Unrestrained shell rotation as per API 650

If the Tank nozzle is on X direction, in which direction the unrestrained shell rotation as per API 650 have to be considered. In Rx or Ry or Rz or do we need to consider in all three Rx/Ry/Rz.
Is it mandatory to consider this shell rotation.

Thnaks in advance for your support.

with best regards,
K.A.Saravanan.

API assumes the torsional direction is rigid, so the longitudinal and circumferential directions may have some flexibility.

So, if we assume "Y" is vertical, and as you say "X" is axial, then:

MX - torsion, asusmed rigid
MY - circumferential bending, may have some "less than rigid stiffness" as computed per Appendix P.
MZ - longitudinal bending, may have some "less than rigid stiffness" as computed per Appendix P.

Richard Ay,

Thank you very much for your reply. I understood the concept. With regard to this it will be helpful to clarify this also.
In CAESAR II when we evaluate the nozzle loads, we will be giving vertical and horizontal displacements as required. Also we need to consider rotation of tank shell, when we are considering this rotation that shall be considered as Rz (X is axial, Y is vertical and Z is transverse) and the Axial stifness will be in Z restraint, Stiffness for circumferential moment will be in Ry restraint and Stiffness for Longitudinal moment will be in Rz.

Please check my statements above, correct me if i am wrong.

Thank you for your support.

with best regards,
K.A.Saravanan.

You say "the Axial stiffness will be in Z restraint". No, axial is in X.

Keep nozzle stiffness and imposed deflections on the nozzle as two separate concepts. Your nozzle may have three flexible "degrees of freedom" (DOF) and three rigid DOF. The rigid directions involve (transverse) shear and torque. Flexible DOF may be axial, circumferential bending and longitudinal bending. These are solely a function of material and geometry. Imposed displacements are caused by tank load - thermal growth, settlement and bulge due to contents.

Those loads are on one side of the "flexible" nozzle and the pipe loads are on the other.

Yes i agree, its my mistake written as Z in axial. Thank you Dave Diehl for your clarification.

Dave Diehl,

While performing piping stress analysis, if Tank vendor has already given the Stiffness co-efficient for the radial thrust load (Kr), Stiffness co-efficient for the longitudinal moment (Kl) and Stiffness co-efficient for the circumferential moment load (Kc) values, do we need to consider the vendor values? or do we need to consider the computed values from caesar2?.

Please clarify this as i am different set of nozzle loads and moments.
1. If i consider vendor values, longitudinal moment is exceeding the allowable of the API 650 nomograms provided by vendor. Exceeding by 1 time (i.e. 200%).

2. If i input the computed stiffness from the caesar2 into the nozzle node and remove the API 650 modeller in the nozzle node, i am exceeding the allowable provided by caesar2 computed values, but closer to allowable provided by vendor.

3. If i consider computed stiffness from caesar2 into the nozzle node and retain the API 650 modeller also in the nozzle node, it is showing two sets of flexiblible and rigid stiffness. I hope this should not be done.

Please clear my doubt.
Thank you.