Cracked Wood Rafter 5

[PSUengineer1]

Thank you for consideration.

I viewed a cracked rafter in an attic. Crack is at bottom of 2x8 rafter which spans about 19 feet and spaced about 16 inches on center. The crack does not appear to be fresh. I have been told the crack is due to wind. I am not buying it. The wind speed was about 60 mph and there were about 5 shingle tabs blown from roof. Metal ridge vent did not blow off of roof. I suspect it would prior to wood framing cracking from wind load.

So, why the cracked rafter? Could it be; inadequate lateral bracing?,,,undersized for span? knots/imperfections in wood?..... bundles of shingles laid upon it during construction (but 8:12 roof slope)?....The crack is located on rafter where it meets wall below and where it meets ridge. I have attached photographs, please view them and comment.

thanks.

 

[BAretired]

There could be other factors, but the detail is very bad. Whether the cracks are fresh or old, there seems to be a clear need for remedial action throughout.

BA

 

[focuseng]

OK, this thread is getting so much traction I had to revisit the original question.
Wind? Not likely. I would expect more damage to the other rafters too this this much on one. Additionally, the overframing at this area would have something to say about this if if were true. The crack is not very "fresh" looking either. So if you are looking for a recent "NATURAL EVENT" or "ACT OF GOD" this is not it. That is unless something strategically fell on it due to the wind. But not damage the others?

My Official Guess: a bad detail with a improperly placed impact construction load and possibly a bad board to boot. I'm calling it "the convergence theory". The top crack is the most interesting to the bad board part. Almost looks like 1st: damaged, 2nd: water got into it and 3rd: then was left in the sun to bake before installing. The way it is separated along the grain and has a bit of a warp to it... The bottom most likely was damaged in place. Strange location within the roof for this one rafter to be singled out. Maybe someone dropped a sheet of plywood or packages of shingles or stepped on it wrong. Wouldn't take much -the bad detail clearly contributed to its demise.

jmjxs263: what did you report?

______________
MAP

 

[Splitrings]

I think we all realize the detail is poor. NDS limits end notches to 1/4 the depth and I have heard of 1/3 depth used frequently on residential projects. These rafters have been notched closer to 3/4 their depths. Another reason this detail is poor is it induces tension perpendicular to the grain when resisting thrust(if the notch resists the force). My point was that one needs to be certain what the caused of failure was prior to suggesting a repair.

Maybe those smaller cracks are from tension perpendicular to the grain caused by the end condition? Then a better detail would solve the problem. I question why a notch is required. Resist the thrust a different way.

 

[manstrom]

Rafter framing is funny. If you sit down and do the engineering, most of the connections just don't work. A lot of times, the thrust reactions are enormous. If you get into multiple gables and levels, the engineering is almost impossible.

The IRC has huge holes in it as far as rafter framing goes. The prescriptive code does not properly address the collar tie connections. I started a job using prescriptive IRC design and scrapped it when I saw a 10x factor of safety concern.

I won't even touch a rafter framed project unless it is very simple. No one knows how to build it and an architect doesn't understand how to design it. Trusses are the way to go.

In your particular condition, the joist cracked because the bearing condition is causing stress across the grain.

 

[Archie264]

Well, on that note, residential structural engineering in general can be a real head-scratcher. Many situations exist that don't work "on paper" and yet they've held up for decades. Sometimes I wonder if wallpaper has a structural component that holds the whole thing together. Residential structural engineering can really be it's own sub-specialty.

 

[manstrom]

I agree. Some stuff "just works". Doing a full analysis doesn't take into account all of the factors in a design. That is often impossible.

What concerns me is the slippery slope of prescriptive design.

"This (accidentally) worked before, let's do that 100 times"
"If that works for a 10' beam, I bet it would work for an 11' beam"
and so on

The ICC has caught some of these errors and has tried to fix them at the reluctance of builders. For example, basement wall bracing was addressed in IRC09 to change wall bracing from 1/2" bolts at 6' o.c. (not even close to working). I was happy they changed it since it was one of my gripes. After enough prodding by builders, they removed the structurally adequate section from the code. "It just works". Sure it works, but try that on a 15' wall and tell me how that works out for you.

/rant

Be careful. Sometimes the prescriptive stuff doesn't work.