0707
Petroleum
- Jun 25, 2001
- 3,357
Some days ago we have had a water boiler tube burst. Burst location was in a radiant rear water tube at the top convection zone on an almost horizontal part of the tube, about 1,5 m from the top water tubes header. First visual inspection of damaged tube was a typical failure of a "thin-lipped" burst of the boiler tube. This type of failure results in a ductile rupture of the tube metal and is normally characterized by the classic “fish mouth” opening in the tube where the fracture surface is a thin edge.
After a closer microscopic microanalysis of the thin edge and good tube zones we didn’t found any differences in the microstructures between thin edge and adjacent good tube zones. No evident distortion of ferrite and perlite was seen.
For that reason we think the failure has occurred well below the lower critical transformation temperature with no evident distortion of ferrite and perlite.
Adjacent tubes don’t reveal any plastic distortion or diameter tube growth. After digital Rx examinations we found that adjacent tubes of boiler tube burst showed great thinning on top of part of the tubes at 12 o’clock location, the thinning was about 250mm long. Nominal tubes thickness is 5.6mm and thinner point in 250mm length was 1.5mm.Thickness of bottom part of the tubes were ok.
Due to the facts my conclusion is: tube failure cannot be explained by a short term overheating.
My question is can we explain the thinning on top of the tubes by cyclic turbulent erosion of wet drops of steam?
Tube goes 15m in a vertical run and then it turns almost horizontal within 2m after connecting with the upper water boiler header. On the vertical part of the tubes there is a equilibrium between water and diffused steam bubbles, but on the horizontal part of the tubes about 1m of the turn bend, by differences of densities the water runs on the lower part of the pipe and the drops of wet steam run turbulently above the water inducing high erosion on that 12 o’clock location of the tubes.
My intention is to warning other operators for this kind of failures and to have comments from the forum on my conclusions.
Tanks
Luis Marques
After a closer microscopic microanalysis of the thin edge and good tube zones we didn’t found any differences in the microstructures between thin edge and adjacent good tube zones. No evident distortion of ferrite and perlite was seen.
For that reason we think the failure has occurred well below the lower critical transformation temperature with no evident distortion of ferrite and perlite.
Adjacent tubes don’t reveal any plastic distortion or diameter tube growth. After digital Rx examinations we found that adjacent tubes of boiler tube burst showed great thinning on top of part of the tubes at 12 o’clock location, the thinning was about 250mm long. Nominal tubes thickness is 5.6mm and thinner point in 250mm length was 1.5mm.Thickness of bottom part of the tubes were ok.
Due to the facts my conclusion is: tube failure cannot be explained by a short term overheating.
My question is can we explain the thinning on top of the tubes by cyclic turbulent erosion of wet drops of steam?
Tube goes 15m in a vertical run and then it turns almost horizontal within 2m after connecting with the upper water boiler header. On the vertical part of the tubes there is a equilibrium between water and diffused steam bubbles, but on the horizontal part of the tubes about 1m of the turn bend, by differences of densities the water runs on the lower part of the pipe and the drops of wet steam run turbulently above the water inducing high erosion on that 12 o’clock location of the tubes.
My intention is to warning other operators for this kind of failures and to have comments from the forum on my conclusions.
Tanks
Luis Marques
