Ridge beam undersized--any ideas on how to reinforce it now that its in place? 1

[Qwikone]

Members-
I was called to look at a 2-2x12 ridge beam spanning 11'9" on a mostly finished porch roof yesterday afternoon. The rafters are 2x8, and the roof pitch is 5:12. The issue is that the inspector noticed a difference in depth of the 2x12's, one is 11-1/2" and the other about 11". He said that the Contractor needed an Engineer to determine if the ridge beam was adequate.
I ran StrucCalc on it and it is undersized by 6% for Moment. Required depth is 11.58"
I am not coming up with any solutions to alleviate the issue without replacing the existing 2x12's with an LVL set.
I know this Board has many folks who must have come across something like this in their travels-- so does anyone know of a solution that could correct the the the existing 2x12's without taking them out?
DusterMick

 

[KootK]

For 6%, I'd just leave it alone. Find a way to justify slightly lower loads or a shorter span etc. I suppose you could just nail a 2x4 on the flat to the underside with a 1/2" shim. That would have to be worth 6%.

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]

I agree with Kootk. Also, why are you using StrucCalc on something like this? It is a 30 second calculation.

 

[Qwikone]

KootK- The County wants a sealed letter, and I'm not concerned with the 6% either, just looking for a solution that would put it back in the green.
I may bolt/screw something to the bottom......

XR250- Just lazy I guess.......or old. LOL

 

[oldestguy]

What does it look like as to defects, knots in the critical places? How about the working stress as compared to ultimate? Who sets that allowable and why?

Adding a steel strap underneath as a possibility?

 

[RPMG]

Can you lower your risk category/importance factor? Or reduce your roof live load? Refine your dead load? Use LRFD?

 

[XR250]

FWIW, I don't worry about overstress in lumber unless it is over 20% or so. I am more concerned about serviceability. See alot of stuff sag that is not overstressed. I rarely see wood members fail. There is just too much redundancy in the systems, the lumber is much stronger than we give it credit for and the loads are rarely as much as code requires.

 

[RPMG]

I'm not a huge fan of overstress limits because they're arbitrary. Admittedly, all the things listed add strength that is unaccounted for. But there's no breakdown, or reasoning involved. It's just a limit based on how one feels.

With that said, the analysis is not significant to 3 digits (106%), and I wouldn't reinforce this beam, unless I believed the beam was undersized to begin with.

 

[wannabeSE]

Why is one piece only 11" deep? If someone cut it from a 2x14, the lumber grading is no longer valid and it needs to be re-graded or replaced. If it is a 2x12 that is unaltered and shrunk a lot, I would call it a 2x12 and use the 11-1/2" in calculations.

 

[thaidavid40]

I am intrigued by the variability in the width dimensions between the two pieces! Around here, a dressed 2x12 is uniformly 11-1/4" wide, with next to no deviation (maybe 1/8", at most?). How do you get a 2x12 that swells 1/4" to become 11-1/2" wide? And how does an already dried 2x12 shrink that quickly to 11"? I'm with @wannabeSE here, these sound to me like maybe "homemade" 2x12s.
Dave

Thaidavid

 

[Qwikone]

I agree there are many "issues" with this ---I am meeting the folks on site tomorrow to get the real scoop. The dimensions I have been given (and pics) show a difference in depth of the 2x12's. The problem is also with the dimensions of the porch. I am getting the impression after speaking with the Contractor that someone is trying to avoid doing this right.........The owner wants it fixed.
I'll reply after I see what the situation truly is.
Thanks for all your valuable input and responses!!
Dustermick

 

[racookpe1978]

An added piece (underneath) is attractive - BUT - if the two existing members are not the same height, then you'll have trouble "flattening" them to the same height to allow a legitimate bond underneath the two 2x12 (nominal) members.

Appearance matters - since it appears they are visible. A sideplate of 8x1/4 plate would be easier to mount, to reinforce the member against vertical loads (across the center of the span ??)

But the true calculation needs to be correct as well - i am VERY surprised that a 1/2 of wood on a doubled 2x12 member actually lets you fall under the Code strength requirement! Check very, very carefully EVERY assumed condition for the loading.

I would assume a nailed-together-doubled-2x12 member would be OVERSIZED by at least 30% to 50% based on minimum stress levels for commercial wood and nails.

So why is this doubled member only 102% as strong as design theory says it should be?

 

[jayrod12]

I would bet a good refinement on the dead load would bring that back down 6%. A double 2x12 spanning 12' feels right for a porch. Add a bunch of nails between the roof sheathing and the beam and run it as a composite beam if you're really stuck.

But in all honesty, any engineer that has worked with wood more than a few times won't even take a second glance at an existing condition of 106%.

 

[Archie264]

Overstress limits may be arbitrary but so too are imposed loads. Arbitrary, at least, in the sense that snow, wind etc. aren't going to be exactly 30 psf, 20 psf, or whatever was stipulated by the governing boards and/or used in the calculations.

Or, to look at it another way, is there enough interaction between the sheathing and the beam to contribute 6% worth of capacity? Who knows? But the likelihood of it probably falls within the same realm of statistical uncertainty as guided the development of the proscribed imposed loads.

My mentor's mentor had a slide rule with a broken cursor on it. He used it to design many buildings, explaining to his protege that that was all the accuracy he needed...or could rely on.

I'd leave the beam alone.

 

[mike20793]

I think the issue lies with the sealed letter. While I think most of use would leave the beam alone, would most of us seal a letter stating the beam is adequate after its adequacy has been questioned? I'm not sure I would from a liability standpoint, just in case the beam failed and my calcs show it overloaded.

While I agree the beam is okay and isn't worth strengthening, I'd try to justify a reduction in loads to bring the member closer to capacity so you can seal the letter and feel better about it. If they ripped a 2x14 down to a 2x12, then that is a whole different issue.

 

[KootK]

While I think most of use would leave the beam alone, would most of us seal a letter stating the beam is adequate after its adequacy has been questioned?

I've been thinking about this too and I would be willing to write and seal such a letter. And, rather than discussing difficult to quantify bonus items such as diaphragm action etc, my preference would be to simply state that I feel that a 6% calculated overstress is acceptable given the degree of accuracy inherent in our work. I feel that such a response would be the most honest. I would, of course be curious to see how such a letter would be received by the reviewing authority.

I feel that it's a logical error to make arguments such as "loads are less than we calculate" and "wood is way stronger than we assume". It is my understanding that these things have been statistically calibrated to produce, to the best of our ability, a uniform level of risk among our various construction materials. Just because wood has a higher COV than many of our other materials, that doesn't justify FOS encroachment in my opinion.

With that said, the analysis is not significant to 3 digits (106%)

Of all the "leave it alone" arguments, I like this one the best. Unfortunately, round would put us at 110% instead of 100%. Doh.

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]

@Kootk;

I wonder if the statistical calibration takes into account the inherent redundancy in light framed, wood structures? Probably not.

 

[Archie264]

Mike20793, that makes sense. I think of Humphrey Boghart's line in Casablanca, "I don't stick my neck out for anyone." To which I would add, "...at least not in this litigious age."

 

[Archie264]

Kootk, that makes sense to me too. I guess it's all in the way the information is presented. "Managing expectations," to use a popular term.

 

[JAE]

One thing that is also probably (maybe?) present here is that there may be some additional capacity in the overall system.

For most of us structural types - we tend to always try to analyze a structure based on very discrete load paths that are mapped out in our minds, and also in our calculations, such that we tend to ignore other load paths that do exist. It is just easier and more conservative to map out the "load follows stiffness" path and ensure that the path is adequate.

One of the features of wood framing with sheathed walls, floors and roofs is that the load, in reality, gets spread out in a lot of different ways.

For this particular case, my first thought was: "why don't we look at the fact that a ridge member also has sloped rafters that may be propping it up".

But with a discrete load path we'd also reason" "nope - that can't work because there isn't an eave beam that can take the horizontal thrust of the rafters."

But in reality - the sheathing on the roof does act as a sort of large leaning beam or diaphragm that does inhibit the vertical deflection of the ridge beam to a degree.
The problem is - how do we quantify HOW MUCH it helps? The steeper the pitch the more it helps. It also has to have adequate connection to an adjoining roof or system to prevent deflection as well.

But give 1.06 as a unity on this design - and knowing,...knowing...that the sheathing does help - I would be more inclined to write the letter at 1.06 and say its acceptable.

Check out Eng-Tips Forum's Policies here:
faq731-376