I have a question regarding the load case combination for occasional loads and frictional effects.
Consider we have a large occasional load such as due rupture disk blow out. The load cases are
L3 W+T1+P1+H (OPE)
L4 W+T1+P1+H+F1(OPE)
L5 W+P1+H (SUS)
L6=L4-L3 (OCC) (ALGEBRAIC)
L7= L6+L5 (OCC) (SCALAR)

When F1 is very large (say 40 tons) this will impose large frictional forces on the system in L4 case . But practically F1 acts when the system is hot and does not affect the displacement except for the direction of F1(if there is no rigid restraint in that direction). This can lead to erroneous results for case L4 and L7.
I would like to know what is the best method to deal with this issue.

Regards
Siv

Better to mull over the meaning of life than to dwell too much on friction...

First The system is hot and essential not in motion.

Second then the new load is applied to it.

But hell its all subjective Mu is all over the place and as soon as the surfaces slide it changes drastically.

You are engaged in Brain Surgery when a couple of stitches are all thats needed. Account for friction in the design details, just because CAESAR II can incorporate Mu doesn't mean you should always use it.

Is there some piping Code somewhere that allows you to "take credit" for the damping action of friction during an occasional loading?

Regards, John

Use the [Search] option (to the right and just above the Calendar). You'll find many posts on the subject of "friction".

Dear Mr. John Breen and Luf and Richard,
Thanks for the replies.
I did read posts considering friction before posting this query,and again read some of the posts after your reply but could not find an answer. or may be I could not understand it properly?
What I understand is like this.
1. Friction acts on the piping when the system starts up and acts till it is moving.
2. The rupture disk in my example will burst when the piping has attained it's temperature therefore at that time I assume the piping is not moving and therefore friction is not working.

Now consider a case where a rupture disk is installed in a vertical pipe. The pipe is supported at the bottom(say node 30) .There is a displacement of 30 mm on the bottom support in horizontal direction in case L3 (W+T1+D1+P1+H). Now the disk ruptures and reaction force of 40 tons acts vertically downward. Case 4 (W+D1+T1+P1+H+F1) shows a displacement of 20 mm on the bottom support due to increased friction of 8 tons (friction factor 0.2). This creates more stresses in the pipe( case L7) because the thermal movement has been restrained and the anchor or equipment nozzle has to bear this additional load.

In my understanding I am making a mistake in making case 4. But don't know a better way.

That's why I seek your advise.

Any advise is appreciated.

Regards
Siv

nope it may or may not be the frictional load but it MAY be! Thats the point if it may be it needs to be considered,.

It is hard to visualize your configuration. Is the relief load horizontal or vertical? Where are the supports that cause the friction concern? So I offer general comments on friction and relief loads.

Friction is a consideration as the thermal force slowly increases during startup. When the thermal force exceeds friction; the line moves. In the case of relief, the load is instantly applied and the line moves.

For relief, I consider:
Case 1 - system hot & relief line ambient
Case 2 - Case 1 + Force
Case 3 - system hot & relief line hot

For static solution the Caesar load cases are:
L1 W+P+T1 OPE (Normal Operation - My Case 1)
L2 W+P+T2 OPE (Relief Flowing - My Case 3)
L3 W+P SUS (Code Sustained)
L4 W+P+F1 OCC (Code Occasional)
L5 W+P+T1+F1 OPE (Relief Loads - My Case 2)
L6 L1-L3 EXP (Code Expansion T1)
L7 L2-L3 EXP (Code Expansion T2)

Of course dynamic analysis is the most accurate way to solve the case of relief force since it includes the system response to a momentary load. Discussion of DLF is another topic.

Regards and Happy New Year, John

Don't make the mistake of assuming a relief load is constant. Violent things are happening at the edges of the rupture disk (or orifice). For liquids, tremendous amounts of turbulence exist. In some cases, the liquids will flash to vapor. For gases, there will be normal shocks all over the place, often radiating out from imperfections on the periphery of the opening.

All of these phenomena have the potential to produce (relatively small, but significant) loads perpendicular to the direction of the reliving flow vector. In addition, they have the potential to produce significant vibratory loads, particularly if the relief device is simply resting on a support.

We ignore all of these things because they are small in relation to the relief loads, and because we do not want the analysis of one relief device to be a career-long project. But just because we ignore them for convenience doesn't mean we can then pretend they don't exist and make further simplifications to our analysis.

Einstein once said something to the effect that the goal of science was to make everything as simple as possible, but not simpler. IMHO, ignoring friction effects for relief loads is, in fact, trying to make the problem simpler than possible.

Dear Siv,

Friction force acts on your nozzle at the instant when the pipe is about to move or when it is moving.
If the rupture disc reaction force will act during those times, then you have to consider the increased friction force due to increased normal force.
Your problem now is to find out when your disc will rupture.

Why not relocate your rupture disc to avoid such situation?

I also noticed that you have a spring hanger somewhere. I read somewhere on this forum that you have to be careful when using springs on systems with dynamic forces.

I hope I added more confusion...

all things being equal, the inherant violence and vibrations observed during a relief event will have the pipes and shoes on a relief stack bouncing everywhere (within its guide), thus disconecting the shoe from the support and eliminating (significant) friction before damage ocours.
Thermals are applied AFTER the initial BD event decause BD only go once and the (downstream )line is initially cold prior to rupture

If you are trying to remove red ink from the output report and overloads are smallish, then maybe use the above rationale, if not, use worst case and deal with it.(play safe)

oh, and DONT use springs. Bad idea (IMHO) rigid supports only.

The rupture events happen VERY quickly, thus the increased friction loads due to the pressure fronts are transient and may well not have enough time to cause significant damage.

I suggest you walk through every event you are analysing in minute detail priot to generating load cases, you might be over egging your pudding.