Hello,

I am currently working on a stress analysis for a main steam system which has cold spring built into it. Just prior to running it, I'm beginning to have second thoughts about the load cases. I looked at the recommended cases from the software, but changed them to what I though would be the proper cases. I attached a picture below showing the difference.


Why wouldn't L4 be a sustained load case?...Is Caesar telling me that I should not compare the stresses calculated in L4 to the sustained allowables? Why?

Another issue that I was wondering about was warning 78E that I got in this model. It says: "COLD SPRING and HANGER DESIGN are requested in the SAME job. The hanger restrained weight run should be made without concentrated loads. Concentrated loads activate cold spring effects in CAESAR II analyses."

So as a result of that warning, I chose not to put F1 into L1. Is that basically what that warning was trying to accomplish?

And I have one more general question about cold spring...If a client requests displacement information for a piping sytem with cold spring, I would include the cold spring in the displacement calculations, correct? Like in this case, use displacements for L3 as I defined it in the attached picture.

Thanks in advance! Any help would be appreciated!

Please try not to use cold springing in your system. I do not see any advantages on this option, specially in the long period of time. Try to put more flexibility on your pipe (loops, legs, etc.) or try to add expansion joints with tie-rods. Or try to find a much better way, this is what we are suppose to do.

Goodluck & Cheers!

I'm actually not designing the system. I should have been more clear I suppose. I'm doing a stress analysis for an existing system which was built in the 1950's.

See pages 6-6 and 6-7 in the Technical Reference Manual.

Also see page 6-18 in the User's Guide.

Hi Amar,

Somewhere in my reading in a textbook say's that CS are "JUST" use for equipement load, end point reaction, and others and not for code compliance.... If any feed back from our senior im happy to hear it.

Regards!

Cold spring is useful in reducing the time it takes to shakedown a system.
Where no shakedown exists, cold spring can also bias installed and operating loads on equipment. But it is difficult to predict actual loads. B31.3 says use only 2/3's of your cut short in the load analysis and Appendix P says run both 2/3's and 4/3's.

Bom mentioned it rightly. Mainly, Cold spring is used for keeping the end point reactions within nozzle allowable range.

If allowable range for an equipment nozzle load is ("-a" Kgf to "+a" Kgf); and actual end reaction of "x" Kgf is outside the acceptable range;

then

Cold spring can be used to just shift the actual end reaction range("0" Kgf to "x" Kgf) to some accptable range say ("y" Kgf to "z" Kgf) such that:

|"y"| & |"z"| < |"-a"|.

However, a system failing in code compliance (actual stress range > allowbale expansion stress range) cannot be corrected by cold spring. It cannot reduce the stress range.

Guys; please comment for necessary correction.

I appreciate the responses, but I'm still a bit confused.

I guess my real question is, why wouldn't you take cold spring into account in your sustained case? ...That way you could see the stress levels in the sustained case.

Here is my situation, I ran the analysis, and here are two of the load cases:

6 (SUS) W+P1+H+F1+CS
7 (SUS) W+P1+H+F1

The analysis used the same allowable stress value (rated at the hot temperature) for both case 6 and 7. But case 6 failed, case 7 did not. Meaning that adding the cold spring into the load case increased the stress.

Now, I have read the code section and other tidbits in books and I understand that they say you can't take cold spring into stress considerations hoping that it will reduce the stress...but what if it is increasing the stress like I'm seeing here?

Is there some larger flaw in my thinking?

I guess another way to look at it is, what load combination would I need to assemble in order to see the stress in the pipe as a result of the deadweight loads after the pipe has been welded up and cold sprung, but is not operating...meaning the pipe is at, lets say 70F. How would I look at the sustained stresses given those conditions???

Richard,

I read the technical reference manual and it says "...The end forces applied to the elements are then included in the basic loading case F (for force), whereby they can be included in various load combinations."

Does that mean that the CS forces are actually found within the "F1" load? Then what exactly does the "CS" load do?

Cold spring does not change stress range. Stress range is used in evaluating the effects of fatigue.
Here you are trying to include cold spring in the sustained loads used to evaluate collapse (sustained stress).
Your cold spring is a (self-limiting stress) caused by strain and the Code offers no evaluation of this load in combination with collapse loads. While you may have some reason for evaluating the state of stress in an operating position, the Code does not. In fact, because of Code indexing of stress (the SIF), I would have no reason to expect a Code calculation (and then, a CAESAR II calculation) of stress to match any stress value collected from the field.
Engineers can expect some correlation between field measurements and CAESAR II calculations for loads and displacements as long as there is no shakedown. But when it comes to stress, the Code has its own "reality" to ensure safe piping systems. Follow the rules and no one gets hurt.

Regarding your quotation from the technical reference...
I'm not sure of your reference or if it is current. CAESAR II used to activate cold spring with the (at the time) single force set - F. For many years now, cold spring is activated with the CS as you show it above in load case 6.