Page 95 - Handbook of Materials Failure Analysis

References 89

Data of Tables 4.5 and 4.6 show that by applying a safety factor the experimental
residual life of tubes with diameter deformation greater than 1.5% become danger-
ously close to 2 years (the service time planned until the next plant stop), or even
shorter in the case of some tens of centigrade higher.

3 CONCLUSION
Radiant tubes are very critical components being exposed to severe conditions for
long time during service and, even thought they are designed to work over than
10 years, premature failures may be often observed. Their life is primarily limited
by creep damage, driven by a combination of high service temperatures and hoop
stresses that can evolve also into catastrophic failure. In order to enhance creep
behavior, materials have been widely improved from the first high Cr-Ni alloyed
steel to the last generation of Cr-Ni alloys, in which microalloying elements have
significant effects.
The tubes considered in this work, made of a HP grade alloy modified with Nb and
Ti additions,wereinspected in situ, duringthe plant stopplanned every 2 years, through
diameter measurements along their axis performed by a LOTIS. They were put out of
service when diameter deformation reached a limit value of 1.5%. In order to assess the
reliability of such criterion, tube’s degradation was investigated to verify any correla-
tions among diameter deformation, microstructural damage, and creep behavior.
Starting from the observation that the measured diameter values are not uniform
along tube axis, it was ascertained that samples taken from undeformed sections have
microstructural characteristics similar to those of samples in the as-cast conditions;
while a significant degradation is present in samples taken from zones where diam-
eter deformation exceeds the limit value assumed for tube decommissioning. In these
zones, the Nb-rich phase transformed, during service, from Nb carbides into Ni-Nb
silicates, known as G-phase, producing negative effects on creep behavior; but the
large peanut shape voids, that give rise to creep failure, are not observed.
Concerning creep test results, samples taken from undeformed sections have the
same behavior recorded in the as-cast conditions; differently a remarkable shortening
of creep rupture time compared to the as-cast conditions is recorded in samples taken
from deformed zones.
Therefore, from the results of these investigations, a link between tube diameter
deformation and creep damage is evident. Moreover, considering the results of resid-
ual life calculation based on creep experimental data, it can be stated that the limit
value of 1.5% for the diameter growth appears reasonably a conservative criterion for
deciding the tubes substitution.

REFERENCES
[1] Rostrup-Nielsen JR. Syngas in perspective. Catal Today 2002;71:243–7.
[2] Zunhong Y, Cao E, Wang Y, Zhou Z, Dai Z. Simulation of natural gas steam reforming
furnace. Fuel Process Technol 2006;87:695–704.