Modeling WSP with Plates

[medeek]

I've been playing around with RISA3D a bit more lately and I think it might be useful in doing some in depth analysis of portal frames. However, after reading through some posts by others on the board I'm trying to figure out whether or not one can get reasonable data with this sort of thing. I want to model all of the studs, beams and plates and then use plate elements between them to represent the sheathing. A previous thread had some mention about "E" and "G" not being compatible for this type of modeling and using a fictitious "E" value based on a poisson's value of 0.2

I am wondering what others experience with this sort of thing has been and what suggestions they might have in this regard. I did notice that if I input the typical "E" and "G" values that I found listed for plywood my model was too "stiff" and I didn't get any realistic deflections.

I am really curious how the wall panel element within RISA works and what type of "E" and "G" values are used to get deflections that compare with the calculated methods of the NDS SDPWS.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[JoshPlumSE]

Medeek -

That previous post was likely from a number of years ago (prior to 2008-ish). Back then, the plate element formulation was based on E and poisson's ratio and ignored the G value. That's no longer the case and hasn't been the case since the version 8.0 release.

The wall panel element in RISA-3D uses true ortho-tropic plate elements where different stiffness properties are used for axial vs shear deformation. We do this so that the axial stiffness represents the stiffness of the studs, but the shear stiffness represents the stiffness of the sheathing (based on the "apparent shear stiffness" designated in the NDS SDPWS spec.)

As such, the program doesn't attempt to model individual studs or posts or such. Rather, the desire is to get a representative (but uniform) axial stiffness based on the total number of studs. And, a representative shear based on sheathing and nailing.

There is a "Shear Stiffness Adjustment Factor" that can be applied to the walls. This is used to soften the walls in the case that the wall deflection doesn't match up with the NDS estimated deflections. Probably more useful for walls which are very narrow and hold down deformation becomes a major consideration.

 

[medeek]

What is the suggested method for doing wood portal frame analysis? I'm pretty sure I am not the first to want to try and do this and will not be the last. Are there any tutorials specifically dealing with wood portal frames and how best to model them?

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[JoshPlumSE]

I'm not sure what a Wood Portal frame is. My impression is that wood members could not be connected in a way to transfer moment.

I've seen post beam construction with kicker braces. But, to me, that's more of a braced frame than a moment frame. At least in terms of how it gets analyzed.

 

[medeek]

Prescriptive portal frames are found in the IRC and IBC, however I am interested in trying to do a rational analysis of them and then devise a method to "engineer" them.

http://publicecodes.cyberregs.com/icod/irc/2009/icod_irc_2009_6_par042.htm

My method of manual analysis is based on a paper by Zeno Marti "Principles of Mechanics Models for Wood Structural Panel Portal Frames". However, I would like to validate this manual analysis further with an appropriate FEA model.


A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[EngineeringEric]

My understanding is that portal frames are more of a 'Tested Capacity' than a calculated. And i hate that. I have been able to rationalize it by looking at the two jambs as T/C members with the beam inducing a force couple and then the wall panel distributing that on down... Not accurate but i could get some decent values. I would not use on in shear wall design however, only prescriptive.

Since JoshPlum is now involved. I do wish he could talk to his programming buddies next door and have them add in Wood panels in their plates so i could model a sloped wooden diaphragm straight in RISA-3D! Or maybe i need to make a new post if this is possible!

 

[medeek]

The APA publication TT-100F (April 2014) has some published allowables for typical prescriptive portal frames. The problem I run into with these numbers is occasionally I get a garage wall that does not quite fit within the max. and min. of the table or the allowables given. Standard shearwalls won't work either since the aspect ratios are too large. If I can't get shearwalls to work or portal frames I am left with Simpson Strongwall products or Steel Moment Frames, both of which are expensive and the local contractors hate them, even the field adjustable Strong-Wall SB has caused me some serious flack.

My thinking is if I use larger holdowns, triple chords, larger and strong straps, beefed up stemwalls, and both sides sheathed I can get quite a bit more out of one of these portal frames. The trick is to quantify this amount which is what I have tried to do using the methods of Zeno Martin's paper and also referencing Terry Malone's book on the subject. Malone uses FEA along with his manual calculations throughout his book and I would like to follow his lead and see if I can get comparable numbers using RISA3D.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com