Hi Ralph
The creep life study is still ongoing. I have managed to find material constants for the Norton Law from creep experiments. Many thanks to all those who responded to my request for help.
The whole project has not been as straightforward as was first thought.
(OK, no project ever is!). The pipework was originally designed to B31.3
for 100,000 hours. I have copies of the original Caesar analysis carried
out by a third party and they look fine. The system has completed 60,000
hours service and the operator would like to extend the life of the
plant
beyond 100,000 hours hence the creep life study.
I modelled the entire system in one model in PSA5 and again it is fine
according to B31.3. However, B31.3 only checks the thermal range
stresses
not the steady state (DW + Pressure + Thermal) that the system will see
continuously, i.e. the condition that will give rise to creep
I changed the design code to BS806 and then the fun began...
Non Intensified Deadweight (DW only at branches): shows several
overstresses
but these are simple to solve with additional supports.
Sustained Case (DW + Pressure): Some high stresses, but again not
difficult
to resolve.
Thermal Range (Thermal Range + Pressure): I left out the pressure to
compare
it with the B31.3 anlysis. OK, similar results to B31.3, slight
variation
due to the differences in coefficient of expansion and modulus between
the
codes for the same materials. I hope the material knows which code it
has been
designed to!
Hot Stress (DW + Pressure + Thermal): This is a creep case compared
against
an allowable of the average stress to rupture at the design temperature
in the
design life. This case shows that about half the elbows and several tees
are
overstressed.
I used the stresses directly from the Hot Stress case to assess the
creep
damage according to the method in PD6525.
From this it was shown that the creep damage on the worst of the tees
was
about 80% which seemed reasonable.
One of the elbows on an expansion loop showed a creep damage of 700%,
i.e.
it should have failed after about one year in service. This has
obviously
not happened. At the same time as starting this project we carried out a
metallurgical survey of several points on the line including this elbow,
the
replication showed no sign of any creep damage.
Although BS806 does take creep into consideration by restricting the hot
case
stresses to the average rupture stress the stresses calculated are based
on
elastic conditions.
The stresses on the tees are modified by a relaxation factor of 0.44 or
0.5
in the Hot Stress case, this is to allow for relaxation due to creep.
This modifier is not applied to elbows or bends and the stresses
calculated
for these items use the full elastic SIFs, between 1.5 and 2.5 depending
on
whether they are in-plane or out-plane.
If the SIFs are set to 1.0 the stresses are reduced by about 40% and the
creep damage on the elbow comes out to be 31% for 60,000 hours.
The next stage was to calculate the stresses, taking into account the relaxation. We have been using a paper by A.Suzuki "creep behaviour of piping components under combined loading". This takes the pressure, in-plane and out-plane bending, torsion and normal force from the PSA5 analysis to calculate reference stresses for each type of loading. One of the factors used in calculating these stresses is the material constant n.
The stresses calculated using the Suzuki paper show that the creep
damage
will be approximately 10% after 60,000 hours.
The current situation is that we are obtaining independent verification
of
the Suzuki paper and the methodology of obtaining the stresses.
This project has certainly been a learning exercise and shows that there is a need for an analysis method between a simple elastic code based analysis and a full blown finite element analysis. The system I am looking at contains just over 1000m of pipework so it is not practical to undertake FE on the whole system.
If anyone on this list knows of any method that could be practicably applied to a problem of this size I would love to hear about it.
Thanks for reading all of this, apologies to those who not interested in
creep.
Barbara
If you have received this E-Mail in error please return it to the
sender.
We should be grateful if you would also copy the communication to
<a href="/group/PipingDesign/post?postID=IhDuy7zmrk-Fr_ZHXRjf7z5cRgAFicp_SH8NGQNTnW0CxJMhO4Z2Oygj8Q4fk69uXY52koGd2rv3f9mz6pW1h_-7nfqMNKI">postmaster@mitsuibabcock.com</a> then delete the E-Mail and destroy any
copies of it.
It is your responsibility to scan any attachments for viruses.
For further information, visit us at WWW.MITSUIBABCOCK.COM Received on Wed May 30 11:09:00 2001
This archive was generated by hypermail 2.1.8 : Tue Mar 04 2008 - 11:40:12 EST