Wood Shear Walls 2

[CrabbyT]

I'd like to have a deeper understanding of wood-frame shear walls. For context, I'm in the USA and have access to the 2015 AWC NDS and SDPWS. My background is in steel/concrete design, but lately I've been dabbling in wood design.

[li]After calculating forces, where do you usually start when designing a shear wall?[/li]
[li]Does anyone here know of any helpful resources (example problems, design guides, etc) for designing wood-frame shear walls?[/li]
[li]Are there any software packages you would recommend? Any you wouldn't?[/li]
[li]Would anyone be willing to share a set of drawings showing how one would typically call out hold downs, straps, nail patterns, and etc?[/li]

In terms of software, I've tried RISA 3D and it seems pretty nice (I like that it does multiple checks on the wall including axial checks). Woodworks Shearwalls looks pretty good too, but I haven't tried it. At any rate, I'd really like to know what I'm doing before going the garbage in/garbage out software route.

Tables 4.3A - D in the SDPWS seem simple enough, although I wish they included some pictures (I'm a little confused about the nail pattern - is "panel edge fastener spacing" the only fastener spacing that matters? Does the nail spacing to the studs not matter at all?). I'm also a little confused on how the sheathing material works. Say I have 15/32" sheathing on the exterior side of a wall and 1/2" gypsum board on the interior side. Do I just add those capacities together? Am I missing something?

 

[driftLimiter]

The edge nailing is the nailing to studs around the Perimeter of the individual shear panels. The 'field' nailing is the spacing of nails to all other studs in contact with the shear panel.

Easily understood by pictures I agree so here is my typical sw det.

Note that blocking is typically used at horizontal panel edges, however SDPWS does have tables for unblocked shear walls but they take a pretty good penalty on strength and stiffness. We always used blocked.

 

[CrabbyT]

Thank you! That detail and explanation about blocking is a big help. I was going to ask about blocking, but figured my post was long enough.

 

[ChorasDen]

As far as as software goes, I have only used WoodWorks software (which was several years ago), and to be completely honest, I didn't really enjoy using it. As such, I don't recommend their software. Others on this forum may have more experience and may also have a better opinion of it.

Say I have 15/32" sheathing on the exterior side of a wall and 1/2" gypsum board on the interior side. Do I just add those capacities together? Am I missing something?

I'd be very careful relying on (and designing) gypsum shear walls. keep in mind that the detailing for any shear wall requires that the sheathing be connected from the bottom of the wall continuous to the top (SDPWS Table 4.3C). And at the interior side of exterior walls where the gypsum attaches, I do not often see the gypsum connected to both the top AND bottom plates.

 

[ChorasDen]

To clarify, I wouldn't rely on contributions from both the gypsum and OSB/Plywood on a single wall. Not sure where you're located, but if you are in a seismic design area, Gypsum shear walls also have a different R value per ASCE 7, and this would increase the seismic loading onto the structure, which would likely outweigh the nominal benefit that the gypsum would bring.

 

[driftLimiter]

NDS SDPWS addresses this specifically. I never count on the gypsum. As noted it has a different R factor and different stiffness associated with it. Plus.... its gypsum board lol

 

[phamENG]

I prefer not to use the gypsum...but everyone else here does so it can be hard to stay competitive. I will show a 4/12 nailing pattern on a wall before I show GWB as required on the interior, though. I don't have to worry about seismic here for houses (SDC A) so it's all wind. No penalty, but there's something unsettling about the MWFRS dissolving in case of a roof leak. In hurricane country, roof damage and high winds go hand in hand...

For software, I like Woodworks Shearwall for basic buildings. You have to fudge it a bit on more complex roof geometries, or irregular vertical layouts. But it's usually 'good enough'. If nothing else, I'll use it to generate lateral loads on shear lines. It's a lot faster than doing it by hand. It's very input intensive, so it takes a while to learn it and understand what you're doing. In that sense, though, it can be a good tool to help you learn shear walls. If you investigate what every button, toggle, or option does, you'll be a shear wall master in no time. Haven't used RISA for shear walls in a couple years. It was clunky then. Maybe the newer UI has made it easier? It's usually best to just build your own spreadsheet, though. Do a couple by hand, then build the sheet. It's pretty straightforward once you wrap your head around the basics. FTAO is a useful tool once you've learned segmented and perforated. There's a bit of a discussion going on in this thread right now (though it's fizzling out): thread507-499942

Basic design procedure:
1) Determine the load to each shear line based on at least a flexible diaphragm assumption. Some will also do a rigid diaphragm assumption to 'bracket' the solution, but it isn't necessary in a lot of cases and also a lot of work if you're doing it manually.
2) Determine usable walls in each line and distribute shear load to them proportionally based on length.
3) Determine hold down forces and unit shear in each wall.
4) Select a hold down and sheathing/nailing pattern for each wall.
5) Calculate deflection for each wall. Do all the deflections for each wall in each line match?
6) If 5 is yes, move on to step 7. If 5 is no, change the load distribution until deflection is equal (more load to stiffer elements).
7) Check your wall for serviceability - seismic drift and wind serviceability (I use a 10 year MRI wind speed for this, which is roughly equal to 0.45W when using the 700 year MRI for calculating W for my location).

 

[EngDM]

Typically if OSB or plywood is specified as well as gypsum I just neglect the capacity given by the Gypsum. It wouldn't be as simple as adding capacities due to the relative stiffness of each material, the load doesn't transfer uniformly and this just becomes time consuming.

 

[CrabbyT]

Hey everyone - thank you for the input thus far. I want to clarify one thing though. My question about gypsum was more conceptual in nature. I don't intend to use gypsum to resist shear. Conceptually, I was just curious if the values in the table are additive.

Here's a better example. Say that I'm called in on a project where they've sheeted the exterior face of the wall with 15/32 sheathing and siding has been installed. Imagine that I calculate the loads, determine that 15/32 isn't strong enough by itself, and I need to install something on the interior face of the wall. Based on the table, I determine I need a 5/16 wood structural panel.

Do I just add the shear capacities together (va = 680 plf for 15/32 sheathing, va = 360 plf for 5/16 sheathing, va,total = 680 + 360 = 1040 plf)? Are there adjustments that need to be made if one panel is stiffer than the other? If you know whether or not that's addressed in the code, please let me know. I haven't managed to read all the way through it yet.

Gypsum shear walls also have a different R value per ASCE 7

Thank you - I was not aware of this. I'm not in a high seismic area, but maybe someday I will be. At any rate, I don't intend on using gypsum board to resist shear, but I can envision a scenario where I'm tasked with calculating the shear capacity of an existing wall and might need to rely on it eventually.

Haven't used RISA for shear walls in a couple years. It was clunky then. Maybe the newer UI has made it easier?

I'll check out Woodworks Shearwalls, sounds like it could be a good learning tool. I used RISA about 7 years ago and have been using RAM Elements ever since. I downloaded the trial version, and I think the new interface is an improvement, but it still feels very clunky to me. I love the workflow in RAM - modelling a new structure is basically just clicking and inputting from left to right, and I like that they don't use way too many icons (and that the icons make sense). And the selection options in RAM are better. But it doesn't do shear panels, Euler buckling, or time history though, so that's a bummer. Annnnyway...

6) If 5 is yes, move on to step 7. If 5 is no, change the load distribution until deflection is equal (more load to stiffer elements).

Thank you for that basic design procedure. That makes a lot of sense, especially the deflections. It's not good enough for the wall to be strong enough - they should all deflect about the same amount, too.

Step 6 sounds troublesome. Do you generally try to avoid doing that? I'm a little concerned about starting off with a flexible diaphragm assumption and then trying to rework the analysis under a rigid diaphragm assumption. Seems like it would be easier to design by matching stiffness and making things symmetrical. I understand how that could be a problem though.

 

[phamENG]

Do I just add the shear capacities together?

It depends. Read SDPWS section 4.3.3.3. It addresses this in detail along with the various adjustments and exceptions. (I design for wind, so I fall under the directly additive exception.)

As for my step 6, read SDPWS 4.3.3.4, Shear Walls in a Line. It discusses distribution to the walls. There are two methods. One is based on capacity/strength (but has a stiffer than usual penalty for high aspect ratio walls). The other is deflection. The deflection makes more sense to me, so I use that - though it does make things more involved if you do it by hand.

Flexible is likely all you need to do, unless you're dealing with a 3 sided diaphragm or something really abnormal.

making things symmetrical

I love symmetry. It's mathematically superior and pleasing to the eye. But architects post-1890 seem to hate it. So good luck with that.

 

[ChorasDen]

Step 6 sounds troublesome. Do you generally try to avoid doing that? I'm a little concerned about starting off with a flexible diaphragm assumption and then trying to rework the analysis under a rigid diaphragm assumption. Seems like it would be easier to design by matching stiffness and making things symmetrical

For myself, I always prefer to model as rigid or semi-rigid diaphragm and shear walls, think it makes more intuitive sense. If you are trying to get a feel for how to approach this concept in wood shear walls, the 2015 SDPWS has a fairly simple 2-page example in the commentary that discusses force distribution in (2) different walls of various stiffness. You can apply the methodology into a spreadsheet or your own software, if desired. The example is C4.3.3.4.1-1 on page 82.

 

[driftLimiter]

Our spreadsheet here solves the equal deflection problem by successive iterations. The flowchart is something like this.



1)Start out by distributing the shear to walls in the same line based on their lengths. i.e. ratioA = lengthA / sum (lengths in the line).
2)Next solve for sheathing and hold downs required. We use consistent shear panels and nailing for all walls in one line.
3)Now calculate deflection of each wall.
4)Using the calculated deflection and the and the loading for each wall, compute the wall stiffness K.

5)Now you kick off iteration 2. But this time you distribute the forces to the walls based on the wall Stiffness K.
Then you repeat steps 2 - 4 from above and keep iterating until you have some satisfactory convergence.

Its definitely not a trivial solution.

 

[RobertHale]

The APA has a document on shear walls that was pretty useful when I was starting out (APA Form L350 "Diaphragms and Shear Walls - Design/Construction Guide"). My experience with software is similar to ChorasDen. As a consequence, I went and wrote a spreadsheet to design walls. I use the perforated wall method. I think it saves on connectors and aggravation during construction just to take the hit to capacity. The spreadsheet design tool also gives me much more flexibility around things like Anchor Tie Down Systems. I have a few selected sheets to show the typical detailing I provide on wood shear walls. The three projects represent an evolution of documentation.
I do not use gypsum. I don't even like to count on gypsum for the stability of the studs. My team at my firm takes the stance that counting on gypsum walls is problematic if the construction documents are "lost" when they go to renovate. There is no way to tell a gypsum-sheathed structural loadbearing wall from a gypsum-sheathed loadbearing structural shear wall. Yes, it means I'm not competitive on multifamily wood, but I don't have a lot of lost sleep from not getting a job where the developer is pressuring the structural system to save cost.

 

[phamENG]

My team at my firm takes the stance that counting on gypsum walls is problematic if the construction documents are "lost" when they go to renovate.

This is certainly a concern. Especially in single family residential. One thing I've heard of people using to mitigate it is to have SHEAR WALL written in indelible ink on all of the blocking - the hope being that blocking behind GWB, hold downs at the ends of the walls, and the word SHEAR WALL all over the place will cause somebody to stop. I was the 18 year old, sledge hammer wielding demo guy once...so I know that's a pipe dream. But it's something!

 

[WiSEiwish]

I personally use gypsum shear walls, but I've also been lucky and have been in SDC A, so R values are unimportant. The good thing about gypsum is that the building will be a heckuva lot cheaper to build. If we are talking demising walls in a multi-family building, they are sheathed with gypsum already, so if you can just make that a shear wall then all you need to do is specify fastening/hold-downs. If you made it wood then the architect needs to add a layer of gypsum on top of that to get their finishes and then lose that small amount of space. That being said, shear walls need to be CLEARLY identified on plan and try to use walls that are going to be structural anyway, so you can't just knock them out willy-nilly.
For software I've been intrigued by woodworks, but haven't figured it out yet. RISA3d is nice, but controlling the parameters takes some time. I've been finding that the Tedds module is quite nice. Inputs are very easy. You have to design the hold-down by hand, but I kinda prefer that.