thickness.

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

 

[FacEngrPE]

The data supplied rules out completely incorrect materials. Air heaters are abusive to the material they are made of. Had you indicated any of the material was 300 series stainless steel it would be obviously trouble.
My experience was with 1945 vintage Lungstrum heaters. These have mild steel plates in a rotating mass, service life between plate change outs on 3/4% sulfur No 6 oil was about 15 years, limited by corrosion. Now we burn natural gas in new boilers, the design does not use an air heater, so direct experience is missing. I think Ed's idea for sleeves has merit, but as long as the number of plugged tubes is small it will not have much impact on performance.
On root cause analysis, it is not possible to find a root cause until the failure mechanism is understood.

Suggestion - review some reports on these websites, to understand why root cause analysis needs to be methodical:

https://www.ntsb.gov/Pages/home.aspx

https://www.csb.gov/

https://www.bradyheywood.com.au/

Brady Heywood Podcast

 

[dik]

Those Brady Heywood podcasts are great. Thanks for the link.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Do you feel any better?

-Dik

 

[MJCronin]

I have never seen this type of sootblowing used on this easily damaged component before ...

Was the superheat-steam soot blowing of these Air Preheater Tubes a recommended practice by the Air Preheater vendor ?

..... this does not sound right to me based on the fragile nature of these thin walled tubes... and the brutal and uncontrolled nature of sootblowing

Or could this cleaning method be something that the Owner/plant staff cooked up to save them labor ??? ..... Hmmmmmmmmmmmmmmmm

Superheated steam sootblowing works well with the robust, thick walled superheater tubes and other thick walled tubing ...

I would like to hear if anyone else on engtips uses this brutal cleaning method on similar equipment in their plant...

Anyone ????



MJCronin
Sr. Process Engineer

 

[AKAK96]

@Edstainless, a lot of thanks for your valuable input!!!
Based on the available data, I tried to perform vibration calculations (manually) but this did not work as the calculation model I used was for TEMA shell and Tube Heat exchangers (which have a cylindrical shell). Earlier, I suspected it to be a failure due to vibrations but I could not conclude it accurately (critical flow velocity could not be calculated accurately). If anybody has a more relevant vibration calculation model kindly refer it.
The idea of liners had earlier been presented to the management but they did not find it impressive and so this idea was discarded. I'll try to convince them once more.
As for thermal fatigue, It may be due to soot-blowing, I am performing a thermo-structural analysis of Air Preheater model to confirm the operational temperatures which may be used to calculate thermal expansion (thermal fatigue as well).

@FacEngrPe Thanks for your input. Yes, I understand the importance of material type in such applications.

@MJCronin The Soot-Blowing system has been designed and provided by the OEM and is in its original shape. I will get the details of soot blowing (duration, frequency, etc.) for you to have a deeper look. Apparently, the failure is only at the tube ends and not on the tube-spans (where soot-blowing affects the tube directly.), moreover, the duration of soot-blowing is limited to 2 minutes at a time.

Anybody with anymore ideas please?

 

[Alistair_Heaton]

I suspect you are doing static fea on it.

The pipes will always sag due gravity and when they expande they will always go in the same direction with nlgeom on. Which will very quickly show the reason why they go at the interface with the end plate.

As for the welds I suspect they won't have done any pre or post heat treatment or used hot rods.

A suggested fix would be to pre heat the tubes to normal operating temps or even slightly above before installation. This will put them in tension when cold. But will stop the high stress due rotation while operating temp.

They are basically buckling due thermal with fixed boundary conditions at both ends. But due gravity and the slight sag it will always be in the same direction.

Hot installation will keep them in tensile load.