Pipe lift off in dynamic analysis

[alfmar]

Dear all,

This topic may have been already replied somewhere, but I cannot find it.

Since dynamic analysis is strictly linear, and the results provided for ABS combination are always positive, how can we perform a lift off check in such type of analysis?

Very grateful for your replies.

Best regards

 

[saplanti]

Check the mode shapes from the natural frequency analysis. You will see if the pipe separated from the supports.

 

[alfmar]

Thank you Saplanti.

This is a good procedure to infer where the lift may be, however modal analysis does not provide magnitude or the one provided is not real. Then we would need an actual magnitude of displacement.

I suggest to compare the static sustained load case load at each restrain with that of the seismic shock. In case the later is higher, it is sure we will have lifting off.

Best regards,

 

[DSB123]

Is the dynamic load greater than the selfweight of the pipe reating on the support? If it is then you will have lift off. The extent of the lift off is difficult to ascertain but if you have a large dynamic load and the weight load is small then expect significant lift off. Where lift off occurs (if significant) unless you design the supports to prevent lift off then you need to design them for impact loading for when the pipe "slams" back down onto the support!!!

 

[saplanti]

If your design is new you must be targeting certain stiffness and natural frequency. Therefore the mode shapes will give you a pretty good idea where you should use restraints against uplift in addition to others. The modes shape are not with actual magnitude, but the separation (without actual magnitude) from the supports are real.

If you are dealing with an existing piping DSB123 has given you full description where to check.

 

[ZippyDDoodah]

Actually, DSB's advice was not a "full description" of what to check, but a general word of advice as to what to watch for.

If nonlinear supports in the piping system are defined with gaps and friction which is common, dynamic load distribution and reactions cannot be reliably determined with a linear dynamic analysis. It's as simple as that. Results using linear dynamic analysis would be unreliable.

 

[saplanti]

ZippyDDoodah, I guess you generalized the problem on all types supports and too far, however missed the concern was for uplift only.

If we extend the discussion we may start asking the analysis with shell or brick elements not beam element as used by pipe stress analysis software, and do not take local plasticity into account. But these are not the questions of this post and additionally this problem and your concerns are also not a big deal in new designs. In case there is a dynamic load concern you can provide the piping with adequate span lengths and supports (you are the decision maker on supports) as required by following the software user guide on dynamic load and analysis types. I think we agree on that.

For the piping systems in operation with the introduction of new dynamic loads the solution is not easy, we may not be able to change the existing system, and your argument is valid. It requires a lot of site measurement (accelerations, displacements, perhaps the mode shape of the system), and comparing/equalizing with the analysis results by iterations in the analysis takes a long time and effort. The effect of nonlinear supports makes everything more difficult, probably without a proper solution with these tools as you mentioned. So you need to consult the dynamics expert to make the job for you.

In most cases, making the problem solvable with the existing tools are the job of the pipe stress engineer as you may aware. If we think we cannot solve the problem with the available tools in hand we can ask for expert help and leave the job to them.

Hope this clarifies your concerns.

 

[DSB123]

ZippyDDoodah,
My oh my what a wonderful day. Where in the OP's question does it mention gaps and friction? The answer is no-where. I specifically answered the OP's question and did not bring into the equation phenonomen which were not mentioned. Why for instance did you not cover spring supports under dynamic loading as well as gaps and friction? I tend to try and answer the specific question not bullshit and include other aspects!!

 

[ZippyDDoodah]

"ZippyDDoodah, I guess you generalized the problem on all types supports and too far, however missed the concern was for uplift only"

saplanti, by definition, the ability for the piping to "uplift" means that the support is nonlinear, and therefore a nonlinear analysis is required if the possibility of uplift is to be considered. If the pipe is supported in -Y direction, but not in +Y at a support location.. that's nonlinear analysis. No need to muddy the conversation and add confusion with talk about brick and shell elements or material plasticity, we're talking only in the context of pipe/line "stick" elements with the possibility of uplift.. and what you described was clearly nonlinear analysis for dynamic loading, whether you realize it or not.

My statements weren't controversial in the least and my point still stands. With nonlinear pipe supports with gaps for uplift and/or friction (and nonlinear gaps WOULD be there if uplift is possible, a +Y support IS nonlinear), then ANY linear dynamic analysis with nonlinear supports would be unreliable, period. With a linear dynamic analysis of nonlinear boundary conditions, you couldn't reliably determine what the load distribution would be to even begin to know whether dynamic loading would cause uplift.

Not sure why DSB feels the need to respond so aggressively. DSB repeats "Oh my what a wonderful day" over and over in multiple threads in an attempt to mock me.. which is fine.

"Where in the OP's question does it mention gaps and friction?...I tend to try and answer the specific question not bullshit and include other aspects!!"

My, what a classy and professional reply. I've answered this already. You can thank me later for educating you.

 

[saplanti]

ZippyDDoodah, my second paragraph comment is still valid. I guess you are confused with non-linearity definition of the static analysis.

If you use all the support are uplift free you will see that the natural frequency of the piping system will be very small, you need support against uplift at certain locations in addition to the support in other directions and locations to be able to bring the natural frequency to the reasonable level for the purpose of dynamic analysis. This does not mean that all the support must have a support against uplift, and the system is still linear for the dynamic analysis.

If you read my answers I have never talked about friction or uplift supports with gaps, and the post was not asking for them either. Please do not pull the conversation/discussion into different subjects.

I hope this clarifies everything. Regards.

 

[ZippyDDoodah]

"If you read my answers I have never talked about friction or uplift supports with gaps, and the post was not asking for them either. Please do not pull the conversation/discussion into different subjects."

Perhaps you missed the title of this post: Pipe lift off in dynamic analysis

You advised the poster to check for liftoff using LINEAR modal results. Liftoff is a NONLINEAR phenomenon which involves "gaps". You can't reliably check for nonlinear liftoff during dynamic analysis using linear modal or linear time history dynamic analysis. If gaps such as +Y supports are important enough to model in static analysis models, then they are important to consider in dynamic loading too.

Your assertion that I am somehow attempting to "pull" the conversation away from the stated topic of discussion is laughable, especially given that YOU are the one doing the pulling by bringing up off-topic issues about shells, solids, and material plasticity which have zero to do with the main discussion.

 

[saplanti]

ZippyDDoodah, I guess you do not want to read and digest the posts above and go into discussions after. I suggest you to read the entire posts.

If you design and manage the new design model adequately there is no phenomenon in the dynamic analysis unless you create one.
DSB123 gave you the answer in his first post, and you are not satisfied. Have you ever heard the load case combinations between static and dynamic analysis? I understand your claim if the load combinations still shows the uplift at certain supports. In this case the designer should go back and modify the model.
Since we started and open discussion, can you please explain here where and why you consider uplift supports with gap and friction if you are supposed to run a dynamic analysis?

 

[DSB123]

ZippyDDoodah,
I/we seem to have hit a raw nerve here. My response was sufficient. As you did not add anything over and above my response apart from bringing other apsects into the equation, (which was not required) why respond?? Take a "chill pill" as I was not mocking you just a little light hearted fun.

 

[ZippyDDoodah]

"Since we started and open discussion, can you please explain here where and why you consider uplift supports with gap and friction if you are supposed to run a dynamic analysis?"

You seem quite unwilling to read and comprehend. It DOESN'T MATTER whether the analysis is static or dynamic. If uplift is a concern for static analysis, then it SHOULD ALSO be a concern for dynamic analysis, which is presumably why alfmar started this thread. That basic point should not be so difficult for you to understand.

It makes no sense to linearize nonlinear supports such as +Y and analyze them as hold-down supports for dynamic analysis, which is exactly what we are forced to do in Caesar II and other pipe stress programs. If a pipe is sitting on a beam or a shoe or a saddle or any support where it can liftoff, what sense does it make for the program to CHANGE that support to a hold-down for dynamic analysis? Linearizing nonlinear supports for dynamic analysis is a dubious, unreliable approach.. yet that is our only alternative with Caesar II, because Caesar II and the other pipe stress programs on the market are incapable of any type of nonlinear dynamic analysis. You gave dangerous advice when you suggested that linear modal analysis would be an acceptable means of checking for uplift in a system with nonlinear pipe supports.

Gaps and friction are included in pipe stress models ALL THE TIME. Uplift and resulting load redistribution is a common concern. Did it not occur to you and DSB that if these nonlinear supports were so important for static analysis, that it makes zero sense to have them linearized for dynamic analysis? Another question, a question which your clients should also be able to ask you - why did this reality never occur to you before, and why are you are so stubbornly refusing to acknowledge it now?

Unless you're completely out of your depth, there's nothing I'm saying which you shouldn't have already known was true.

"I was not mocking you just a little light hearted fun."

DSB, given that you repeated the same verbatim mocking comment (can you at least be creative enough not to repeat the same exact material?) in the pump thread coupled with an insult toward my post telling me that my post "added nothing", an insult you're repeating here again, there can be no honest doubt that you're not simply having a little "light hearted fun", but instead you were/are badgering and acting obnoxious, as anyone reading here can see for themselves.

 

[nipra03]

Zippy............... Well done !!!!!!

very logical answer.
Some of these guys are so taken up with their half baked concepts they confuse themselves and others.

I wish you could contribute more often on these sites

 

[saplanti]

ZippyDDoodah, I think you really underestimate the capability of available software of the pipe stress analysis, and I started to believe that you do not have a clue what to do when the piping system has dynamic loads after reading your unnecessary statements.

I suggest you to read the attached extraction of Caasar II User's Manual, and you may come up a conclusion why and when the you need to use your judgement with the results and decide iteration in the analysis or stop. I am not going into more discussions on this subject, this forum is for tips not for teaching.
Additionally, this is not a blame game on the computer, please do not insult others, try to understand their positions. Sometimes you may not understand why it was done this way, so you can ask your question to the software provider and get the answer if you were not satisfied by the documentation provided with the software.

Please do not forget that the software is a tool to achieve a reliable result for your problem. In case a particular software does not answer your question, it is not your tool to use find another one. If the tool that you are after is not available, create one so we all can respect you. Or involve and make reasonable suggestions to the software providers to make the piping analysis more reliable. I believe they would not return your reasonable suggestions. But DO NOT ACCUSE or BLAME, we are all human.

 

[ZippyDDoodah]

If the tool that you are after is not available, create one so we all can respect you.

That has to qualify as one of the most bizarre comments I've ever read in an eng-tips forum.

 

[saplanti]

ZippyDDoodah, I see that you have behavioral problem, but I assume that you are young, and inexperienced in social life such as this forum. Therefore I do not want to extent my discussion any more in this post.

 

[ZippyDDoodah]

ZippyDDoodah, I see that you have behavioral problem, but I assume that you are young, and inexperienced in social life such as this forum.

You gave dangerous, incorrect advice to the person who started this thread as has already been explained in detail, whether you care to acknowledge your errors or not. When called out on your misguided advice and wrongheaded understanding of the subject matter, instead of admitting that you didn't have a clear understanding of what you were talking about, you're now reduced to hurling insults at me ("you obviously have behavioral problems") like a 10 year old child instead of adding constructive input for other engineers. That sort of comment defines the type person you are, and it drags the discourse down to your level.

Hopefully others will take your statements on this thread including your denial of facts into consideration before listening to ANY other advice you may have to offer on this or any other forum. You're simply out of your depth, and now you're lashing out now that someone has pointed a finger at the inconsistencies with your misguided statements. That's not an insult, it's fact. I teach my teenage children to take responsibility for their actions, especially when they're wrong. You should do the same.

 

[DSB123]

Saplanti,
Spot on. Better watch out for the "dummys being thrown out of the pram"!!!!