Choosing which method for Seismic Analysis (Eqiv. Static Procedure v. Dynamic Analysis)

[Mohzus]

Hello!

I'm looking for some clarification on when to use either method of analysis (in NBC 2015) when dealing with wood. An example would be a 5-storey irregular wood structure:

1. From my understanding is that Equiv. Static Procedure can be followed if it meets the criteria outlined in Clause 4.1.8.7(1), but I've been told that doing a dynamic analysis (spectural response method) on wood is not applicable because it's neither rigid or completely flexible. Instead using the Eqiv. Static Procedure with Flexible and Rigid method of analysis for wood structures (such as WoodWorks software or by hand). Leaving only dynamic analysis for steel and concrete structures since they are rigid bodies. Is this correct? and if so, where does the NBC 2015 comment on wood structures able to ignore dynamic analysis even with irregularities.

2. The Eqiv. Static Procedure is laid out clearly in the NBC 2015, but I haven't used the dynamic analysis method before. What does that look like? Is it completely software based using ETABS? or is there a procedure by hand that can be followed?



Forgive me for I'm an EIT on the scene and wasn't lucky enough to have a seismic course offered. Any help is greatly appreciated!

Thanks!

 

[Guest090822]

Mohzus,

Where does it say you can't use spectral response methods for wood? For seismic analysis, the engineer usually has a lot of freedom to make the analysis as complicated as they want. I've used specula response methods for wood structures, however, never anything greater than 3 stories.

If you can get acmes to the NDS Wood codes, take a look at what they recommend.

 

[JoshPlumSE]

Wood structures are more difficult to do via FEM whether it be static or dynamic analysis.

The elastic stiffness of a wood shear wall (for example) has a lot of factors in it. For concrete or masonry walls, you normally just worry about flexural and shear deformation. Then apply some factor against the gross properties for cracking.

For wood you've got the following for a single story solid wall (not even talking about openings yet):
1) Flexural deformation
2) Shear deformation
3) Fastener slip / nail slip. This is really non-linear. But, NDS has us take a look at the slip at failure and apply that extra flexibility as an "apparent shear stiffness" which modifies item 2 instead.
4) Hold-down deformation. How do you account for this? Do you just soften up the wall by a certain percentage to account for this effect (which can be pretty significant).

Now, I tend to assume that openings don't need any special treatment if you use the Force Transfer Around Openings method of analysis. This assumes that if you've adequately accounted for items 1,2 and 3 this will automatically account for any openings.

Multi-story walls mean you don't have hold down deformation in the upper panels. But, you will likely have similar issues associated with whatever straps connect the various walls together. Or, do you have continuous rods at those hold down locations (which is becoming popular for multi-story wood buildings).

In summary, this is nowhere near as automated in analysis programs as it is for concrete and steel. So, the users may have to re-invent the wheel a bit when they do this.

 

[JIMEY]

I'm actually having problems with this very issue right now. I've got a 5 storey wood frame building. Due to irregularities, I need to design it with a period of 0.5 seconds - otherwise the building code requires a dynamic analysis. The problem is that in reality the period is probably at least double that, so if I want to stick with static analysis, I have to design with loads that are far higher than they should be. I would like to be able to just do a dynamic analysis to keep the loads down, but I don't know of any software with that capability. I usually use ETABS for my analysis, and I could just manually input the shear wall stiffness, but as JoshPlumSE rightly pointed out, wood shear wall stiffness is very nonlinear. The backbone curve is really curved. I wonder if anyone knows a way to "fake" ETABS (or any structural analysis software) into modeling wood shear walls accurately?

 

[molibden]

Woodworks probably. In Europe I use Timbertech.