I am curious how folks are handling the section within ASCE 7 stating that flexibility in piping connections to tanks and vessels needs to exist in such a way to allow many inches of displacement (exact amount depends on specific design) in all 6 directions and not allow rupture in the piping. This section is nearly identical to API 650 Annex E. API 650 Annex E is a suggested practice (and we've seen most clients choose to not apply this to piping), whereas ASCE-7 is generally adopted as law in most US states to my knowledge. Are most folks running this displacement rupture analysis on all piping connected to their tanks and vessels?


Per ASCE 7-16 section 15.7.4 "Flexibility of Piping Attachments":


The piping system and tank connection shall also be designed to tolerate Cd times the displacements given in Table 15.7-1 without rupture, although permanent deformations and inelastic behavior in the piping supports and tank shell are permitted.

I can also confirm this table has been in this specification since or before 2005.

Having the vast majority of my onshore work come from a non-seismic region, even by ASCE-07 standards, no, this is generally deemed to not apply to my work.

Have I ever seen anyone else trying plugging 12" of displacement into CAESAR and back check against UTS? Also no. I've heard of it mentioned, although seemingly derisively.

What I would suggest, though, is that the following statement is our "get out of design specification hell" card.



My offshore work utilizes CSA-calculated accelerations.

Thank you, Michael for the response.

I agree with you in practical sense with everything you say. But alas, ASCE-7 is adopted as law in most states I can think of. And the code doesn't differentiate the requirement for piping flexibility at piping to tank & vessel connections depending on seismic zone. I'm looking for get out of specification hell cards left and right!

I tend to think the "Unless otherwise calculated, the minimum displacements in Table 15.7-1 shall be assumed." as not a way out, but as a way to actually increase the displacements if additional info is available (like a soils report). Piping to nozzle connection displacements differs from calculated accelerations. The accelerations would apply to ASCE-7 section 13 for piping, right? Calculating seismic movement in equipment is a specialty and not sure I've seen anything except a settlement value given leaving the other 5 directions left to be pulled from the table. I have no issues running seismic analysis per section 13 with accelerations in my models. Section 15 is specific to the piping to equipment connection and the movement created in the piping as a result. As you seem to also be aware, some of these displacements are enormous and every facility I've ever seen obviously appears to be completely designed without applying the requirements of Table 15.7-1. Which I totally understand. It is generally not feasible, economical, nor practical to design every single connection to a vessel or tank to these measures. Hence, my question since it is considered law. So I'm curious how others in the industry view this topic.

By the way, UTS = Ultimate Strength I presume? Again, thank you!

Few years ago I tried to understand API 650 seismic section basis.
For self-anchored tanks, the bases are related to a "double-hinged model" of the annular bottom. It may be understood as a limit of the theoretical collapse (by developing two plastic hinges) of the lifted part of the bottom.
The point is this "state" of the bottom is not really seen as a response to a set of the seismic accelerations and has no connection to a particular design spectrum or seismic event. It is a limit state for a calculation focused on tank overturning (and an increased compression in a part of shell) under a design seismic event. Doesn't means plastic double hinges would appear in bottom for the design seismic event.

For me the intention of API 650 "yu" calculation appears strange because was linked directly with "double hinged model" rather to a design seismic event. IMO, event the intention failed; I was able to reconstruct, formula by formula, the results of Appendix E but not "yu".

So "yu" can be understood as the maximum displacement of shell before the theoretical collapse of the bottom will occur (interpretation that explains why there is no need to be multiplied by 1.4Cd).
In API 650 "yu" is just wrong calculated within the "double hinged model" and a corrected formula should give results reduced by a factor more than 2. I mean the coefficient in formula doesn't meet even the theory! And I guess (with no evidence) that the table giving displacements was a tentative to offer more acceptable results adjusting results of a "theory"/formula which is just unrealistic.

Now, trying to say something about your question... which is the relevance to calculate piping attached to a tank when the bottom is ready to develop plastic double-hinges, in an unpredictable seismic event? Or which is the relevance to use as imposed displacement for piping the tabulated values based or not on "yu", anyway based on obscure criteria?
Frankly, I don't know.

Thanks Mariog for your response. My specific question to folks is are you considering ASCE 7 Section 15.7.4 every single connection to vessels and tanks in states/locations that have adopted ASCE 7 as law?

It is interesting to me that there's not a more specific direction of this requirement in ASME B31.3 and ASME VIII Div 1.

I disagree. If you were to pick the location in America with the worst seismic events, and then use the worst possible "normal" piping arrangement, this singular chart is supposed to cover that much... and then some more. My opinion is "unless otherwise calculated" here means "you take responsibility for determining better/useful/realistic numbers."

All our codes inevitably push some amount of "engineering judgement" onto the engineer at some point.



Unless a crack in the ground opens up between the tank and the first support, it's problematic to think that the original input to the free body diagram should be a displacement.

Instead, the displacement itself is either:
a) a displacement caused by a singular thrust of acceleration of the entire system and
i) can itself be calculated and then input into caesar or
ii) can simply be included into CAESAR and have CAESAR calculate the displacement
b) a final or worst case scenario location of where the pipe lands after an extended sequence of random thrusts and vibrations during a seismic event. The only way I've seen this addressed is through a dynamic study where you physically input seismograph data into CAESAR.



I can't say at what point in which state it became law or what point ASCE-07 included this table. Some simply become grandfathered in.

Those that did try to comply with either these or other onerous requirements frequently circumvented the problem via flex hoses, that may or may not have been replaced by hard piping after the flex hose expired.



Yes. This would be my first "sanity check" for "won't rupture." If the stresses exceed UTS, then we can say without hesitation that this pipe or nozzle should not be able to withstand this scenario. Brand new? Perfectly installed? Not even one single time.

Of course, simply exceeding YS means you're likely dealing with material that needs to be replaced.

Michael, I really appreciate the feedback.

"If you were to pick the location in America with the worst seismic events, and then use the worst possible "normal" piping arrangement, this singular chart is supposed to cover that much... and then some more."

This isn't exactly accurate. We are dealing with a client right now where seismic displacements are determined to create displacements above and beyond this table and seismic vertical settlement are orders of magnitude more than this table. This is per a soils report from a geologist. The soil is extremely poor and it is a high seismic zone.

"All our codes inevitably push some amount of "engineering judgement" onto the engineer at some point."

I agree, although in this instance I just can't convince myself this section is open to engineering judgement. It reads completely black and white to me and as a law that leaves little room for interpretation in my opinion. If this were a standard and not code adopted as law, then I would certainly pursue the engineering judgement rule here. But I'm having difficulty finding any amount of grey area as much as I think it makes sense to. Thanks again for your input!

Then I would be inclined to respect the end user's trust of the geologist and recommend moving towards ball joints or flex joints to accommodate, even if their application within CAESAR is a challenge, as hard piping will be difficult to manage here.