Hello All,

Subject of Discussions :
Differences in Temperature within a piping system of 3 parallel lines connected together with 1 line working at a time and other 2 lines having stagnant fluid withtendency to have cooler or hotter ambient temp.
.
Service : Crude Oil
Pipeline condition : vertically laid outdoor & uninsulated
Temperature Condition :
Service temperature = +65°C
Min. metal temp. = minus 3°C (ambient condition)
Installation temperature = +21.1°C
Max.metal temperature = +85°C
(due to Solar radiation)

We have a specific case as furnished below :

3 vertical riser pipes of 3 different pipe sizes of 6”, 8” & 10” are connected to a header of 20” both inlet and outlet of risers. The header at the outlet of riser is connected to inlet nozzle of a storage tank.
All the 3 riser pipes are carrying same crude oil of different flow rates with same operating pressure & temperature of 65°C. At operating condition as well as during shut down, (3 risers being outdoor & un-insulated), the metal temperature may rise to 85°C due to solar radiation. The design minimum metal temperature is furnished as -3°C. (Calculated from minimum predicted ambient temperature from 100 Yrs data)

The issue is “at a time only 1 riser will be used depending on the crude flow available and the other 2 risers will be closed through upstream valves. Since near the header at the outlet of the risers, there is no valve to isolate the risers from each other, all the 3 risers will be flooded with crude oil of operating temperature 65°C. when only 1 no. riser in operation and other 2 risers will contain stagnant fluid since there is no isolation from the line in operation.

Our query is :

“Is it that we require to analyze this piping system with temperature profile of running riser that of operating temperature of 65°C ,while other two riser pipes carrying stagnant fluid with temperature profile of far less temperature like minimum metal temperature of -3°C / maximum metal temperature +85°C because of the fact that the stagnant fluid may may tend to cool down or heat up the 2 riser lines by the direct influence of ambient temperature even though both the risers are connected open to the running riser carrying crude oil of 65°C.
or

“We analyse all the 3 risers exposed to same temp. condition - be it ambient max. (due to solar radiation), be it operating temp. , be it installation temp. , or be it exposed to minimum metal temp. of the ambient condition and analyse the thermal displacement from one temp. condition to other temp. condition or any other combination of these temp. conditions to be analyzed.”

Please Clarify as early as possible.

Thanks,
P.Manivannan.
Piping Design Engineer,
HEISCO, Kuwait,

This is a classic design issue that also pertains to, for example, pumping stations where one or two pumps may be out of service (spares) at any one time.

The normal solution is to run an analysis case for each possible combination of "cold" (out of service) branches. It's easiest to check with ambient conditions first (or, if the piping is heat traced, the minimum design heat trace temperature). If all of these load cases pass, you are done.

Usually, one or more load cases will not pass this check. Then it becomes necessary to reconsider the temperatures of the "cold" branches. If engineering judgment permits increasing the design temperatures, do so. Perhaps this set of runs will pass. If so, you are done.

If this still does not work, you will need to add flexibility to your piping system to reduce the stresses and/or equipment loadings.

From past experience, these sort of problems can be quite time-consuming. However, it's better for a stress analyst to do a lot of work to get the system designed right than to do a lot of work to try to justify his omission of analysis in the design of a system that has just failed.

Or, as I told one of my co-workers the other day, "I'd rather blow my budget making sure the pipe is hanging right than have you guys try to determine the right way to hang the stress analyst."

This is a straight forward sectional system complicated by there being three "ambients"; installation, frost and solar. Operationally you will have one out of three pipes in duty (1oo3). Thats 3 load cases. However you may have just changed duty pipe so consider 2oo3 as well. Thats 3 more load cases. Because you have 3 "ambients" thats 18 load cases. It is probably best to do 6 then copy the run and change the "ambient". The stresses will be coped with by looping away from then back to the upper 20" header. You might have to play with loop size a little. This should leave you with loads on headers which are largely from deadweight.
You say the upper riser connects to a nozzle. But how close? Your real task is to ensure you meet the nozzle allowables be they API 650 or other requirement. That is why you want the loads tamed of large thermal variation so loads are largely deadweight.