Hello Everyone,

I am having a hardtime in obtaining g's for seismic loading in caesar input. can anybody please give me some thoughts on how/where to collect data regarding this informations?

my s/v gave me only the horizontal sole acceleration = 0.62m/s2.

this would greatly be appreciated. thanks.

>>> obtained from ubc code, i have used Fp=(((0.4(ap)(Sds)(Wp))/((Rp)(Ip)))*(1+2(z/h))

Fp=0.32Wp, hereby used 0.32 as g, where, Sds(design spectral acceleration)= 0.6 (horizontal sole accleleration)

I hope I am doing the right thing? please advice. thanks again.

Hello Ianpinoy,

It may be beneficial to you to download and read this document.

http://www.americanlifelinesalliance.org/pdf/Seismic_Design_and_Retrofit_of_Piping_Systems.pdf

Also, have you tried to look at previous postings via the "search" utility that this board offers?

Regards, John.

Hi John,

Thanks for the quick reply, I have been searching about this topic but seems like I am not so lucky finding it. But, I'll try to check this link you have provided and see if I can get some answers in it... I hope I would.

Thanks again.

Cheers!

Hello again,

The link is a very good reference John, thanks again!

But, my concern is if our project horizontal sole acceleration is the same as the project spectral acceleration (SDS) stated in IBC 2000 code. And if it is, then, SDS=0.6m/s2:

FP = [0.4 aP SDS W I / RP] (1 + 2 z/h)

SDS = Project Spectral acceleration for short period
I = importance factor (1.5)
W = weight
FP = horizontal load
aP = 1.0 for any piping system
RP = 3.5 for high deformability piping systems
z = height of attachment to structure
h = height of structure
g = weight load multiplier used for CAESAR II input

From the equation above: (z=h)

FP = 0.32W, therefore g = 0.32???

Can anybody please confirm if I am doing the right computation.

Thanks again.

I'm not going to check your math.

And without knowing your background and abilities, I cannot say if YOU are doing the right thing here.

An issue I have with that equation is the z/h. If you are at the top of a structure, that (1+2z/h) will triple your ground acceleration. If that structure is a tall building, I can see your g load rising. But what about a short structure? If it is very stiff horizontally, I think that acceleration increase is too great.

I am uncertain how to vary the g load as you travel up through the building with your piping. Do you change g load abruptly at the pipe midpoint between floors or is some sort of smooth transition required to make the numbers look right? Or, will using the maximum g load throughout produce conservative results?

Another important consideration in these seismic evaluations is the building drift. We have simplified calculations for building acceleration but the structure deflections are still in question. Maybe it's not an issue in a single building but if piping connects two "independent" structures, these deflections may control safe design.

Hi Dave,

Thanks for your comments.

Regarding the z/h issue, I assumed to take the maximum approach for the piping location/attachment to the structure which is in the safe side of considerations, that is why z/h=1.

But, I am not really sure if I should be using this Building code equation, I just can't find any other way to get this "g" (weight load multiplier used for CAESAR II input) based on the only given data for seismic considerations for our project which is Horizontal sole acceleration = 0.62m/s2. I am not even sure if Horizontal Sole Acceleration is the Design Spectral Acceleration on a short period of time at the project location.

Anyway, if somebody could enlighten me more about this topic it would be much appreciated.

Thanks again.

Pinoy,

You can refer to Anil Chopra's ( University of California at Berkeley) Dynamics of Structures ,a Prentice Hall Publication and probably Chapter 18 where the author has discussed Structural Dynamics and Building codes.Also as advised by Dave , study the effect of Story Drift also ( Probably Chapter 12, 13 and 14).

Regards

pinoy,

Ratio of Fp/Wp should be your horizontal accleration Gx and Gy(vertical accleration) = 0.2*SDS also Z/H = 1 is a resonable assumption.

Refer to ASCE 7.6 if you need clarification.

PS: The numbers do change if your parameters change. 0.32 for GX seems to be a bit higher.As said it all depends on the parameters. please check the seismic Group,Site Class, S1 and SS,

Reub

Hey Reub,

Thanks for your thoughts! I'll check ASCE 7.6 out, but, you know what, I shouldn't even looking at this things because its civil's job. huh... it's really hard to get money this days...

Dave,
I have got exactly the same issue with the way z/h should be considered. Just wondering if you have had clearer picture by now. Thanks.

To make a joke... I think the clear picture was always there, the problem is it seems to be an abstract Kandinsky picture and not all of us are ready to accept what represent.

Basically the first mode of vibration of a complex structure is one associated with oscillation between the extreme lateral deflections of a vertical cantilever. Intuitively is clear that the displacement in top is greater than the displacement in middle and because the oscillation in both top and middle are performed in the same time T, the acceleration amplitude in top must be greater than in middle. The relative location factor, intended to scale-up the ground acceleration is 1+2*z/h for ASCE and may be different in other Codes.

What ASCE 7 considers is a linear law between 0.4SDS at ground and 1.2SDS at top . Sounds 0.4SDS at ground realistic? In case SDS is about 2.5 times the peak ground acceleration, yes.
Is 1.2SDS acceleration at top realistic? Considering an oscillator at ground can see maximum SDS acceleration (depending on its natural period), to predict that a rigid placed at top structure will see 1.2SDS acceleration (and an oscillator in resonance with structure up to 2,5 times 1.2SDS) may be acceptable, why not trust the Code? But here the concern is: the structure may be quite rigid (and in this case the top structure will see also 0.4SDS) or flexible and in post-elastic behavior and in this case you are not sure 1.2SDs is underestimated or not.

In case you are not satisfied- ASCE 7 allows you to consider a formula based on the "i"th level acceleration form modal analysis. This would eliminate the concern as is your structure rigid or flexible, because the modal analysis will reveal the truth. Yes, somebody must perform the structural modal analysis to give you the input in case you are in doubt. When nobody will do it, just accept z/h correction!

Ideally is to have the "i"th level spectrum based on post-elastic calculation instead a single "i"th level acceleration and some software is able to give such stories spectra. When nobody will do it, just accept z/h correction!

Because such spectra construction is missing in ASCE, there is a "more engineering correction" with "ap" coefficient which is component amplification factor. "ap" is tabulated (ASCE tried to guess the behavior of your component) but there is also a NCEER graph (and a corresponding formula) based on the ratio Tp/T (period of vibration of your component vs period of vibration of the structure). More or less, this approach is intended to predict the effect of resonance between your pipe/ component with structure vibration.

Attached a paper from Web.

Regards.

Just a remark. In fact ASCE 7, in structural calculation, estimates the Seismic Base Shear (V) based on some points on the Response Spectrum graph and correlates V figure with T- fundamental period of structure; later asks for Vertical Distribution of V (in 12.8.2) -which is equivalent to estimate the acceleration "at any level". The latest is (12.8-12) formula based on a distribution on vertical by (hi)^k, where hi is the height from the base to the level "i" and "k" is an exponent related to the structure period. In fact that formula just tries to guess the first modal response.

What I try to say is ASCE 7, in structural calculation, is able to estimate more correctly the horizontal accelerations at level "i" based on the behavior - rigid or flexible- of the structure. This approach is not longer applicable in ASCE when asks for horizontal accelerations for components attached to the structure, where is replaced by the factor 1+2z/h applied to 0.4SDS.
I guess ASCE found 1+2z/h factor approach more practical (you don't need to know the details of the structural calculation, just h and SDS are enough), but in fact is too much simplified and quite conservative because fails to consider the behavior of the structure itself. ASCE 12.8-12 is more accurate for estimating the horizontal accelerations at level "i" but is not what ASCE asks for component attached to.

To conclude about the "clear picture" you've asked for: the fact the acceleration amplifies by height is a physical fact that is counted in different ways through sections of ASCE. Under ASCE you still have the freedom to estimate the level horizontal accelerations by a more accurate calculation, but in this approach you need to be involved/ to deeply collaborate with structural specialists that may be impractical for a stress specialist that needs to be in schedule with piping stress calculations. When this tentative approach fails, you need to consider 1+2z/h factor applied to 0.4SDS, as ASCE asks for.