OD thermal fatigue of waterwalls

[Gem1]

Hi All,

I'm trying to pin down a cause for an OD thermal fatigue cracking problem in boiler waterwall tubes.

The cracks are slow growing and have the characteristic vee shape and S spine associated with thermal fatigue, but there is minimal thermal degradation of the tube. There is a very thick OD scale in the cracked area and it is shedding, but not due to water cannons or any other cooling method. I'm having trouble understanding how the cooling aspect of the thermal cycling is occuring.

Has anyone had any experience that could shed some light on this for me?

Thanks!

 

[metengr]

Gem1;
Circumferential OD cracking in waterwalls is caused by thermal stresses from several factors; these stresses can be caused by heavy tube ID oxide deposits (magnetite) due to infrequent boiler chem cleanings. There could be a feedwater flow or distribution problem within the tube circuit due to orifice problems where you have preferential subcooling. If you are a member of EPRI there are several excellent EPRI reports on this problem.

 

[Gem1]

Thanks metengr. I've checked the bore scale and, although slightly above reportable limits, I'm sure that isn't the issue. I've read the EPRI reports, which are good, although the aren't helping me too much with this particular problem. I'm leaning towards a flow or distribution problem at this stage, there have been rumours of some imbalanced burning.

Do you know how rapid the thermal fluctuations need to be to result in thermal fatigue?

 

[metengr]

>250 deg F

 

[davefitz]

There are several causes.

If the unit is a subcritical drum boiler, startup subcooling of the waterwalls and downscomers can cause thermal fatigue, especially in the vicinity of windobox supports and other waterwall stiffeners. Refer to some B+W papers from the 1980's.

Other subcritical drum waterwall events relate to upsets in cooling of the waterwall, probable root cause is upsets in drum water level local to the bellmouth inlet to the downcomer causing a temporary drop in watewall circulation to the sections of waterwall fed by that downcomer. This leads to an alternating dryout event ( mid furnace or upper furnace) that may temporarily cause a single tube's hot side surface temp to increase by ( + 250-350 F). As the other waterwall sections prevent this tube from freely expanding , the local metal at that tube's hot side yields in compression, later generating residual tensile stresses that lead to surface cracking. See correlations by Kon'Kov and Dorsochuk and papers by KWU Siemens authors in Intl journal of multiphase flow circa 1988-1994 for calcualted upset increase in metal temp due to dryout events ( to be distiguished from DNB events in th elower furnace).

Simialr behavior occurs in supercritical units under some scenarios.