Verification of modal analysis Results

[mecheil2000]

Why If I solve one small model (concrete building 10 floors) by using Etabs and the same building By Using robot I see the different results for modal analysis from each program
If the basics and concepts are the same in each one of them then why we get different results.

I hope any one can answer my question

[medo]

Please post more details about structure model and some details about the input and output from modal analysis.

[mecheil2000]

Please post more details about structure model and some details about the input and output from modal analysis.

thanks for your helping
But did you model any building by robot and Etabs and compare dynamic analysis result?
I will try to post the model

once again I have to thank you for your post and assistance

[troy_the.best]

My model etabs v9.7.1 and RSA2011 sp2
results are same. ıts perfect run.
no problem modal analysis. bye

falconado

[mecheil2000]

My model etabs v9.7.1 and RSA2011 sp2
results are same. ıts perfect run.
no problem modal analysis. bye

falconado
I found one thing
Sometimes when I use Riigid diaphragm With Salb (Flexible Rigid Diaphragm XY) I got this problem , I do not know why
I think rigid diaphragm some times creats a problem with dynamic analysis

[troy_the.best]

Yes you are right. when use rigid diaphram or semi rigid diaph. its given wrong results. İ hope from autodesk this bugs is fix on next Sp3 patch. Good works.

[mecheil2000]

Yes you are right. when use rigid diaphram or semi rigid diaph. its given wrong results. İ hope from autodesk this bugs is fix on next Sp3 patch. Good works.

Robot with a lot of Bugs!!!
See, when I try to use Seismic Loads by Using UBC code
and applied the value of base Shear say Vx = 1000 KN,
I found that robot does not multiply the results with a scale factor to be matched with the new values

For example :
Static seismic load say = 1000 KN
Dynamic seismic load say = 500 KN
now robot should Multiply the Seismic dynamic load by factor = 1000/500 =2
this as per UBC because dynamic loads should not be less than static load
But Robot did not do that, and this is also one more Bug there.
Thank you

[troy_the.best]

mechiel,
look this jpg UBC scale factor in robot. have a nice days.

falconado_turkey

[mecheil2000]

mechiel,
look this jpg UBC scale factor in robot. have a nice days.

falconado_turkey
Hi falconado,
thank you for your helping,
But this is not hte scale factor,
this is the direction of ritz vector
when we apply the seismic in X direction we put X=1
when we apply seismic in y dir we put Y=1

After that we use a respondse spectrum curve UBC
then program we calculate the seismic forces in X and Y
then we will get Vx= 1000 say (Dynamic) ; and Vx = 500 (Static) say
so all results of dyanimc seismic should be multiply by scale factor =0.5 = Vstatic/Vdyn

Regards

[length zero]

Mech
Maybe it can help you
When u use directional X=2 it means that you have doubled seismic force in that direction

[mecheil2000]

Mech
Maybe it can help you
When u use directional X=2 it means that you have doubled seismic force in that direction

Hi Z.G
X ,y and Z are only vectors not values
it is like i, j, K in vectors
X means the program will put the mass in X- direction
and Y and Z same Things

[troy_the.best]

Edward and Length zero
Which you are right?
where is the scale factor of seismic loads. Please can you show in jpg. screenshot? I'm so confused. thanks

[length zero]

If u need to define seismic regarding to EC - i can helep, and you dont need scale factor
If u need to perform via UBC - mecheil can help you
Regards
Z.G. ( Lengh Zero)

[length zero]

Mechil
In EC i use this values to determine direction of seismic excitation,
And if i want to use 120% of seismicX loads and 35% of seismic Y loads, you must put X=1.2 and Y=0.35, ( this mean in this case you have apply 1.2SX+0.35SY - in one case)
i MEAN SO : Put 1 on X and read Base shear And for example 100kN, Go again and put in x=2 and you will get base shear 200nK, and i mean this you can use to scaling forces by seismic excitation, And maybe this is an SCALE FACTOR to EC, ahhah
Regards
Z.G.

[mecheil2000]

Edward and Length zero
Which you are right?
where is the scale factor of seismic loads. Please can you show in jpg. screenshot? I'm so confused. thanks

Hi troy,
see follow this steps to show meaning of x,y,z directions byspectrum analysis by ubc code
1-go to menu analysis
2-select : analysis type
3-one form will open and you will see seismic ubc x .....
(these cases you defined before)
4-select seismic ubc _x and under the window click on parameters
5-you will get new window click on direction
6-u will get new window contains X Y and Z
when you type 1 in X this means the program will generate seismic forces in x direction
when you type 0 in x and 1=y and 1=z this means the program will generatw forces in y directiona+z direction and so on

usually we will put for seismic_x we put x=1 y=0 z=0
for seismic_y we put x=0 ;y=1; z=0
and if you would like to generate one seismic load to consider x+y we put in this case
x=1 ;y=1
hoping this clear

[troy_the.best]

Edward,

yes you are righ but It means when we say the same things.

direction in robot=scale factor in etabs

isnt it?

best regards
falconado_turkey

[mecheil2000]

Edward,

yes you are righ but It means when we say the same things.

direction in robot=scale factor in etabs

isnt it?

best regards
falconado_turkey
See falconado, Now I try one other thing
Forget now scale factor ok
run the seismic analysis ok.
get the base shear from dynamic analysis say 1000 KN ok and say the name of case is Seismic_x
calcultae the static base shear say 500 KN.
now go to combinations and make a new case say Comb1 and use the case: seismic_x multiply into combination scale = 500/1000 = 0.5
Now you get the same idea of scale factor
but when you start design do not use Seismic_x but use Comb1
Hoping this is good

[mecheil2000]

Mechil
In EC i use this values to determine direction of seismic excitation,
And if i want to use 120% of seismicX loads and 35% of seismic Y loads, you must put X=1.2 and Y=0.35, ( this mean in this case you have apply 1.2SX+0.35SY - in one case)
i MEAN SO : Put 1 on X and read Base shear And for example 100kN, Go again and put in x=2 and you will get base shear 200nK, and i mean this you can use to scaling forces by seismic excitation, And maybe this is an SCALE FACTOR to EC, ahhah
Regards
Z.G.
Z.G thanks for your post
But believe me x,y,z is a only a direction vector in masses by using ubc
x=1 means but forces of masses from newton equations in x-direction
x=2 same thing
but yuor concept is valid only when you apply x=1 and y=2
this means you have a vector force F=1i +2j
then the program will prepare a new unit vector
hoping you get my point

[troy_the.best]

Z.G thanks for your post
But believe me x,y,z is a only a direction vector in masses by using ubc
x=1 means but forces of masses from newton equations in x-direction
x=2 same thing
but yuor concept is valid only when you apply x=1 and y=2
this means you have a vector force F=1i +2j
then the program will prepare a new unit vector
hoping you get my point

edward,

I understood what he meant. Please do not be attached to the term. x = i, y = j say. all misunderstood anything. sonuçte F = 1 i + 2 j is the mean resultant force vector. that can not scale well. How well do you scale factor logging. (If you're changing coefficients in the combination?) jpeg screenshot please attaching menu.

Thank you again for interest on.

falconado_turkey

[mecheil2000]

Hi Troy
Hi Z.G
Please get the link for this pdf file
this file show u step by step seismic load by ubc
But still scale factor without solution by robot
as I told before we can generate a new combination and multiply the seismic load by the required scale factor
Hoping it will be useful
http://www.4shared.com/file/QoEHcxiR/ToTroyAndZG.html

[troy_the.best]

Hi Troy
Hi Z.G
Please get the link for this pdf file
this file show u step by step seismic load by ubc
But still scale factor without solution by robot
as I told before we can generate a new combination and multiply the seismic load by the required scale factor
Hoping it will be useful
http://www.4shared.com/file/QoEHcxiR/ToTroyAndZG.html


many thanks edward brother,


I will first review. I write again later. bye

ps: you have a big archive in 4shared. its suppper.. thanks.
can i upload your file from archive? Tower40FloorZiajdenRobot2011.rar


best regards
falconado_turkey

[troy_the.best]

Hi edward,

You've done a great work. was very surprised. Thank you very much. I now understand more clearly what they want to say.nobody could tell that so well. into the hands of health care.

about robot model.
1- my question is floor analiz method is rigid diaphram or only shell?
2- in our seismic standart have a criteria for peryot. N= number story of building
peryot not greater than Tx,Ty < 0.1N ( sample number story 40 then T max = 4 second.) this criteria at UBC yet? and why your bulding restraint simple restraint, not fixed. if fixed this base nodes then your period itself will decrease.


and you are lucky mans. cuz your code and standart exist in Robot and etabs. We are assigning for static lateral load shell is rigid. we are doing rigidity center lateral load. with hands calculate and define value for fx,fy. look we areuse turkish standart but not exist etabs and robot these. :/

many many thanks

falconado_turkey

[troy_the.best]

Hi Troy
Hi Z.G
Please get the link for this pdf file
this file show u step by step seismic load by ubc
But still scale factor without solution by robot
as I told before we can generate a new combination and multiply the seismic load by the required scale factor
Hoping it will be useful
http://www.4shared.com/file/QoEHcxiR/ToTroyAndZG.html



dear brother,

I want to say a few things.

first ..
we use scale factor in units ton.meter., but your definition is kip.in. is it mistake?
second...
we doing seismic loads analysis, if Vspec > Vstatic never reduce scale factor. because it will be wrong. but seismic (response analysis) are essentially. therefore we cant response spectrum scale factor in etabs and other programs. why do you this reduce? i m not understand, brother. is it right?

see later

best regards
falconado_turkey

[mecheil2000]

dear brother,

I want to say a few things.

first ..
we use scale factor in units ton.meter., but your definition is kip.in. is it mistake?
second...
we doing seismic loads analysis, if Vspec > Vstatic never reduce scale factor. because it will be wrong. but seismic (response analysis) are essentially. therefore we cant response spectrum scale factor in etabs and other programs. why do you this reduce? i m not understand, brother. is it right?

see later

best regards
falconado_turkey
See,
when you open Etabs by default units will be kpi and you have to change to KN-M
when we run the model as a first time we put an accelarion of Spectrum diagram 9.81 m/s2
you can select any value you can put 1 2 or 3 as you want
We will say this is first scale factor Say FS1 ok
when we run analysis and we will know the natural period of building then we can get a static shear force
simply static shear force means the building is rigid i.e the the full energy of seismic will transfere to the building ok

Last step we have to callibrate the accelarion of seismic for response spec
this is what we say scale Factor
so the dinal scale factor of response diagram say SF2= SF1 x Vstatic/Vdyanmic
ok
But now you will be confiused and you ask how Vstatic< vdynaimc
it is not like that just we adjust the accelarion of spectrum curve
you get my point
Hoping it is clear

[troy_the.best]

See,
when you open Etabs by default units will be kpi and you have to change to KN-M
when we run the model as a first time we put an accelarion of Spectrum diagram 9.81 m/s2
you can select any value you can put 1 2 or 3 as you want
We will say this is first scale factor Say FS1 ok
when we run analysis and we will know the natural period of building then we can get a static shear force
simply static shear force means the building is rigid i.e the the full energy of seismic will transfere to the building ok

Last step we have to callibrate the accelarion of seismic for response spec
this is what we say scale Factor
so the dinal scale factor of response diagram say SF2= SF1 x Vstatic/Vdyanmic
ok
But now you will be confiused and you ask how Vstatic< vdynaimc
it is not like that just we adjust the accelarion of spectrum curve
you get my point
Hoping it is clear


edward,

if Vspec > Vstatic ( Vspec too large then Vstatic) then probably a mistake your model or specrtum analysis. values are usually very close. Please pay attention to this point. Have very different results, if the model is broken or not stabil. and dont reduce spec factor (if Vspec is great) absolutely. I know this very well.. please check again this point edward. usually spectrum analysis is essentialy in our solutions. equal lateral static loads for comparison. and have any period restrictions in your Code.


To use this equation. SF2= SF1 x Vstatic/Vdyanmic (have to be Vstatic>Vspec)

sorry my english.

best regards
falconado_turkey

[troy_the.best]

hi friends,

i have a question. please look this screenshoot. whats the mean main mode number in seismic case.?

thanks in advice.


regards.
falconado_turkey
A.S.C.

[romanich]

troy_the.best,
Run modal analysis, then go to Results--Advanced--Modal Analysis. Look at columns name as Cur.mass UX, UY & UZ. Find maximum value in each column and look at number of mode for this value. And now you have main modes for X, Y & Z direction! Just type it in Combination Sign menu (your screenshot).

[troy_the.best]

troy_the.best,
Run modal analysis, then go to Results--Advanced--Modal Analysis. Look at columns name as Cur.mass UX, UY & UZ. Find maximum value in each column and look at number of mode for this value. And now you have main modes for X, Y & Z direction! Just type it in Combination Sign menu (your screenshot).

thanks romanich,

I had already guessed. then what we do? typing mode number and again analysis???? or what? in results will change anything ;??? i can understand this point.
thanks in advice.

best regards
falcoando_turkey
ASC

[romanich]

Yes, we run analysis again!

[pascal.bogdan]

hi friends,

i have a question. please look this screenshoot. whats the mean main mode number in seismic case.?

thanks in advice.


regards.
falconado_turkey
A.S.C.


Main mode – The number of the main mode for determining the sign of the given combination, so that (as you can see in the attached picture) in the seismic combination on Y direction you will have some braces tensioned and the other ones working in compression.

[troy_the.best]

Main mode – The number of the main mode for determining the sign of the given combination, so that (as you can see in the attached picture) in the seismic combination on Y direction you will have some braces tensioned and the other ones working in compression.

Thanks Pascal, but i m not understand. Cuz romanich and you say two very different things, i m confused.
Main mode number is max comb. mode number or tension - compression sign? if tension, seismic x main mode=1, if compression, seismic y main mode=2, really??

please explain this either of you??

regards..

falconado_turkey
ASC

[pascal.bogdan]

Thanks Pascal, but i m not understand. Cuz romanich and you say two very different things, i m confused.
Main mode number is max comb. mode number or tension - compression sign? if tension, seismic x main mode=1, if compression, seismic y main mode=2, really??

please explain this either of you??

regards..

falconado_turkey
ASC

I have attached another pictures. There you will see the seismic combination on +Y direction and how the elements are subjected to axial force. Because I set there the main mode for Y direction, the vibration mode 2, the program will be able to tell which braces are tensioned and which are compressed.
I think that if you look at the pictures I have attached you will be able to understand why is so important this option.

[troy_the.best]

I have attached another pictures. There you will see the seismic combination on +Y direction and how the elements are subjected to axial force. Because I set there the main mode for Y direction, the vibration mode 2, the program will be able to tell which braces are tensioned and which are compressed.
I think that if you look at the pictures I have attached you will be able to understand why is so important this option.

Pascal thanks for working. yes i m understood now better. But which structural type use this option. or if have a braced object in model. use this option. well If we do not use, Would it have made a mistake? Can you explain a little bit or if you have the chance to suggest any resources please, because As far as I understand this option important.

((Under what circumstances use it? please any document or website suggest to me.))

thanks in advice
best regards

falconado_turkey
ASC

[pascal.bogdan]

Pascal thanks for working. yes i m understood now better. But which structural type use this option. or if have a braced object in model. use this option. well If we do not use, Would it have made a mistake? Can you explain a little bit or if you have the chance to suggest any resources please, because As far as I understand this option important.

((Under what circumstances use it? please any document or website suggest to me.))

thanks in advice
best regards

falconado_turkey
ASC

For structural steel is very important which brace is in compression and which is in tension, because one will be checked for buckling, one not( because it is in tension). Also, the program take into account about this issue by itself when using steel design modulus so the results will be adequate.

[troy_the.best]

thanks Pascal..

yes i a m understand this case clearly now. see later..

regards
ASC

[mecheil2000]

edward,

if Vspec > Vstatic ( Vspec too large then Vstatic) then probably a mistake your model or specrtum analysis. values are usually very close. Please pay attention to this point. Have very different results, if the model is broken or not stabil. and dont reduce spec factor (if Vspec is great) absolutely. I know this very well.. please check again this point edward. usually spectrum analysis is essentialy in our solutions. equal lateral static loads for comparison. and have any period restrictions in your Code.


To use this equation. SF2= SF1 x Vstatic/Vdyanmic (have to be Vstatic>Vspec)

sorry my english.

best regards
falconado_turkey
Hi Falconado,
How r you?
See Scale factor means the accelration of spectrum diagram curve
first run we apply with scale facctor = g = 9.81 m/sec
this is not possible it means just to calebrate the results
next time we we apply the exact scale factor = 9.81 x Vstatic/Vdynamic
say = 2.3 ,
this procedure is based on the ubc code
because the response spectrum curve should be acceleration of seism/9.81
hoping this clear

[troy_the.best]

Hi Falconado,
How r you?
See Scale factor means the accelration of spectrum diagram curve
first run we apply with scale facctor = g = 9.81 m/sec
this is not possible it means just to calebrate the results
next time we we apply the exact scale factor = 9.81 x Vstatic/Vdynamic
say = 2.3 ,
this procedure is based on the ubc code
because the response spectrum curve should be acceleration of seism/9.81
hoping this clear

thanks. edward i m fine.
No problems for me. if you are sure you are doing is correct, is okay. reduction, but we do not do that.
Thanks again for your interest.

best regards
falconado_turkey
ASC

[troy_the.best]

edward,

reduction in the scale factor we can not do. if Vstatic / Vdynamic <1 is okey. but in other situation that Vstatic / Vdynamic> 1, the reduction made by the new scale factor would again be the solution. We do not look at the the equivalent loads value if spectrum of behavior loads value is enough. (Vstatic <Vdynamic then no problem. Ok process.)

thanks in advice

falcoando_turkey
ASC

[quangntn1992743652]

I have trouble Robot and Etabs! You can help me? I have results period so different! Can you email your model Robot and Etabs in picture to me. My email :quangntn1992@gmail.com. Thanks you

- - - Updated - - -


My model etabs v9.7.1 and RSA2011 sp2
results are same. ıts perfect run.
no problem modal analysis. bye

falconado

[Ed Jobe]

I have trouble Robot and Etabs! You can help me? I have results period so different! Can you email your model Robot and Etabs in picture to me. My email :quangntn1992@gmail.com. Thanks you

- - - Updated - - -

If you want help, tell what your problem is.

[quangntn1992743652]

If you want help, tell what your problem is.

104748

I have 2 model Robot and Etabs, I assign same load gravity (Self Weight, Superdead Load, Brick Load, Live Load). I find period different about 2 model, Can you helf me?

Model Robot: https://drive.google.com/file/d/0B0b19LZR90c0UEY5VUVjM3k1alU/view

Model Etabs: https://drive.google.com/file/d/0B0b19LZR90c0b3A3MkdMRTJ3b3c/view?usp=sharing