Based on the wisdom of having the stress analysis
performed by the GRE vendor.
On a certain job, it was specified that the vendor perform not only the thermal analysis, but also the
Surge and the dynamic analysis.
The GRE system scope for the vendor was an ambient temperature 16" water injection pump
suction header and manifolds. There is a spool of CS between the GRE piping and the pump suction.
In order to perform the surge analysis, the vendor needed a lot of data.
It turned out that it would take a comparable amount of effort to have all that data compiled, as much as it
would probably take to do the analysis itself.
It was later decided to remove the requirements of the surge and the dynamic analysis from
the scope. The likely surge pressures that the system will experience was concluded to be
less than the rating of the GRE piping being requisitioned. There was no vibration data
available for the pump. Besides the GRE piping was not directly connected to the pump.
Eventually the vendor was to do only the thermal analysis.
The vendor did the analysis using the CAESAR software. The modelling was the same as the CS
pipe analysis except that the property values for the GRE piping were used.
Overall, It occured to me that it may take only less effort to do this in-house.
The thermal expansion coeffecient of GRE is about 1.8 times that of CS and the Modulus of
elasticity is about 1/15 times that of CS. This attribute of GRE allows 2 anchors in a
straight line with only marginal restraint load increase, in comparison to loads that would
have resulted, were it a CS line.
There was a note about the GRE stress analysis being different from a CS some time back.
Could not trace that email by subject.
Sajit
This archive was generated by hypermail 2.1.8 : Tue Mar 04 2008 - 11:40:43 EST