My compliments to Mr. Srinivasan for all the point especially his prescription regarding unlikelihood of pure thermal fatigue here.

I agree with you all my friends and thank you for your good, constructive and helpful replies. We are scheduling a shutdown for the repairing, the line is still working and for this reason SCC, Sigma phase or other fail can only be checked through metalographic replicas, which at the moment are impossible to perform. Because of the thermal socks procedure, to spall catalyst particles on rotor expander blades, we think that thermal fatigue is only a hypothetic cause for the failure not the definitive one. Keep on touch, I will try collecting more elements to clarify our discussion and will let you know. Meanwhile your inputs are always welcome. Regards Luis Marques

Luis I suggest that the following may be carefully looked into while doing detailed analysis to establish the predominant and contributotry causes for this failure: 1. First symptom of cracking - when noted and after how many years. 2. Periodicity of steam injection to spall catalyst dust in turbine exhaust blades. 3. Frequency of FCC shut down history during the recent 5 year period. 4. Whether insulation is provided for the piping and valves or by chance some one has removed the insulation after the cracking was noted. 5. distance between steam injection nozzle and butt weld crack (described as circular cracking in your note) 6. Thickness used for the piping and scheduled rating of the valve - 150# or 300# 7. cracking pattern - is it branched or only straight line? 8. Process design and any inadvertant use of excess quenching opertaion done due to steam valve throttling during operation? 9. Distance between the cracking and steam inejction nozzle ? 10. Sampling of the corrosion product - for doing Augur Spectroscopy and X-ray diagnosis or SEM approach in addition to conventional metallography? Metallography will reveal the nature of cracking and this is to be done during the proposed shut down of FCC unit. Augu will help you to quantitatively determine the corrosion product elemental analysis. Based on these, you can do further diagnosis to determine the predominant and contributory causes for this crack propagation. Trust this is of help to you C.V.Srinivasan Nishi Engineers Pvt Ltd India August 19, 2005 E-mail : nishi@vsnl.com >Cracks were found on a circular weld on top part of the >48-inch pipe at each side of the pipe diameter. Cracks are >600mm length and 1 to 3mm wide. Left crack is on center of >the weld. Right crack is on HAZ; a small distortion of about >3mm is seen on HAZ of right crack. >Steam injection is done on top of the 48 inch pipe 1 meter >way and downstream of the circular cracked weld. Diameter of >steam injection pipe is 6 inch. Flue gas temperature is >700ºC and the cooling rate for the thermal shock is 150ºC/h. >When pipe temperature reaches 400ºC is injected superheated >steam of 24-bar g misted with boiler feed water to spall >catalyst particles of rotor expander blades to bring machine >vibrations to limited acceptable levels. > >Regards >Luis Marques