I am doing a stress analysis case involving Siesmic loads. The G forces can be in all X,Y and Z directions simultaneously. It can also be in positive and negative directions. Some compnents of the uniform load can be dorment as well. There can be many permutations and combinations these G forces can occure.Naturally we are looking at the worse case. This is for code compliance B 31.3 and Nozzle Loads limitations using API 610, 617 etc.

However Ceasar allowes only 3 vectors for inputing these uniform loads.

Question 1. How are these multiple vectos be input in Ceasar.
Question 2 How are output Load cases defined to get the Code Compliance ( Sustained & Occasional cases) and for the Nozzle Loads unless X, Y& Z input separatelly treated at input stage.

I fear that I have to run the model with many Uniform Load input scenarios. I still have a difficulty in defining output load cases, if both X and Z unform loads are mixed in one Vector in the input spread sheet.

I also have to add Wind condition.

What I realy mean is that, for Wind and Unform Siesmic load have to have input cases for each +x, -X, +z ,-Z,

Adding to this case is that Siesmic case can be in any combination of different directions such as +X, -Y, +Z. Or some components be dorment as well.

Your comment is very much appreciated.

Thanks & Regards,

Rienzie

See if this helps

U1--- siesmic acc. in x dir
U2--- siesmic acc. in Y dir
U3--- siesmic acc. in Z dir
Now load cases can be made like this

W+P1+T1+U1
W+P1+T1-U1
W+P1+T1+U2
W+P1+T1-U2
W+P1+T1+U3
W+P1+T1+U1-U2
etc........

Load cases for wind can be made in similar fashion.

If I understand correctly B31.3 does not ask for combining wind and seismic in one load case.

Regards
Siv

What design code are you using ?

B31.3 for example does not require wind and seismic loads to be considered as acting together. Maybe you could simplify your analysis ?

nor does B31.1

Thanks,

Yes 31.3 does not ask for wind & Earthquake together. I would assume the same situation for API 610 and API 617 nozzle load cases.

Would this be right?

Thanks & regards,

Rienzie

It's common practice to ignore wind loads on equipment such as pumps and compressors, usually installed close to ground level in a plant, unless directly exposed. API loads are concerned with the operating condition. Seismic events will normally shut a plant down, and should be considered as exceptional. If you must use API loads in this case, try to agree a factor with the vendor, say 10x for seismic.

I would like to see more opinion on occasional load limits for equipment.

The only reference I have found so far is in the nuclear code and, if I recall correctly, it is focused on pumps that must remain in operation through the event. It was no where near 10x.

I appreciate your statement regarding vendor consultation, in a perfect world this conversation should take place BEFORE the job begins. Welding Research Council Bulletin 449 (Guidelines for the Design and Installtion of Pump Piping Systems) makes that point as well.

I don't want to hijack this thread so if one of you wish to begin a new one in response, please do. And Thanks.

In case the system is intended to be operational and leak free the seismic loads should be considered on the equipment.

Some can claim that the control systems are there to stop the system running and therefore there is no need to consider the seismic events. This statement to me is not totally correct if you need to use the equipment after the seismic events. If one can face the full investigation on the equipment including replacement or repairs before the start-up there is nothing to say. Most cases the pipe is a lot stronger than the equipment nozzle. The stress analists are asked to design the pipe to take the seismic events but not consider the same loads on the equipment. Do we think the equipment is replacable, the pipe is not?

The standards give the maximum allowable loads on the equipment nozzles for proper operation. Some say that there must be a margin for the seismic event on the equipment. There may or may not be. Who knows?
I would recommend anyone who does pipe stress analysis to consider the occasional loads not to sleep worried.

Ibrahim Demir

In my view, the best we can do with regard to seismic loads is to design for product containment. Seismic design is not a precise science by any means. Fault lines don't behave themselves and exhibit that precise 0.7g or whatever response spectrum that the design spec quotes.

In trying to reduce nozzle loads due to seismic forces there is a tendency to rely on the mathematical solution, add restraints and particularly snubbers, to get acceptable results. In fact, adding snubbers close to equipment is probably the worst thing to do since significant movement is needed to lock up, thus damage is already caused. Better to be practical, allow some flexibility and accept some big transient loads.

Thanks for all the contributions.

I found it difficult to get sufficient flexibilty to piping when dealing with very tight space restrictions on an Offshore Platform. Added to this was the clients "Basis of Design" which required Load Cases and Nozzle Loads to be considered for (a) "Maximum Design Temperature" (b) Minimum Design Temperature, rather than the maximum and minimum operating temeperature.

Regards,

Rienzie

Please note that the definition of "Maximum Design Temperature" means different things to we stress analysts than it does to, say, process engineers. See, for example, B31.3, 301.3. This essentially defines "maximum design temperature" as maximum fluid temperature, and makes no mention of any requirement to add a design allowance.

I have personally seen piping carrying saturated steam, from package boilers with no possibility to create any significant superheat, where the process "maximum design temperature" is 50-100 F higher than saturation temperature at the "maximum design pressure." (I have also seen water systems with "minimum design temperatures" below 32 F!!)

One of the big problems in the engineering world is the tendency of EVERYBODY in the design chain to add a conservatism factor. Nowhere has this impacted me quite so much as the client that chastized me for not adding a larger safety factor in a fluid flow calculation, when the calculated line loss of 75 feet of head broke down into 72 feet of elevation change and 3 feet of calculated friction loss. I guess he thought the end user equipment was capable of self-levitation.