Hi !
My understanding is that AA-2.6 does not require the recalculation of the stresses in the shell or channel. It recognizes that the shell/channel is overstressed and consequently calculates a required thickness for the tubesheet assuming the plastic deformation at the junction with the shell/head. Why does Codecalc 6.20 calculate stress in the shell and compares it with an allowable value. This is misleading. Please comment.

Michael

This is a very interesting point.

After careful consideration and consultation with an ASME committee member dealing with this appendix, we agree with your comment. The elastic plastic iteration reduces the strengthening of the tubesheet due the shell or channel and so we have to check to make sure that the tubesheet and tube stresses are still acceptable.

One of the pre-conditions of the Elastic-Plastic iteration is that the shell or channel stresses be less than 3 *S. After an elastic plastic iteration is performed, this condition is already satisfied. Hence, we do not need to recalculate the shell or channel stresses.

For information purpose, in case of elastic-plastic iteration, the comparison of the shell (or channel) stress should be with 3*S, using the actual stress from the initial pass.

We will implement this change in the next version of the program, to be released in couple of months.

Regards,
Mandeep

Thanks for your prompt reply

Since you plan to revise the ASME fixed tubesheet design, I would like to suggest the following change: Under tab "Tubesheet", there is a check box titled "Is this a Pressure Only case". If a user is not very familiar with the Code requirements, he might overlook the importance of this . I suggest that Codecalc runs (2) sets of calcs: one "Pressure + Thermal" and the other "Pressure only". Also, in the "Pressure only" case, Codecalc should automatically set the value of gamma (differential thermal expansion) at zero regardless of the data entered by the user.

Michael

[This message has been edited by nawarm (edited October 27, 2000).]

The next version will address the multi load case requirement. I think you will find it very useful.

Regards,
Mandeep