Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
ETABS Floor seismic and wind acceleration 4
- Thread starter JohnB965
- Start date
-
1
- #2
slickdeals
Structural
ETABS cannot calculate dynamic wind forces. Wind loads generated in ETABS are based on code prescribed static forces based on the analytical method. In order to get forces from a dynamic analysis, a wind tunnel analysis is required.
As for seismic accelerations, you can get response spectrum diaphragm accelerations from Display --> Show Tables --> Displacement Data --> Story Accerations. The prerequisite is that you should have performed a RSA analysis.
As for seismic accelerations, you can get response spectrum diaphragm accelerations from Display --> Show Tables --> Displacement Data --> Story Accerations. The prerequisite is that you should have performed a RSA analysis.
-
1
- #3
slickdeals
Structural
You have to do a dynamic analysis + Response spectrum analysis for that option to be activated.
slickdeals
Structural
Which version of ETABS are you running? I am running 9.2.0
-
2
- #9
Stressed,
Thanks for your help. We were trying to get the floor acceleratio for a floor but i wasnt able to get that, still my ETABS is not showing that and I dont understand why. I have checked my RS cases again, I checked with some other engineers in my office also but still I am unable to extract floor acceleration.
Thanks for your help. We were trying to get the floor acceleratio for a floor but i wasnt able to get that, still my ETABS is not showing that and I dont understand why. I have checked my RS cases again, I checked with some other engineers in my office also but still I am unable to extract floor acceleration.
I am not sure about what you are using the accelerations for.
Are you designing parts/portion of the building or you checking acceleration for wind sensitivity? ETABS can not do the dynaminc wind response so take it the former.
In any case, you should be careful because the accelerations in the ETABS output are, as i understand, reported at the center of mass at each floor level. If the building has some torsional compornent, there may be larger acc at far corner of the building.
My two cents, let us know your opinion.
Are you designing parts/portion of the building or you checking acceleration for wind sensitivity? ETABS can not do the dynaminc wind response so take it the former.
In any case, you should be careful because the accelerations in the ETABS output are, as i understand, reported at the center of mass at each floor level. If the building has some torsional compornent, there may be larger acc at far corner of the building.
My two cents, let us know your opinion.
I'm pretty sure that the last two posters are mistaken in some of their assertions:
1) Accelerations versus period is a common LINEAR response spectrum analysis. You do not need to run a nonlinear analysis in order to obtain actual accelerations
2) I don't understand the comment from saburo stating that ETABS cannot perform dynamic wind response. We have used ETABS for this on buildings utilizing wind tunnel data, running dynamic time history for wind loading.
3) ETABS accounts for torsion.
4) ETABS give you the option to model the structure with rigid or flexible diaphram
1) Accelerations versus period is a common LINEAR response spectrum analysis. You do not need to run a nonlinear analysis in order to obtain actual accelerations
2) I don't understand the comment from saburo stating that ETABS cannot perform dynamic wind response. We have used ETABS for this on buildings utilizing wind tunnel data, running dynamic time history for wind loading.
3) ETABS accounts for torsion.
4) ETABS give you the option to model the structure with rigid or flexible diaphram
slickdeals
Structural
@Stressed:
Can you provide some more information on dynamic time history analysis for wind loading? Does this mean you get a wind pressure vs time data from the wind tunnel?
Seems rather odd, but maybe you can explain.
Can you provide some more information on dynamic time history analysis for wind loading? Does this mean you get a wind pressure vs time data from the wind tunnel?
Seems rather odd, but maybe you can explain.
slickdeals, why would you describe it as "odd"? Have you ever been involved with wind studies before, as time history wind loading as part of the study is commonplace? RWDI, as one company we've used, performs studies and lab tests including use of wind tunnels, then provides wind forces as a function of time to be used in the design since ASCE wind load calculations can be overly conservative in some cases. Wind studies can save money.
Any load, whether it's wind or seismic or whatever, can be analyzed as a dynamic time history, which it was so odd to read the post claiming that ETABS cannot do dynamic wind response.
Any load, whether it's wind or seismic or whatever, can be analyzed as a dynamic time history, which it was so odd to read the post claiming that ETABS cannot do dynamic wind response.
slickdeals
Structural
@Stressed,
Whenever RWDI has provided us with information on tall buildings they have typically provided us with equivalent floor by floor static loads and the combination factors to be used for a static test. I know that these results have been derived by them from tests on a HFFB and post processing them with the provided modal characteristics.
Whenever RWDI has provided us with information on tall buildings they have typically provided us with equivalent floor by floor static loads and the combination factors to be used for a static test. I know that these results have been derived by them from tests on a HFFB and post processing them with the provided modal characteristics.
Stressed, What I meant by "ETABS can not perform dynamic wind analysis" is that ETABS can not perform time history wind response analysis with time dependent wind pressure (each time step)around the building(building each face has different wind pressure/suction distribution at any given time and need to consider turbulent effect, neighboring building effect ground surface roughness etc too). If ETABS can perform that, we do not need to do the Wind Tunnel Testing. I pretty much doubt that we can make the wind data required for that analysis too(may be theoretically possible, but the data will be huge and ETABS is not made out for that. EQ time history ground record input is simply time vs acc).
Could you tell us what form of time history wind forces you were given by Wind Tunnel Testing and how you analyzed with ETABS? Are they segmented at each floor levels(sort of equivalent time dependent forces)? I am very interested to know the procedure and why you do wind dynamic time history analysis in the first place because we already know the answers from Wind Tunnel Testing. Are you guys trying to calibrate the stiffness differences between WT model and ETABS analysis? Matching torsional frequency in the WT model with ETABS would be difficult.
My experience is that(I am actually designing a tall building now), I get equivalent static forces(peak forces) from Wind Tunnel Testing Report and simply input them to ETABS as static forces. ETABS is used to get the max wind action. The strength requirement is relatively easy to be satisfied(you just design it) but dynamic acceleration response at serviceability state(usually compared with some ISO standard **milli-g criteria) is hard to obtain/evaluate from analysis. Wind Tunnel Testing is the only way to get these info in a reliable level, I understand.
About story acceleration that ETABS outputs, off course the values were derived with 3D effect but they are reported one value/level at CM considering a lumped mass system. So if building twists, near the edge of the floor plate away from CM may have larger acceleration. Assumption of rigid or semi rigid diaphragms does not matter. Answer may be slightly different between the two but still a "Storey Acceleration", right?
Sorry about my English.
Could you tell us what form of time history wind forces you were given by Wind Tunnel Testing and how you analyzed with ETABS? Are they segmented at each floor levels(sort of equivalent time dependent forces)? I am very interested to know the procedure and why you do wind dynamic time history analysis in the first place because we already know the answers from Wind Tunnel Testing. Are you guys trying to calibrate the stiffness differences between WT model and ETABS analysis? Matching torsional frequency in the WT model with ETABS would be difficult.
My experience is that(I am actually designing a tall building now), I get equivalent static forces(peak forces) from Wind Tunnel Testing Report and simply input them to ETABS as static forces. ETABS is used to get the max wind action. The strength requirement is relatively easy to be satisfied(you just design it) but dynamic acceleration response at serviceability state(usually compared with some ISO standard **milli-g criteria) is hard to obtain/evaluate from analysis. Wind Tunnel Testing is the only way to get these info in a reliable level, I understand.
About story acceleration that ETABS outputs, off course the values were derived with 3D effect but they are reported one value/level at CM considering a lumped mass system. So if building twists, near the edge of the floor plate away from CM may have larger acceleration. Assumption of rigid or semi rigid diaphragms does not matter. Answer may be slightly different between the two but still a "Storey Acceleration", right?
Sorry about my English.
You're right, HFFB is popular, and fine for many applications as long as you're ok with the limitations and assumptions with this method. My understanding is that load data measured at the base is extrapolated along the height of the building making a number of assumptions regarding idealized modes of vibration, ignoring coupled modes and ignoring phasing of wind loads. Accurate prediction of accelerations can be an issue with HFFB depending on the structure, and the static equivalent wind loads don't enable you to reliably explore some design alternatives. Also, HFFB correction factors seem black box.. maybe some of you have better insights.
It doesn't cost much more $$ or time to have a surface pressure measurement which gives you force vs time info at different levels, and the data is relatively easy to extract for time history analysis. Typically they divide the building wind tunnel model into 6 levels or so (multiple stories per level) with multiple pressure taps around each level in order to record pressure/suction distribution every 15 degrees (as one example) for a specified sampling rate and period of recording. You can then determine tributary areas based on pressure tap coordinates
ETABS can most definitely perform time history wind response analysis with different time dependent wind pressures around the bldg on different faces, and the structural model you have in ETABS will almost certainly be a more accurate structural representation than a scale model.
You seem to be confusing the generation of wind forces vs. time through test measurements, CFD, databases, etc. with the ability to properly analyze those loads. No one is claiming that ETABS can generate TH wind loads without data from an external source, only that ETABS can handle the TH analysis of time dependent forces. It's not nearly as involved and complicated as you seem to suggest, which is the point - it's not much more costly or time consuming go the surface pressure measurement route with wind TH analysis, and you'll have a more realistic model to explore design alternatives if you need to. You've already created a detailed analytical structural model for strength design that you can also use for dynamic analysis.
Regarding the time it takes to run the ETABS TH analyses, I'm going by memory here, but as I recall, for a similar number of TH load cases with a similar number of time steps for the same model, there is not much difference in run time between running a TH acceleration load for seismic, and TH loads for wind, even though the wind loading may have more load points. With wind TH loading, you'll be looking at different TH load cases for each angle between 0 and 180 degrees. Don't get me wrong, it still takes time to run large models, but the incremental time cost is not much different than running a TH acceleration for seismic.
If project budget is an issue, there is archived time history wind pressure data available from RWDI and NIST that you might be able to utilize.
It doesn't cost much more $$ or time to have a surface pressure measurement which gives you force vs time info at different levels, and the data is relatively easy to extract for time history analysis. Typically they divide the building wind tunnel model into 6 levels or so (multiple stories per level) with multiple pressure taps around each level in order to record pressure/suction distribution every 15 degrees (as one example) for a specified sampling rate and period of recording. You can then determine tributary areas based on pressure tap coordinates
"ETABS can not perform time history wind response analysis with time dependent wind pressure (each time step)around the building(building each face has different wind pressure/suction distribution at any given time and need to consider turbulent effect, neighboring building effect ground surface roughness etc too)."
ETABS can most definitely perform time history wind response analysis with different time dependent wind pressures around the bldg on different faces, and the structural model you have in ETABS will almost certainly be a more accurate structural representation than a scale model.
You seem to be confusing the generation of wind forces vs. time through test measurements, CFD, databases, etc. with the ability to properly analyze those loads. No one is claiming that ETABS can generate TH wind loads without data from an external source, only that ETABS can handle the TH analysis of time dependent forces. It's not nearly as involved and complicated as you seem to suggest, which is the point - it's not much more costly or time consuming go the surface pressure measurement route with wind TH analysis, and you'll have a more realistic model to explore design alternatives if you need to. You've already created a detailed analytical structural model for strength design that you can also use for dynamic analysis.
Regarding the time it takes to run the ETABS TH analyses, I'm going by memory here, but as I recall, for a similar number of TH load cases with a similar number of time steps for the same model, there is not much difference in run time between running a TH acceleration load for seismic, and TH loads for wind, even though the wind loading may have more load points. With wind TH loading, you'll be looking at different TH load cases for each angle between 0 and 180 degrees. Don't get me wrong, it still takes time to run large models, but the incremental time cost is not much different than running a TH acceleration for seismic.
If project budget is an issue, there is archived time history wind pressure data available from RWDI and NIST that you might be able to utilize.
Stressed
Thank you very much for your input and sorry about my lack of study!! It seems you guys are well ahead of us. I was aware that ETABS can not generate dynamic wind loads for sure but did not know ETABS can handle TH wind dynamic analysis to be honest. I only have some experience with WTT similar to what slickdeals expressed.
I am working on very tall/large building and it takes 14 min or so to just run some static cases(I know it all depends on how we effectively model and machine capability etc).Just wondered how long would it take to do the TH dynamic wind analysis. Can we not do something better with frequency domain?
Thank you very much for your input and sorry about my lack of study!! It seems you guys are well ahead of us. I was aware that ETABS can not generate dynamic wind loads for sure but did not know ETABS can handle TH wind dynamic analysis to be honest. I only have some experience with WTT similar to what slickdeals expressed.
I am working on very tall/large building and it takes 14 min or so to just run some static cases(I know it all depends on how we effectively model and machine capability etc).Just wondered how long would it take to do the TH dynamic wind analysis. Can we not do something better with frequency domain?
It's my understanding that many/most highrise projects over the past several years have had the surface pressure model in the wind tunnel, usually accompanied by HFFB for validation. Two different scale models, one for HTTB and another for surface pressure taps.
The only logic I can see to emphasize the frequency domain loading in such highrise structures if there is pressure measurement data also available might possibly be the runtime savings.. but even then I'm not convinced there would be that much time savings without familiarity with structure being analyzed. And you would lose a more accurate, more versatile time domain design tool by doing so.
For example, it's not so uncommon to have a 35 story concrete building under static and and multiple time history loading run in under 15 minutes, and doing so with a relatively average PC. Other models can take longer, but runtimes under 1 hour for 35+ story structures with time history loading is not uncommon.
The only logic I can see to emphasize the frequency domain loading in such highrise structures if there is pressure measurement data also available might possibly be the runtime savings.. but even then I'm not convinced there would be that much time savings without familiarity with structure being analyzed. And you would lose a more accurate, more versatile time domain design tool by doing so.
For example, it's not so uncommon to have a 35 story concrete building under static and and multiple time history loading run in under 15 minutes, and doing so with a relatively average PC. Other models can take longer, but runtimes under 1 hour for 35+ story structures with time history loading is not uncommon.
- Status
Similar threads
- Locked
- Question