Hi,
after having carried out a pushover I would need to be able to extrapolate from the modal analysis on the inherent structure, the modal participation factor of the first mode of vibration, in order to reduce the curve, as requested by NTC 2008.
I see that Seismostruct gives in output this factors but (also written on help) I think they are not factored. Is there some setup options to get this or how else I can obtain them?
Thanks for any answer.
Modal participation factors
Re: Modal participation factors
Hi,
in order to compute the modal participation factor "gamma" as defined in paragraph C7.3.4.1 (NTC 2008), I suggest to proceed as follows:
1. Run an "Eigenvalue analysis"
2. Go to the "Post-Processor" area and select the "Step output" module
3. For each principal mode (1st mode in X and 1st mode in Y, if you have a 3D structure), click on the corresponding output (e.g Output No.1 - Period ....)
4. At this point, select all the nodal displacements, copy and paste them into a spreadsheet.
5. From those values, compute the normalized displacements (e.g. normalized to the maximum) in order to obtain "phi" vector and then apply the equation C7.3.5
I hope everything is clear. If not, let me know...
Federica
in order to compute the modal participation factor "gamma" as defined in paragraph C7.3.4.1 (NTC 2008), I suggest to proceed as follows:
1. Run an "Eigenvalue analysis"
2. Go to the "Post-Processor" area and select the "Step output" module
3. For each principal mode (1st mode in X and 1st mode in Y, if you have a 3D structure), click on the corresponding output (e.g Output No.1 - Period ....)
4. At this point, select all the nodal displacements, copy and paste them into a spreadsheet.
5. From those values, compute the normalized displacements (e.g. normalized to the maximum) in order to obtain "phi" vector and then apply the equation C7.3.5
I hope everything is clear. If not, let me know...
Federica
Re: Modal participation factors
Hi,
you were very clear. Also because while I was waiting for answers I also supposed this approach. The only problem that remains is that of the masses at the nodes to complete the paragraph C7.3.5, which should be derived manually, because I used the masses as distributed loads along the beams. Is there any way to extrapolate the loads at the nodes with this method?
Thanks for any answers.
you were very clear. Also because while I was waiting for answers I also supposed this approach. The only problem that remains is that of the masses at the nodes to complete the paragraph C7.3.5, which should be derived manually, because I used the masses as distributed loads along the beams. Is there any way to extrapolate the loads at the nodes with this method?
Thanks for any answers.
Re: Modal participation factors
Hi,
unfortunately, you have to compute the nodal masses manually...
Federica
unfortunately, you have to compute the nodal masses manually...
Federica
-
seismosoft
- Posts: 1271
- Joined: 06 Jul 2007, 04:55
Re: Modal participation factors
Hi both,
Just a quick note on the fact that in the forthcoming release of SeismoStruct (March/April), nodal masses will be provided by the program.
Seismosoft Support
Just a quick note on the fact that in the forthcoming release of SeismoStruct (March/April), nodal masses will be provided by the program.
Seismosoft Support
Re: Modal participation factors
i am having a load of 19.24 ton (in the form of concrete blocks or Slabs)on top of circular column. How can i calculate the Nodal mass ???
Re: Modal participation factors
zsaeed,
There is something not quite clear to me about your query, since you refer to loads as a quantity that is defined in tonnes (when 'tonnes' is typically used for masses), and you also ask on how to get nodal masses right underneath a (4 years old) post that says that the program now readily provides such information.
In any case, I have noticed that you have raised the same issue in a number of other posts, and thus refer you to my answers in those: http://www.seismosoft.com/forum/topic.asp?TOPIC_ID=1159.
Rui
There is something not quite clear to me about your query, since you refer to loads as a quantity that is defined in tonnes (when 'tonnes' is typically used for masses), and you also ask on how to get nodal masses right underneath a (4 years old) post that says that the program now readily provides such information.
In any case, I have noticed that you have raised the same issue in a number of other posts, and thus refer you to my answers in those: http://www.seismosoft.com/forum/topic.asp?TOPIC_ID=1159.
Rui
Re: Modal participation factors
Hi,
I want to ask 2 quick questions:
1)To calculate effective modal mass for X direction of 3D structure in excel I followed the steps below:
-I took nodal displacements of corresponding period and also modal masses in X direction. Then used in excel with the equation to calculate effective modal mass using the equation: sum[M(i)Phi(i)]^2/sum[M(i)Phi(i)^2] but this value is not same as the effective modal mass value of the related mode in eigenvalue analysis. What can be reason? (I also tried with spear model in verificatio report there is 8 tonnes difference.)
2)To calculate participation factor, I can normalize based on one of the nodal displacements in roof level. But those roof displacements are different from each other and when I change to normalize with different node my participation factor dramatically change. I need this factor to convert my pushover curve to capacity curve(sa vs. sd). So, which node should I use for normalization of mode shape to get proper participation factor?(for instance centre of mass or the highest one?)
Thank you.
I want to ask 2 quick questions:
1)To calculate effective modal mass for X direction of 3D structure in excel I followed the steps below:
-I took nodal displacements of corresponding period and also modal masses in X direction. Then used in excel with the equation to calculate effective modal mass using the equation: sum[M(i)Phi(i)]^2/sum[M(i)Phi(i)^2] but this value is not same as the effective modal mass value of the related mode in eigenvalue analysis. What can be reason? (I also tried with spear model in verificatio report there is 8 tonnes difference.)
2)To calculate participation factor, I can normalize based on one of the nodal displacements in roof level. But those roof displacements are different from each other and when I change to normalize with different node my participation factor dramatically change. I need this factor to convert my pushover curve to capacity curve(sa vs. sd). So, which node should I use for normalization of mode shape to get proper participation factor?(for instance centre of mass or the highest one?)
Thank you.
Re: Modal participation factors
Hi morello. It would appear to me that you have tried to extract only a portion of the mass and stiffness matrices when you used only x-direction dofs. I am certainly no expert, but I believe you will need to work with the full matrices, [K} and [M], rather than extracted portions of those matrices to get precise values for modal mass and participation factors in post-processing outside of SeismoStruct.
Is there a reason that the modal masses and participation factors reported by SeismoStruct do not meet your needs? Or are you simply wanting to verify them?
Best of luck morello.
Is there a reason that the modal masses and participation factors reported by SeismoStruct do not meet your needs? Or are you simply wanting to verify them?
Best of luck morello.
Tim Huff
Re: Modal participation factors
Hi Tim,
Actually my main reason to do that I need participation factor which uses mode shapes normalized by top floor so this value should be something between 1 and 2. So I just wanted to verify values that I used to calculate for this factor. (Also second post of this topic from fbianchi suggested to use these nodal masses and displacements.)
However, as I read from manual, seismostruct computes that participation factor using mode shapes normalized by mass. Can you suggest any solution to calculate the participation factor that I need?
Regards.
Actually my main reason to do that I need participation factor which uses mode shapes normalized by top floor so this value should be something between 1 and 2. So I just wanted to verify values that I used to calculate for this factor. (Also second post of this topic from fbianchi suggested to use these nodal masses and displacements.)
However, as I read from manual, seismostruct computes that participation factor using mode shapes normalized by mass. Can you suggest any solution to calculate the participation factor that I need?
Regards.