Residential diaphragm design or lack there of 1

[jeffhed]

Our office (only two of us) are having a discussion on residential roof diaphragm design. I have attached a sample drawing of what I am talking about. Say you have a typical house with a nice long wall at the back of the garage you want to use for shear. However, the trusses run perpendicular to this wall so to provide a drag strut would involve blocking and coil strapping. My colleague says you can use the shear wall without a drag strut by limiting the diaphragm length to the length of the wall at the back of the garage as long as the unit shear on the shear wall is lower than the unit shear on the diaphragm. He also says that the collector for the transverse shear wall will also work as the chord for the longitudinal diaphragm. All this makes sense and could be proven with calculations and I have no doubt that the demand on the diaphragm would be well within the ability of a standard residential diaphragm. I agree the diaphragm he proposes will be strong enough to support the load, but I am unsure of how the remaining portion of the home ties into the diaphragm without a collector. My colleague says that the roof sheathing is all staggered and nailed and the rest of the home just goes along for the ride. This is the part I'm having a hard time accepting, especially for tension drag loads, which is why I say if you use the shear wall at the back of the garage for the longitudinal direction, you need a drag strut or at the very minimum blocking and edge nailing with coil strap out into the diaphragm a sufficient amount to develop your drag force into the adjacent diaphragm. What are others thoughts on this? Can you just rely on the sheathing and the nailing along the left hand side to go along for the ride? Or should a collector be provided. I realize residential loading typically is very low, however, this also can be a problem on large residential projects as well as commercial where loads are higher, but providing a drag strut the full length of the building would cause additional costs that may not be required.

 

[AELLC]

Jeff,

I now do almost all residential, from tract homes to custom, up to 8000 s.f.

Most are in Phoenix area, very low seismic. We never do drag struts, collectors, etc, in this area for residential.

 

[msquared48]

A couple of things here:

1. I normally, at the top story of a trussed roof residence, only use the exterior walls for shear to avoid the internal blocking required between the trusses.

2. The wall top plate of the exterior walls will function as a drag strut if properly detailed for connections between the top plates members.

As for the situation you have shown, I would try to eliminate the interior shear wall and just use the exterior walls. If that cannot be done, then a strap tie similar to a CMST would definitely be in order here, nailed off to the shear wall top plate and the blocking between the other roof trusses.

Mike McCann
MMC Engineering

 

[jeffhed]

AELLC,
The area I am in is typically a Seismic Design Category D, once in a while a C. Most times our wind and seismic loads are almost the same. Longitudinal direction is often controlled by seismic while transverse loads can be wind controlled but seismic will control sometimes. If you only have that shear wall behind the garage, how do you justify it bringing the rest of the home along for the ride without a collector? What happens if your unit shear on the wall is greater than the unit shear in the diaphragm? You would have to have a collector or the shear force could never be delivered to the wall.

msquared48,
1. I would rather not use the wall at the back of the garage for shear, however, many times it is needed to reduce the lateral load at the front of the garage, especially if the garage sticks out a ways. If the exterior walls jog back and forth quite a bit how do you justify it working all together without collectors? Judgement?

2. The same as above, if the exterior walls jog back and forth a bit, how do you justify the collector working if it is not a continuous straight line?

3. If you could not eliminate the interior shear wall, would you just run blocking and coil strap our long enough to develop the drag force? Maybe my colleagues idea of designing the diaphragm on one side and the rest of the home going along for the ride isn't too far off?

 

[msquared48]

1. Well, remember and use your allowable 4' offset for the shear walls in one line. As for the drag forces, usually, unless this has a lot of brick on the walls or tile roof, seismic will not control (which would have a greater effect on any need for drag connectors) and wind will be your main consideration. You need to take a look at the magnitude of the wind pressure/suction forces and see if you feel the drag connectors are needed.

2. If the offset is 4 feet or less (a requirement for them to be considered in one line), I generally would not worry about it - the plywood should be able to do this. With some higher end houses, it could be a problem though.

3. Yes, I would use the strapping to create the drag lines rather than create sub-diaphragms.

Mike McCann
MMC Engineering

 

[AELLC]

jeff,

I have done custom houses in California where all the engineering comes into play as you are describing.

On the other hand, here in Phoenix, builders would object violently to long CMST straps, tight nailing and blocking in roof plywood, etc, and the plans checkers don't require them. The only notable failures here have been garages racked by windstorms when the concrete roof tiles have not yet been stacked on the roof combined with deficiencies in garage door jamb shear and hold downs.

That entire roof acts much better than you think w/o collectors etc - the way I explain it to architects, etc - imagine you have cardboard sheet about 3 feet long x 3 feet wide folded to the shape of that entire roof and you try to distort the cardboard in the sense that lateral shear would - the cardboard actually has substantial stiffness against buckling and racking as long as it was properly attached to little trusses a few inches apart.

But you can't use this analogy for seismic, why do I think it is allowable for wind? Perhaps the shaking and repetitive cycling accelerations with seismic - that is progressively making nails bend and pop out. At any rate, if I insisted on the collectors and straps, all my clients would probably leave me.

Look at the attached detail- about the highest shear I calculate typically is 400 plf, more than the basic roof sheathing is rated for, but with wind forces here probably very conservative.

 

[AELLC]

ps - I forgot -

If I was designing this house for seismic, I would calculate reqd length of roof deck for shear and then subtract out the length of the shear wall - the remainder would be the reqd length of drag collector.

 

[jeffhed]

msquared,
1 & 2. Right, I did forget about that 4'-0" offset. As I said before, seismic is always a concern as it is usually about the same magnitude of the wind loads (almost every house is tile roof also). As far as the magnitude of drag forces, that has often been my question as well. How much force can the diaphragm transfer where there is no edge nailing? In the past I have figured if the allowable diaphragm shear is 230 plf with 6" o.c. edge nailing, without blocking the nailing would be at 24" o.c. (or whatever the truss/joist spacing is). So in the case of 230 plf and 24" o.c. truss spacing leaves me 230/4 = 57.5 plf. Not much, but definitely within the realm of a majority of single story wood framed structures. We also have set the drag force at 2000 lbs before we worry about it. So if my unit shear in my shear wall is less than the unit shear in my diaphragm, the diaphragm shear is less than 57.5 plf and my drag force is less than 2000 lbs, then in the past I have not worried about collectors. Then I have a way to rationalize why no collector is required. I don't know if this is over conservative or not. a 2000 lb drag force isn't all that hard to get. But I have a hard time getting a handle on when I "feel" a drag strut should be provided without coming up with some kind of numbers. Too conservative?

3. OK. So you would consider what my colleague has suggested as creating one large subdiaphragm? Do you think that his feeling that the sheathing and nails is enough to bring the rest of the home along for the ride?

 

[jeffhed]

AELLC,
Here framers aren't all that adverse to providing coil strap to tie drag trusses to shear walls. We usually only make them coil strap when the truss ends at the wall. If it runs next to the wall we will use a detail similar to what you have shown that specifies some sort of shear transfer between the wall and the truss. If the trusses are perpendicular, in the past we have run the coil strap and blocking out far enough just to develop the drag force into the diaprhagm. If we had a 2300 lbs drag force and a 230 plf diaphragm, we would need to run straps and blocking out 10'-0".

I also agree that the diaphragm is much stiffer than what we would imagine. However, I like to have a concrete reason why I am ignoring something, rather than just saying we don't do it that way just because. If I have a rational reason for doing something, I feel like I am at least giving my clients something with some numerical backing rather than a guess that may not be correct. Perhaps I am just anal, paranoid (or a robot), but my mind doesn't like to accept feelings without numbers. I have only been licensed for 6 years, so maybe that is part of it too.

 

[AELLC]

jeff,

If a custom home builder sees all that (what I call advanced structural engineering) detailing, they add it to their estimate and the homeowner pays for it. Frequently no problem.

However I deal with a lot of tract and semi-custom, and to them, it would be like telling the Chevrolet corporation that all the their seats need to be covered in hand-stitched leather instead of vinyl and cloth.

We all need to have numbers and calcs to justify what we do, it is something instilled during University and helps keeps us honest. But houses have so many secondary load paths etc, we get into the non-engineered, well that's the way it is always done around here mentality.

It seems only in California, houses do fall down due to earhquake, and in tornado areas, there's no way to keep a house from being destroyed.

 

[msquared48]

AELLC:

Actually there is a way. You just do not have a view.

Mike McCann
MMC Engineering

 

[AELLC]

Mike,
Are all new tract homes in tornado alley now designed to withstand the worst tornado?

 

[jeffhed]

AELLC,
You are right with homes having multiple load paths. All those partition walls with sheetrock on both sides are definitely contributing, as well as the ceiling diaprhagm. I am in southern Utah (St. George) and the Santa Clara river flooded and washed out some homes on the banks. But many homes were left standing with one corner or the entire half of the home cantilevering over the edge with no foundation. They stood there for a week while the flood was occuring then they were underpinned, back filled and they are still there. Imagine what the engineering would be like to make a home do that. I guess the we have always done it that way is my stumbling block since I haven't been at it all that long. So having numbers helps me develop some engineering judgement. I was trained from the get go on commercial buildings and there is peer review on all commercial projects here, even lattice structures. So I have been trained from the beginning to have calculations for everything so that it stands up to scrutiny in peer review. Many of my clients do large custom homes, so the rules still apply, but on the smaller homes I would like to eliminate some of this stuff but be able to do it with piece of mind. Which can be hard to when you have been doing it one way for a long while. So I am guessing that you probably have no problem with my colleagues method of using a large subdiaphragm and the rest goes for the ride. Looking at our typical roof diaphragm which has 8d nails and 12" o.c. in the field and using increases of 1.6 for wind and seismic and 1.1 for diaphragm nailing I get an allowable load per nail of 133 plf. Which is higher than the diaphragm loads I get on many of the smaller homes. I have no doubt the sheathing can resist a tension/compression force of that magnitude when nailed off every 24" o.c.

 

[AELLC]

jeff,

Yes I have a more relaxed attitude about residential design after being an employee at 3 different all-residential firms.

The important thing is to calculate shearwalls and hold downs accurately, or else you slide down and get sloppy/negligent after a lot of years of doing this.

The other thing is showing accurate details that match the calcs.

 

[jeffhed]

AELLC,
Woking at multiple firms definitely is a help. I have worked at the same firm for 15 years starting out as a drafter while I was going to school and now as an engineer. Over that time I have had one mentor who is 65 and has pretty much turned eveything over to me. His comments to me most times is that's how we have always done it. He laughs when he sees the look on my face trying to accept it, without numbers and tells me he remembers feeling the same way. He proved much of this stuff to himself years ago, so long ago (probably 30+ years)that he can't tell me how he proved it to himself, just that he applied numbers to something years ago and isn't worried about it. I think doing all commercial design for my EIT years was great and helped me learn how to design all kinds of commercial buildings and with all kinds of materials. Then you move to residential stuff and it is hard to leave that commercial design mentality behind. But now that I am licensed and he is going to be gone in less than a year, I am trying to cement my engineering judgement as far as residential engineering goes. And asking questions and seeing how others are doing residential design also helps in developing my engineering judgement.

 

[AELLC]

jeff

I am 66 but work as a one-man shop. I can see how your mentor's attitude is typical, I am that way also. A lot of things that get established as standards and then you forget exactly how the numbers are run.

The important thing is to stay sharp, even though you may forget how to use classical hand-calculation methods that were taught at University. For example, I see a lot of questions on these forums about results of continuous beam and space frame analyzers - basic engineering principle answer a lot of those questions.

Myself, I can't even remember most calculus anymore (but you have to always remember how load-shear-moment-deflection diagrams generally work), but I think I engineer stuff better than ever. If something looks fishy on those diagrams, I can recognize it is not correct.

 

[jeffhed]

AELLC,
Thanks for all your input, it helps to hear others mentalities as far as residential goes. I just want to do things right, but I don't want to make people do stuff that they don't need to. I have also been taught by my mentor that it is important to have calculations in a lawsuit. So where I am just on the cusp of being on my own, I want to be able to have some sort of numbers for what I am doing. Even if it just a one time calculation I scan and keep on file for later reference. I think I am going to start looking into this subdiaphragm method more since it may be the easiest way to show some sort of calculation. Another thing is msquared48 has said to look at the magnitude of the drag force to determine if a collector has to be detailed or not. He didn't say what kind of magnitude he uses for that sort of judgement. I feel 2000 lbs or less is a fairly small drag load, what level would drag forces need to get to before you started to think maybe you needed a collector?

 

[AELLC]

jeff

This is the method I always use - see attach

 

[AELLC]

Also, do you guys calculate actual tension or compression in the wall top plate along shear wall lines...that was always a big deal when I was doing houses for CA? I got some interesting straps and nailings for those.

It was so tedious I have an Excel that does all that now.

 

[jeffhed]

AELLC,
Our lateral spreadsheet has a drag force diagram that we just enter in the walls and diaphragm length and it calculates the drag force for us. We then run blocking and strapping out far enough into the diaphragm to develop that drag force. The numbers will come out the same as yours. And yes, we do check the chord forces in the top plates. We also have a spreadsheet that we just enter in the diapraghm that is the worst case and see what the chord forces would be. Residential is almost always just the minimum nailing because I almost always shear the interior garage wall and tie it through so that keeps all of my diaphragm ratios to a point where diaphragm shear and chord forces are pretty low. Typical construction method here is to sheath the entire exterior, so that interior transverse garage wall is sheathed most times. I have attached a sample of our spreadsheet for a single wall line.