Hello,
I'm modeling a tower and the pipe system from the top of the tower.
The temperature at the bottom of the tower is 228℃ and the Temp at the top is 167℃.
The tower is 40 meters high.
The diameter of the pipe is 20".
The pipe run down 22 meters then run -z direction 6.5 meters.AT last the pipe run down FTF to a nozzle of a heatexchanger.
while the pipe run down along the tower,there I put 3 restraints,the topest one from the top of the tower is 1.732m ,the second one down 7m,the last one down 7m.At the middle of the horizontal pipe I put a +Y restraint.

My question is:
1,The tower is modeled as a rigid,this means it only expand ,not bend,isn't it right?
2,How can I conpute the average Temp of the tower?
3,The toppest restraint on the vertical pipe in most common case,is a +Y or a anchor?
4,Then the second and the third ,they should be a guide or a hanger?
5,The static output show that the restraint at the horizontal pipe doesn't take effect.Should I put a hanger there?
however ,the sus stress and the exp stress all passed.
6,when I input a hydrotest pressure,what can i get from the output report.
7,The pressur I input is a gauge pressure ,isn't right? So it could be a negative number?

Thanks for all your reply.

Here are my answers, with assumptions:

1) Your tower should be modeled as plain pipe, not a rigid element. This will allow the tower to deflect due to applied loads - such as wind. Yes the tower will see thermal displacments.

2) That is hard to say, because it depends on whether the diameter is constant, insulation properties, etc. Why not use several elements to build the tower, and set the temperature of each element according to process conditions.

3) Probably not an anchor, since this is difficult to achieve up in the air. A +Y is correct if the pipe is sitting (via lugs) on a support hanging off of the tower.

4) This depends on the specific tower and loads.

5) What do you mean "it doesn't take effect"? Are you saying this is a +Y that lifted off or something else? There was some thought in laying out the restraints for this system - right?

6) You can setup a hydrotest case, which will give you another loading condition for restraint design.

7) No, you don't perform a flexibility analysis with a negative pressure. The PD/4t longitudinal stress component should always be positive (or zero).

yuandb,
I would only add to clarify on 3) that supports hanging off the tower need to include CNode to get thermal growth of tower at the corresponding elevation.

While analyzing a tower system, one should ensure that the very first +Y support on piping after the tower nozzle should be as close as possible to the nozzle.This will ensure min. thermal loads to be transfered to the nozzle.After guiding the vertical pipe,sometimes you may go for the spring/rigid support in the horizontal run to keep the sustained stresses/trunion loads within the limits.

Saneh Gupta
FluorIndia

Thank you for your reply.
My new question is:
I define the tower as a plain pipe ,the highest restraints along the tower as a +Y(node 40),the second as a hanger and guide(node 50).Their cnodes are on the tower.Then in the sus case ,the restraint on 40 is 0 ,the restraint on 50 is so big.
IF the tower be modeled as a rigid,then 40 and 50 will share the weight.
which of above will reflect the true thing.

Do not model the tower as a rigid.

Supports along a tower are best modeled using "dummy rigids", extending horizontally off of the tower centerline, to the shell. From the shell out, you can use some other "non-rigid" piece of pipe or strucutral steel to get to the centerline of your pipe riser. Here is where you want the CNODE.

If you feel the hanger is carrying too much load, review your input. Did you accidently "free" a restraint you shouldn't have?

hello,
I don't know how to input the diameter of the "non-rigid" piece of pipe .

Its usually a number greater than zero...