I hope my question is not repeated, because I spent some time searching the forum and could not find it.

The Installed Load Case section of the Hanger Sizing Algorithm in the Technical Reference Manual reads as follows:

"The theoretical cold, or installed, load is the load on the spring when the pipe has exactly zero displacement. The actual installed load may differ from the theoretical installed load by (K)(d), where (K) is the spring stiffness and (d) is the displacement of the pipe in the installed condition. In essence, the actual installed load is calculated by taking the piping system and "freezing" all displacements at zero. With the pipe in this condition, the hangers are installed and the theoretical cold load is applied. The pipe is then "defrosted" and allowed to adjust its weight position due to the hanger, restraint, and anchor stiffnesses and the installed hanger loads. Once the system settles out, the total load on each of the hangers is read and recorded as the "actual" hanger installed load."

I understand what the passage says, but I have a concern (and please bear with me as I explain):

The typical construction scenario that I have experienced is that after the piping and supports are put in place and after all welds are completed from anchor to anchor, the system is hydrotested. Once safely hydrotested, the locks on the spring hangers are released. It is true that the system will initially be unstable as the loads are being redistributed naturally to balance out with the introduced springs' stiffnesses. However, once the springing of the system is settled, all spring hangers are adjusted to their Cold Loads (usually the white mark on the spring can). Once this "manual" adjustment is done, the system "freezes" and THAT is considered your zero position. From there, the pipe heats up and expands to the predefined Hot Load (usually the red mark on the spring can). Accordingly, CAESAR II should use the theoretical Cold Load and report the SUS case with zero displacement.

My concern is that if the SUS case reports a non-zero vertical displacement, the derived load (whether Hot or Cold) will be biased, because the program will calculate the load based on the "OPE" movement NOT the "EXP" movement. It is accurate to use the "OPE" movement to calculate the loads because of the non-linearities, but the "SUS" case should be based on a zero vertical displacement.

Please advise if there is any way to resolve this. I want to have the program use the theoretical cold load and zero displacement in spring calculation. Is it possible?

Thank you for your time

Is there anybody who can help me on this...?

Mr. Dave/Richard,
1. Is this not doable using the latest version of CAESAR II?
2. If not, could you please consider this as a refinement to be considered in future releases of the program?

Thanks,

The SUS case may have non-zero displacements, depending on the spring stiffness as stated in the Documentation. The reason for this is that the system is balanced in only one position of the pipe, and that position is assumed to be the hot position. You do not need to have zero displacements in the SUS case. What would happen if you had more than one SUS case, they can't all have zero displacement at the hanger location?

Thank you Mr. Richard.

I think that the piping could be balanced in either Cold or Hot position depending on what it is designed for. Nevertheless -as I have stated above-what really happens at site is that the springs are adjusted to their Cold Loads after the locks are removed and then the pipe "freezes" at that adjusted position. Therefore, there is really no displacement at the spring locations during the SUS condition.

I agree with you that there could be more than one SUS case, but in reality and practically speaking, most of the piping systems will experience only one W load, while the P loads could change (P1, P2, P3, etc.), and pressure does not significantly influence the vertical loading. With that said, is it possible to have the program use the theoretical cold load and zero SUS displacement in spring calculation?

Thanks again.

Change to "cold load design", that is as close as you're going to get.

Thanks. I will do that, but one more thing: Is it possible to consider my proposal to have zero SUS displacement with theoretical cold load (only as an option) in future releases of the program?

Best Regards

Not without further study, at the moment I disagree with your argument.

Mr. Richard/Dave

I finally was able to sketch my concern on a piece of paper, which I hope will make things clearer (please find it attached here):
http://imageshack.us/photo/my-images/20/springsexample.jpg/

Now what CAESAR II does is:
1. Get the balanced load from the W case (CASE 1 in my sketch)

2. Calculate the operating displacement from the W+P+T case

3. Assign the balanced load to the Hot condition to be the HL (assuming that we select Hot Load (HL) design)

4. Theoretical Cold Load (CL) is calculated by taking the HL (W case) and adding the vertical movement (delta) from the OPE (W+P+T) multiplied by the selected stiffness of the spring (k).

If you look at the output screen for any CII run with a spring, you will notice that the SUS case reads a load different than the Theoretical Cold Load in the "Hanger Table" screen. When we buy our spring, we specify the Theoretical Cold Load (CASE 3 in my sketch) and HL from the "Hanger Table". Meanwhile, CII calculates all reactions in the system in the SUS case based on the Actual Cold Load (CASE 2). This creates an inconsistency in the results, because the installer will set all the springs to their Cold markings (CASE 3), while CAESAR II SUS calculations are based on different numbers from CASE 2.

Do you understand my point or did I confuse you (sorry if I did :-))? If I am mistaken, I would really appreciate if you clarify this point for me because I am having doubts that I am missing something; otherwise, please confirm my understanding.

Thanks for your help.