Hi all

Please provide me with information of cold spriging effect in Piping stress analysis my questions are like

1.How it is done at site
2.How the Forces & moments on nearest anchor are reduced
3.Is there any effect in stresses in the system
4.On what basis the cold springing point in the piping system is selected
5.Is this practice is recommended or not.

More clarification on this topic would helpful

Thanks in adavance,
Rakesh

Having no direct experience with cold spring, I'll leave 1, 4, and 5 to others. The theory, as I understand it, is:

2. Cold spring pre-loads the anchor in the opposite direction. If you expect 5000 lbf of compression on the nozzle, you might cold spring the line to put 2500 lbf of tension on that nozzle. Thermal expansion would reduce the tension to 0 before applying the remaining 2500 lbf of compression to the nozzle.

3. Stresses, yes. Stress range, no.

Hope that helps some.

1. Big chainfalls, cherry pickers, or other lifting equipment. Usually, a particular deflection is selected based on the stiffness of the pipe as analyzed - this is usually easier than finding calibrated load cells of appropriate precision.

4. See Steven Perry's answer to (2) above. Simple math.

5. NO. NO. NO. It becomes a real pain to maintain cold sprung systems - if you have flanged in-line components, you often need to cold spring the pipe every time you replace them. If your system is all butt-welded, cutting it apart could be extremely exciting. (This is also true of old systems that have self-sprung.)

Essentially, cold springing is an attempt to replicate the behavior of a fully self-sprung (several to many cycles of operation) piping system, where the loads on terminal equipment have settled out so that they are about equal in absolute magnitude in the hot and cold positions (and of opposite sign). Usually, we use 1/3 of the total free movement of the terminal point for our cold spring, primarily because the sensitivity of the load reduction vs. deflection due to cold springing is very high and we don't want someone to be able to hand-wave away a large nozzle load by simply specifying 50% cold spring when it would be impossible to do this that precisely in the field.

It's much better, from every standpoint (except capital cost and, occasionally, space considerations), to simply make the piping system more flexible. The ratio of capital cost of the piping to capital cost of the equipment is extremely low, so an extra leg of pipe and a couple of elbows is cheap insurance for protecting equipment.

We also have much better analysis technology available today than was available when cold springing was common, so we can locate axial line stops, and a full complement of restraint hardware to keep our nozzle loads in line.

1. Sometimes with ease sometimes with great difficulty. Basically you erect the line minus the cold pull weld. When the pipes are stress free with specified gaps in X, Y and Z direction inspectors are checking and confirming and the gap is closed and weld welded.

4. Try to select a spot that is possible from erection point of view. Discuss with experienced erection people. I had a case many moons ago where 3 dimensional cold pull was originally split into 3 welds. Could not be erected. Recalculation and one point only at the end. I was the one who signed off the fitt-up. Definitely not what the calculation engineer dreamed but to best of my knowledge line is still operational some 20+ years later.

5. There are many far more experienced on this issue on this forum but what I figured from some literature on main steam lines it was very common long time ago. Then the use was reducing to a point where it was not used at all. Now it has turned out that some new materials are such that cold pull is about the only solution. Some codes do not give any benefit to the use of cold pull. With "low" temperature lines where compensators are used cold pull is sometimes the only solution to make sure that the compensators remain within allowable movements.

Try to make your design without a cold pull.

Ref Steven's response to point 3 (stresses)

"3. Stresses, yes. Stress range, no"

I was under impression that cold pull does not alter the developed stresses in the system in any way. Perhaps I was wrong - will go through the theories again. Any volunteer for simple explanation of the stress phenomenon in a cold sprung pipe?

cheers,

Let me modify Steven's response.

3. Code Stresses - no. Stress Range - no. Actual stresses - yes.

Cold springing (or cold pull) has no effect on sustained loads - the pipe weighs the same whether or not you cold spring it.

Cold springing has no effect on the thermal stress range, which is defined as the algebraic difference between the system's hot and cold state of stress.

Cold springing DOES affect the state of stress within the system in both the hot and cold load cases - its effect on the terminal equipment loads is caused by the deformation created by the cold springing. This deformation, in turn, imposes a state of stress within the pipe that is different than one would find if the system had been installed with no cold spring.

BUT (to emphasize the Code rules once more) the cold spring has no measurable effect (since we assume linear behavior of the piping system) on the algebraic difference between the hot and cold stresses.

Think of it using this analogy.

I get up in the morning to go to work. I get in my car, and set the trip odometer to 0. I drive to work, note the trip odometer reading, put in a full day, and go home. When I get home, the trip odometer records how many miles I have driven, and the difference between the reading when I got into my car after work is X. The next day, I am driving to work, and I stop for gas 1/3 of the way there. I reset my trip odometer to 0. When I get to work, I note the trip odometer reading. It is different than on the previous day. I work a full day, get in my car, and drive home. The difference between the reading when I left work and the reading in my driveway is the same as the day before, but the absolute trip odometer readings are different.

That's what cold spring does - it changes the minimum state of stress within the piping system from the installed temperature to some temperature intermediate between the installed and operating temperatures. By making this change, the absolute magnitude of the larger thermal load imposed on the terminal nozzle is decreased, at the expense of increasing the magintude of the nozzle load at the other end of the operating thermal cycle.

That's ALL it does, when only Code stresses are considered. For those of you paying attention to the "hot operating load case," cold spring will change that, too.

I think your understanding was correct, Sanjay, but I hope this discussion increases your understanding of how the B31 Codes look at fatigue stresses.