I have to analyse steam piping for steam turbine. Some part of piping is vendor's responsibility (up to first shut-off valve). They have specified displacements at the locations, but no allowables loads are available at presentIs it Ok to consider these valve outlets as anchors with specified displacements or shall I model the piping up to turbine shell and ask the vendor for the displacements at the shell nozzles and also the allowable forces and moments.
Thanks for the suggestions

Hi Siv,

IF the valves are rigidly mounted to a robust structure you can model them using the displacements provided by the OEM. You cannot call thes points anchors as you will be specifying displacements.

However, many turbines have the stop valves and/or intercept valves mounted on springs. In these cases you will have to model the valves accordingly. Some times the piping between the valves and the turbine casing is special non-standard pipe. You would have to get the pipe specifics from the OEM.

I would suggest that you get a copy of the GE manual on piping to turbines "GEK-27060 - Design Recommendations for Steam Piping Systems Connected to Large and Mediun Size Steam Turbine-Generators" and a copy of the NEMA SM-23 Standard and read them thoroughly for some background information.

You might look through previous postings here to see what has previously been discussed. You might benefit from communicating directly with those who have posted on the subject - you can send them "Private Messages". Some of them may actually answer you.

Regards, John.

Mr. Breen, Thanks for your kind reply. The valves in this question are spring supported.
What i meant by anchor is to define the valve outlet as anchor with a C node and define the displacements at C node.
I can understand the concept of displacements and C nodes when applied to equipment nozzles as equipments can be considered to be rigid so that not to be affected by pipe forces as far as displacements of nozzles due to thermal effects are concerned .
However in the present case the vendor piping has 3 bends and few meters of piping before the valve.How can we maintain vendor specified movements at the valves, becuase our piping will also push/pull the vendor piping.
Kindly suggest the right approach.
Regards
siv

Hi Siv,

You should model the piping right up to the turbine connections (nozzles) and use the turbine OEM's provided displacements at that point (use the method of C nodes that you describe. Typically the turbine manufacturer will tell you that their springs, supporting the valves and their piping were chosen (by them) such that they support exactly the weight of their valves and their pipe (up to and including the valves). Of course to get the flexibility of their piping right in the model you will still have to know the diameter and wall thickness of their pipe. So you will want to model the turbine OEM's pipe and the rigid links that you use in the model to represent their valves as having zero density (as that weight is supported by the springs). You will have to support your piping fairly close to the valves and then appropriately place your supports on the piping that you design.

Also, you may have a very significant wall thickness difference between "your pipe" and the valve nozzles that you are going to be welding the pipe to. There are Code limits on how much wall thickness mismatch you can have at butt welds. To get the butt welds right you may have to use a "transition piece" with a tapered wall thickness between "your pipe" and the valves. Pay attention to the choice of material for the "transition piece".

Please send me a private message.

Regards, John.