eh, well, I think I figured it out on my own. Much of my mass comes from the modeled infrastructure (the foundations) and the bottom floor slab. These elements do not participate in the vibration modes. This is why the accumulated percentage of mass will never come close to 100%.
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 JAE on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Modal analysis of LSF building 3
- Thread starter fp77
- Start date
be careful with the modelling if you just did the OSB properties and those shells are continuously attached to the CFMF studs/post and floor framing you are likely over estimating the stiffness of the panels, in the real construction there will be slip of the fasteners and slip of any hold down devices.
We'd generally back into equivalent wall panel stiffness and use a single shell for the lateral analysis if even using software for that analysis at all.
Open Source Structural Applications:
We'd generally back into equivalent wall panel stiffness and use a single shell for the lateral analysis if even using software for that analysis at all.
Open Source Structural Applications:
Lateral resisting elements for these types of structures, in my experience, usually consists of Cold Form Strapping which would be modeled more like a steel braced frame.
Open Source Structural Applications:
Open Source Structural Applications:
Erik Panos Kostson
Structural
That does make sense - if most/a lot of your mass is at locations where the mode shapes have almost zero disp. it will not add much to the mass P.F.
- Thread starter
- #5
Thanks for the replies.
The structure I'm modeling isn't exactly your typical LSF structure. It's a modular type of construction developed by the company I'm working on. The walls are produced in the factory and are then placed and connected onsite. Neighboring wall panels are connected from the sides but also from the top and bottom. There are no steel bracing frames, all the lateral stiffness is provided by the OSB panels. All the wall modules (interior and exterior) serve both a structural and insulation purpose.Celt83 said:
There may be an overestimation of the stiffness of the walls, but right now I don't exactly have a better way to model these structures, because I don't really know any better. Unfortunately, I also do not have anyone else to discuss these issues with.Celt83 said:
-
1
- #6
Grab a yourself a copy of this document: AISI D113-19
Several free publications are also available here: [URL unfurl="true"]https://www.cfsei.org/free-publications-by-development[/url]
Open Source Structural Applications:
Several free publications are also available here: [URL unfurl="true"]https://www.cfsei.org/free-publications-by-development[/url]
Open Source Structural Applications:
NCSEA just hosted a webinar on the topic of cfs shear walls a couple days ago. You can access ($$$) the recorded version on their website.
Out of curiosity, have you figured out what was different with the previous analyses of your colleagues that resulted in the high mass participation and low frequency?
Also curious what you decided to do as a workaround with the mass participation issue in your model.
Out of curiosity, have you figured out what was different with the previous analyses of your colleagues that resulted in the high mass participation and low frequency?
Also curious what you decided to do as a workaround with the mass participation issue in your model.
- Thread starter
- #8
Now that I know the reason for the low mass participation, I don't think there's an issue at all. It's just a matter of fact that most of my model's mass does not participate in the vibration modes, because I decided to actually model the foundation and bottom floor elements (I could have just inserted restraints to the columns and not model the bottom floor slab).
My colleagues, when they modeled their structures, did not model the foundation or bottom floor elements. That explains why they got a high mass participation in their modal analyses. Nevertheless, the low frequencies of the structures they modeled are not expected in this type of structures (the best book available in Portugal regarding LSF structures says it's expected for the natural frequency of these structures to be above 10 Hz).
So their results denote that the innate stiffness of these structures is absent from their models. It seems to me that they did not model the walls or any equivalent bracing frames.
My colleagues, when they modeled their structures, did not model the foundation or bottom floor elements. That explains why they got a high mass participation in their modal analyses. Nevertheless, the low frequencies of the structures they modeled are not expected in this type of structures (the best book available in Portugal regarding LSF structures says it's expected for the natural frequency of these structures to be above 10 Hz).
So their results denote that the innate stiffness of these structures is absent from their models. It seems to me that they did not model the walls or any equivalent bracing frames.
I think you already have it figured, but if you consider a simple cantilever. It will not require a huge number of modes to reach 90%. But say that you add a large mass at the base, the fixed base. Then it is impossible to reach 90%.
I have experienced similar issues with box-like concrete buildings for primarily the vertical seismic response. You mention 400 modes, check the forces you have with 400 modes. Then increase to the number of modes to 2000 och 4000 and check the forces again. If there is no signifacant change than 400 is enough include the full loading.
Thomas
I have experienced similar issues with box-like concrete buildings for primarily the vertical seismic response. You mention 400 modes, check the forces you have with 400 modes. Then increase to the number of modes to 2000 och 4000 and check the forces again. If there is no signifacant change than 400 is enough include the full loading.
Thomas
-
2
- #11
JoshPlumSE
Structural
A number of comments:
1) You will need to "soften" the G value for your shell elements modeling OSB. This is to account for nail slip and such. Wait until you know what your nailing pattern is going to be. You can use the NDS which has the concept of "apparent shear stiffness" to account for nail slip.
2) In my experience wood shear wall can be overly stiff if they're not meshed enough.
3) I might have two separate models. One for gravity only system where the vertical studs and beams are all modeled. And, another for only the lateral system. The problem is you might be getting a lot of lateral stiffness out of your studs. Where each of them acts like a hold down point.
4) You might manually "discretize" your mass. Meaning that instead of relying on member self weight, try lumping the mass at the floor locations only. That will get rid of local modes and allow you to get closer to 100% mass participation.
5) Alternatively, you could go with Ritz Vectors. Ritz vectors are very similar to true modes, but they are biased towards a certain deflected shape. This gives them a few advantages:
a) They inherently get rid of lot of local modes that aren't consistent with the deflected shape.
b) They already account for the "residual" mass associated with the base of the structure. Though they acceleration used for this mass can be overly conservative at times.
1) You will need to "soften" the G value for your shell elements modeling OSB. This is to account for nail slip and such. Wait until you know what your nailing pattern is going to be. You can use the NDS which has the concept of "apparent shear stiffness" to account for nail slip.
2) In my experience wood shear wall can be overly stiff if they're not meshed enough.
3) I might have two separate models. One for gravity only system where the vertical studs and beams are all modeled. And, another for only the lateral system. The problem is you might be getting a lot of lateral stiffness out of your studs. Where each of them acts like a hold down point.
4) You might manually "discretize" your mass. Meaning that instead of relying on member self weight, try lumping the mass at the floor locations only. That will get rid of local modes and allow you to get closer to 100% mass participation.
5) Alternatively, you could go with Ritz Vectors. Ritz vectors are very similar to true modes, but they are biased towards a certain deflected shape. This gives them a few advantages:
a) They inherently get rid of lot of local modes that aren't consistent with the deflected shape.
b) They already account for the "residual" mass associated with the base of the structure. Though they acceleration used for this mass can be overly conservative at times.
- Status
Similar threads
- Locked
- Question
- Question
