Wood Gable Roof Diaphragm 6

[Jerehmy]

In my Breyer book they don't go into any detail about the discontinuities of a roof diaphragm at the ridge due to ventilation requirements. In the Breyer ridge detail (Page 9.28 Figure 9.10e of the 6th ed.) they show the sheathing as continuous. Not sure how that's realistic.

For the job I'm currently doing, I have a gable roof. The attic is finished so I was just going to use the rafters to transmit the lateral wind load into the attic floor diaphragm. But what if the attic is unfinished, how have some of you resolved this? Treat each roof face as two separate diaphragms?

You could put blocking at the sheathing edge near the gap at he ridge, but this would disrupt air flow from the eave is the ceiling is finished. But I guess you have a finished ceiling, you'd have an attic floor to use as a diaphragm. So is that how you guys do it?

Curious as to what others do. Breyer advises to use "ATC Guidelines for Design of Horizontal Wood Diaphragms 1981" for steep roof diaphragms, I might have to pick it up. Anyone else us it?

Thanks

 

[ChiefInspectorJeff]

The problem is the wood butchers trying to build a precision engineering product! A trim carpenter would be a better choice.
Most inspectors do not look at the top plate nailing so if that is defective you have no load path anyway.

 

[jdgengineer]

Our typical ridge detail involves a structural ridge (homes in my area are generally very custom / vaulted where trusses dont make sense). We drop the ridge down 1.5" and put in a 16" long flat 2x block on top of the ridge. The diaphragm nailing is clustered to the block and there is roughly 6" of airflow space available at each rafter bay.

 

[KootK]

I don't quite get your detail jdengineer. I'd like to though. Can you elaborate or post a sketch?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[wannabeSE]

FYI
Woodworks has some workshops entitled "Advanced Diaphragm Analysis – A Technical Presentation of Complex Diaphragms" with Terry Malone.
They are reasonably priced. For locations, see

http://www.woodworks.org/events-calendar/upcoming/

 

[Guastavino]

This is a great thread. It's one of the reasons that I pretty much don't do wood design. I would love to do it, but getting wood to be designed correctly is difficult enough. It seems as though getting a complete load path is next to impossible with the way wood buildings are typically framed and detailed (by that I mean non-structural components). That, and contractors seem to think you're crazy if you do detail a complete load path.

My experience has been that some of the other engineers in my area are ignorant of complete load paths and I can't compete with them.

Anyway, to the thread subject. one thing to consider is that if the building is symmetric, for the wind load parallel to the ridge, the shear should be close to zero at the ridge (IE the midspan). Obviously that's not the case for wind load perpendicular to the ridge. The only major wood building i designed that had gable roofs had spray on insulation on the roof sheathing, and therefore didn't have ridge vents.

As for considering the roof as two separate diaphragms, personally, I wouldn't do that. The diaphragm really needs to be rigid or close to it in my opinion to consider that.

With all that considered, we structural engineers aren't building a swiss watch, but it's easy to get bogged down into the details and stress the small (IE small gaps) details

 

[RFreund]

this may have already been mentioned:

There is APA literature called "Introduction to Lateral Design" which states that "A continuous ridge vent can be used with an "Unblocked Diaphragm". An unblocked diaphragm has no requirement for blocking between unsupported edges of adjacent panels. It also states you could stagger the blocking in a "blocked diaphragm" allowing vents every other truss bay. You then provide double the nailing requirement at the blocked bays. So you have two choices: design the diaphragm as unblocked or stagger your vents. I don't know your ventilation requirement, so that could dictate which one you use.

I stole this from a previous thread:

http://www.eng-tips.com/viewthread.cfm?qid=184481

EIT

http://www.HowToEngineer.com

 

[KootK]

Nicely done RFreund. Sadly, I already have that document but never got around to reading it.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[medeek]

When you really think about it an unblocked diaphragm could just as well have a row of vents along every unblocked seam (perp. to rafters/trusses) and still be qualified as a diaphragm. I think we have been thinking about this all wrong. The ridge vent seam is then no different from any other seam other than it is a little more opened up and it occurs at the center of the diaphragm. At some point though this gap width is going to impact the overall performance of the diaphragm. My question is when does it start affect the performance, strength and deflection. (Note: I am specifically talking about unblocked diaphragms here.)

The APA should do some ridgevent/diaphragm testing to better quantify this behavior. My suspicion is that the minor discontinuity caused by the ridgevent air gap is so slight that it does not affect the performance in a measurable amount when the stresses are low (ie. unblocked diaphragm, typical residential roof).

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

http://www.medeek.com

 

[KootK]

I proposed the unblocked diaphragm strategy at the top. In my mind, there is a fundamental difference however. For a normal unblocked diaphragm, you transfer shear across a continuous framing member that spans the sheets of plywood. At a ridge, you're transferring that same shear across a joint between framing members that likely has questionable weak axis moment/shear transfer capacity (truss plate). From what I've seen, there is a lack of testing on pitched diaphragms in general, for both plywood and steel deck.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[jeffandmike]

The ridge vent seems to be the primary (though not necessarily the only) problem in this discussion. Why not omit it, and get your ventilation elsewhere... like down at the eave? Aside from the diaphragm issues discussed above, ridge vents eventually leak at the poorly (non-existent) sealed fasteners (nails) that eventually withdraw (or completely come out), or are at least susceptible to infiltration of wind-driven rain at higher wind speeds (if you doubt this, just Google "ridge vent leak"). Common sense states that the more penetrations you have in your roof, the more likely you are to have leaks. Problem avoided = problem solved (at least in this instance)

 

[Jerehmy]

Because there already is eave ventilation. You need top and bottom side ventilation so the air can flow. Having only ventilation at the eave wouldnt do much of anything. Gable end vents count as top side IIRC so thats out too.

 

[jeffandmike]

Jerehmy,

As far as I know, there is nothing in the code that typically requires "top" ventilation, as you refer to it. "Cross ventilation" is required, but this can be accomplished by having ventilation at the eaves on either side of the roof, as I previously described. If 40% - 50% of the ventilation is not located in the upper portion of the attic, the "minimum net free ventilating area shall be 1/150 of area of the vented space", as opposed to 1/300 (

http://publicecodes.cyberregs.com/icod/irc/2012/icod_irc_2012_8_sec006.htm).

Essentially, there is a "penalty" applied due to having slightly less than desirable ventilating conditions; however, my point was, if you're already providing ventilation down at the eave, why not just increase it, avoid the possibility of leaks through a ridge vent, and remove most of the concern about analysis and diaphragm force transfer at the opening created by the ridge vent?

I would further, maybe naively, ask the question... if you must have a ridge vent, and we assume the building is rectangular in shape, with the ridge vent running parallel to the long direction, and being at the center of the roof (midspan of the structural roof members), and with lateral (i.e. wind/seismic) forces being applied perpendicular to the long direction, exactly what forces are you really concerned about? Visualizing the horizontal diaphragm as a I-beam (

http://nisee.berkeley.edu/lessons/a_figure26.jpg),

tension/compression forces are resisted by the flanges (i.e. the diaphragm chords down at the exterior walls/eaves), with the bending stress being lowest at the neutral axis (i.e. the ridge)... correct (

http://www.expertsmind.com/CMSImages/761_Determine the maximum bending stresses.png)?

So then, what you have left to be concerned about are the shear forces in the diaphragm, which are highest at the ends of the "beam" (

http://www.awc.org/calculators/connections/images/diaphragm_pic.gif).

Restrict placement of the ridge vent to within the middle third or "middle half" of the long direction, where the shear forces are lowest. Have you seen these beams before (

http://quick-calcs.net/designs/images/holes.jpg)?

Those are some awfully big "ridge vents" in the web of the beam.

I personally think we're making a mountain out of a molehill here. Where are the widespread structural failures of wood-framed, single-family residential roofs, at the ridge vent, due to wind/seismic forces?

 

[KootK]

As has been pointed out a few times above, the peak shear demand across the ridge would come from loads acting perpendicular to the ridge. Additionally, even for loads parallel to the ridge, it`s hardly reasonable to assume perfect symmetry in loading.

I myself mentioned the lack of evidence for real world diaphragm failures above. However, this becomes quite problematic for designers. All the codes, articles, books, and seminars make it plane that the expectation is for wood diaphragms to be rigorously designed for lateral loads. When we choose not to do this, for reasons that seem vary pragmatic, it puts us in a precarious position ethically.

Large scale wood residential projects are notorious for ending up in court. I shudder to think what might happen if some savvy lawyer gets wind of the fact that most of our wood diaphragm designs are full of load path holes that we`ve knowingly turned a blind eye to. I imagine my day in court unfolding like this:

Prosecutor: This building leaked! Now you owe $1B KootK.

Me: No, I`m a good engineer!

Prosecutor: If you`re such a good engineer, why did you omit ridge diaphragm blocking that has been clearly documented in industry publications as a requirement for diaphragm load transfer?

Me: *tubmle weeds rolling bye*

Me: Because I`m SMARTER than the codes damnit! And a bunch of randoms on the internet agree with me!

Prosecutor: put the check in the mail.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[Jerehmy]

Nothing requires it, but in the experience in our office, it's never a good idea just to have eave vents. A good portion of the claims we
do is with attic ventilation issues. And most of them are on newly built structures. Just having eave ventilation doesn't seem to do the trick. So yes it's allowed by code,
but with all the issues we see with attic ventilation, we wouldn't do ventilation that way. At least not in our area where it can get pretty cold making ventilation more of an issue.

I don't see how the air would flow with just eave ventilation anyways. The cold outside air isn't going to push the warm inside air out. The warm air
rises and sucks the cold air in out the ridge. I'm not an expert in thermodynamics so maybe there are other mechanisms at work. I'm just going based off
our offices stance on attic ventilation. We also don't like those fibrous ridge vents. They always seem to get caked with snow. I've actually seen a few with mold in them.

 

[Jerehmy]

Warm air rises out the ridge and sucks the cold air in the eave*

 

[BUGGAR]

No time to draw a picture. Saw a condo being built here in SoCal with conventional double 2x12 ridge beam, roof sheathing both sides boundary nailed to ridge beam, 3" holes drilled through roof sheathing just outside the ridge beam on both sides and spaced so only a percentage of the diaphragm was penetrated, then a wide ridge cap was placed covering the holes. The composition roof tiles were held back to expose the holes but were covered by the vent.

 

[medeek]

@KootK

You're absolutely right, at the ridge or any other pitch break not only is the diaphragm discontinuous (unblocked diaphragm but the rafters or trusses are also "broken" at these locations. In the case of MPC wood trusses all of this load will be shouldered by the truss connector plates and the wood-to-wood contact of the chord members.

I'm sure there is some way to model this behavior in RISA or some other method.

My thinking is that this location (center) of the diaphragm is not critical so that is why we don't see any failures in the news from ridge vents.

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

http://www.medeek.com

 

[KootK]

My thinking is that this location (center) of the diaphragm is not critical so that is why we don't see any failures in the news from ridge vents.

Not so. For lateral load perpendicular to the ridge, the shear at the ridge, and adjacent to the gable end, will be the maximum value of the entire diaphragm for load in that direction. And since gable building trusses generally span the short direction, that often makes the shear demand at the ridge the critical value for either direction.

If you really want to give yourself a headache, contemplate shear transfer in the sheathing across hip roof ridges where there is no blocking along the ridges which is, of course, always the case.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[XR250]

Has anyone ever even seen a wood frame roof diaphragm failure?

 

[ChiefInspectorJeff]

Here is the new 2012 IRC roof ventilation requirements:
R806.2 Minimum vent area. The minimum net free ventilation area shall be 1/150 of the area of the vented space.

Exception: The minimum net free ventilation area shall be 1/300 of the vented space provided one or more of the following conditions are met:

In Climate Zones 6, 7 and 8, a Class I or II vapor retarder is installed on the warm-in-winter side of the ceiling.
At least 40 percent and not more than 50 percent of the required ventilating area is provided by ventilators located in the upper portion of the attic or rafter space… with the balance of the required ventilation provided by eave or cornice vents.