Seismic acceleration factor using IBC simplified method

[vishkan]

Dear Experts,
I am pretty new to this structural engineering world, and now only i am facing real-world problems. I really need your helps and advises from your long term design experiences.
I am involved in a structural design of steel structure is build in the USA.
I just now involved in seismic analysis for my FEM model. According to the client information Ss = 2.51 g, and S1 = 0.654 g
with site class D.
I have followed the simplified method given in IBC using the equation.
V = 1.2 * Sds/ R * W;
From this equation 1.2 * Sds/R = Ah = acceleration factor.
This is the acceleration factor i give to my FEM program to calculate the horizontal seismic load.
According to the IBC standard, when i start to calculate the Sds;
Sds = 2/3 * Sms
Sms = Fa * Ss; Fa = 1.0 (for site class D) and Ss = 2.51
Then my Sds = 1.67,
then my Ah = 1.2 * 1.67/R ; here I have take the response modification factor according to the client concern as 3;
So finally i end up with Ah = 0.67 !!!!!!!!!!!!!!!!!!!!!!!!!!

Now i am asking my questions....
Can you believe this much of value for seismic acceleration factor? because when i calculate for my structure with the self-weight plus operational load with 12,000 kN , i end up with a seismic load around 8000 kN....
Because according to the equation V = Ah * W ; here W is effective seismic weight
so V = 0.67 * 12000 = around 8000 kN..
My kind question is, i am working in a German system with m,kg , s...
is there any role for this g in the seismic detail (2.51 g and 0.654 g)
I really did not consider the g.. the calculated values from US seismic data only for acceleration with ft/s2 ?
do we need any conversion?
The effective weight of the structure W; can i take it in N? I believe this is just a weight, according to my fundamental, weight is in N, some people confused me that this can be in kg according to american usage -like mass,
but in the IBC standard it is clearly given that effective seismic weight of structure...
I am looking forward for your suggestions, please correct me where i am going wrong..
I am thanking you in advance for your help..
greets from Germany
Vishva

 

[RobertHale]

1. Yes, I can believe such an acceleration factor. Think about it this way at 5 Hz (a moderate close approximation of the building's fundamental period) the ground is accelerating at about 2.5 times gravity in the lateral direction. At least you get to reduce that force because of building height/response and inelastic behavior.

2. In glancing though the Response coefficient table there is not a a system with an R=3 that is permitted under the Seismic Design Category D, which your spectral response accelerations place the building in. Typically it is incumbent upon the structural engineer to select the appropriate system from the table based on the design intent and site location. You would do well to verify that you meet the SDC limitations, you meet the height limitations, and you know what the detailing requirements are for the system you select.

3. Your use of Force (kN) with the acceleration parameter is consistent with the method. The coefficient you calculate is a multiple of the gravitational acceleration. Your building weight (Mass x Gravitational acceleration) has the "g" built into it so they cancel out.

4. I think most structural engineers (at least in my area) completely skip trying the simplifed method and just go to Equivalent Lateral Force Method. From my memory, there are a lot of onerous checks in that you have to perform to make sure you can apply the method, and you don't have to do those checks if you use the equivalent lateral force method.

 

[vishkan]

Dear RobertHale,
Thank you very much for your response.
Your answers and words from your experience cleared me most of my doubts. Actually I had the feeling too that we do not need to bother about this gravitational acceleration "g". This is also in the FEM program already included.
I would like to clear one more point.. please you or any other experts can advice me how should i proceed....
My question is...
If i adopt Equivalent lateral force Procedure (ASCE 7), I need the fundamental period T of the structure...
In deciding this T, i have many question always.. Because my structure has various parts such as under construction, operational parts and equipments and roof structures and so on... and there are many parts hanging on the structure too.. they go for pendulum oscillation ... In deciding the correct eigen frequency i have really headache...
For my understanding, in a real FEM structural analysis, no one create a FE model exactly like the real structure, there are so many assumption and and approximations. in this case how can i look for exact eigen frequency of the whole structure...
Can you advice me, which eigen modes are very important? In general the important part i consider the under construction, because which is taking the horizontal ground acceleration, this structure is very stable and giving in general high frequency. But at the same time, the upper part of the structure.. example roof and roof bearing structures are very flexible,they have small frequencies in general..
in this situation how can go for a suitable eigen frequency.

I have just now tried with equivalent lateral force method...
I have used the approximate method to determine the fundemental periond of my 21.9 m tall structure.I got a value of Ta = 0.49 s goes around 2 Hz
Can i take this values as T? Do not want to have a FEM model analysis for eigen frequency?
For the same seismic conditions, when i calculate using this T value, i end up with Cs = 0.39 g...
this is very less than the simplified method value...
What you suggest me... this is not a building structure, kind of industrial steel structure.. my suspect is.. how can i go with a suitable fundamental period..
Could you please share me your experiences and knowledge in this regard..
Thanking you in advance
Vishva

 

[RobertHale]

Without more information on the use and proposed construction of the building I can only comment generally. Typically, I will consider any superimposed dead load as a smeared area load. This changes if anything is "unusual" by my eye (e.g. water tanks concentrated in a specific location, large pipe loads, etc.) The ELF Method assumes the structure's response will be dominated by the response of the modes at the fundamental frequency, i.e. the frequencies associated with a mode shape of [1,1,1] in the X, [1,1,1] in the Z, and [1,1,1] in the rotation about the vertical Y. This is the purpose of checking for what the code refers to as "Irregularities." Those irregularities typically mean that the response of a higher frequency mode will be significant.

If you have already developed an analytical model then it should be a simple task to identify the modes I listed above and pull that frequency/period out of the model. Be careful to observe that most codes provide an upper limit to the period calculation for strength limit states.

With industrial structures, I would be very careful about applying either ELF or the simplified method as both assume some time of diaphragm to collect the force and deliver it to the Lateral force resisting elements. Without those diaphragms the structure is likely to behave like a distributed mass system rather than the discreet mass system ELF assumes.

 

[vishkan]

Thank you.. I got some points..your answers made me some more readings..