Dear sirs,

With a pressurized pipe, there are two enlongation: in hoop direction and in longitudinal direction. I think elongation in longitudinal direction behaves like thermal expansion. But how can i check it in Ceasar II?

You select the options for 'Activate Bourdon Effects' available in 'Special Execution Parameters'.

Many thanks Mr. Moverz.

In addition, i want to model in Ceasar II a pressure vessel nozzle as anchor with relative thermal expansion of the vessel.Do i model node 15, 40, 75 (attached file) as normal point and add displacement in x,y,z direction for them?

I'm looking forward to your respone.

That's correct.

Remember when you model a displaced anchor point, define displacements X, Y and Z and set rotations RX, RY and RZ to zero, otherwise there will be free rotation.

There are other ways to deal with nozzles too. See the User Guide.

You could also improve your model by including vessel shell stiffness using the WRC 297 facilities provided in Caesar.

Thanks Mr. Moverz,
I do not include vessel stiffness because it is more conservative than included case. It cause reducing stress.

If you ignore the real flexibility of a connected shell, you risk adding pipe loops that are not required, thereby increasing cost and reducing plant efficiency.

Simply ignoring inherent shell flexibility will not prevent its effects, so your analysis may be incorrect.

Yes, i agree with you.

Thanks for your advise.

Dear Mr. MoverZ,

I see that pressure enlongation is seldom included in pipe flexibility report. Do you have any idea about this?

Thanks.

Pressure elongation is real. However, it may or may not be significant in any particular analysis.

For example, I have dealt with gas transmission pipelines where pressure elongation had been ignored by software other than Caesar II. To illustrate this I equated pressure elongation to an equivalent thermal expansion. The pipe 'thermal dT' was 50 deg C, and 'equivalent thermal' from pressure was 25 deg C. Thus, in the previous analysis, expansion effects had been severely under-estimated. Numbers may seem small, but where long pipe lengths are concerned, the resulting error may be large.

Remember also, that in high temperature systems such as FCC pipework, although pressure may be low, so is the Young's modulus, so to ignore pressure elongation may give dangerously inaccurate expansion results.

It is a lot easier to model the column, tower, exchanger, etc. from their support anchor points all the way to the nozzle. The nozzles are tied to centerlines of the vessels etc. with rigid rods.

I agree with you, that's why I wrote ..." There are other ways to deal with nozzles too. See the User Guide".

Your suggestion for modelling simplifies multiple thermal load case work too.

One word of warning ... if it's a tall tower or similar, be sure that your 'rigid' element is actually rigid enough, by using a large diameter, heavy wall and zero weight. If a small diameter were used, because Caesar simulates a rigid element by increasing the wall thickness, you may end up with significant and unreal flexibility.

Better not to model tanks with rigids.

Regards,