Notches in sawn timber beams 1

[Settingsun]

Using my local code, a sharp notch on the tension side of a sawn timber beam reduces the bending capacity to about 10% of the un-notched capacity, varying depending on the details of the notch and type of timber. The notch is where a post was previously located but has been move to one side leaving the notch in the span.

Say the beam is 300mm deep notched to 230mm. If I had an un-notched 230mm beam, capacity is 59% of the 300mm beam. This would be adequate due to reduced loading and favourable geometry, but 10% from notch calcs is not.

Is a notched beam really that much weaker than an un-notched beam of same depth, ie the extra timber is bad? How do they fail? I picture splitting along the grain so the notched beam will turn itself into the 230mm beam under load (this wouldn't bea danger as access is restricted), but I guess it could also fracture at an upward angle through the grain.

 

[MIStructE_IRE]

Compare the moment capacity of the reduced section to the moment at that particular point - not against the maximum moment.

If the notch and the point of max moment coincide then it is what it is and the carpenter should be taken outside and shot!

 

[dhengr]

Steveh49:
Basically, the notch does turn the beam into a 230mm deep beam. But, the notch is a reentrant corner in a tension stress field, so as the tension stress tries to move around the corner of the notch it is a significant stress raiser. You are right that the tendency will be to split along the grain, in a horiz. fashion, like we consider horiz. shear (or shear stress). For this one case, look to see that the member grain is running horiz., and straight, around (over) the notch, parallel to the bot. edge of the beam, and you are probably o.k. with your thinking about the beam’s action. It is another story if the grain is running at an angle which a crack might follow, or if there are other knots, imperfections, etc. in the immediate area of the notch corner. You can also improve the notch corner slightly by cutting away from the corner, at an inclined angle down the bot. edge of the original beam. This causes the tension stress field to move around the notch corner in a less abrupt fashion.

 

[r13]

I've rarely practiced in wood, but I assume wood grain is not a perfect horizontal, straight line, so the cracks started at the corners of the notch will continue to lengthen by splitting along the grain, then leave you with a beam with non-uniform depth over the split region.

I suggest to replace the beam, or double it up. Other than strength requirements, check deflection and other services criteria.

 

[r13]

By the way, do you know why the middle post was removed at the first place? Obviously it wasn't done under directive of an engineer, as he/she would have required replacing the beam while removing the post. Does the owner require permit to make structural change?

 

[Settingsun]

Thanks everyone for the responses. I suspect the old post had rotted and a new permanent post was installed to the side before removing the old post, to save the temporary propping stage required to install the new post without offset. Just one of those cases where the code says it should have failed whereas it's standing without any sign of distress.

 

[r13]

Just one of those cases where the code says it should have failed whereas it's standing without any sign of distress.

1) Conservatism in material properties, and safety factors in code.
2) Either the environment producing loads have changed, or the member hasn't encountered the loads with intensities that match the design.

 

[moltenmetal]

@retired13- you forgot the important word "yet" in your otherwise excellent sentence.

 

[r13]

moltenmetal,

Thanks for your feed back. Can you rephrase the sentence for me? As I always struggle with proper use of English language in writing.

 

[r13]

moltenmetal,

Never mind. I get it. You are correct, it "hasn't yet...". Hope I didn't make mistake again. Thanks.

 

[KootK]

I believe that the section loss and notch reduction factor should be cumulative in so much as the reduction factor in your code is intended to account for stress concentrations.

The section loss is a given of course. And, because your beam wont be smart enough to know that it's not supposed to make flexural use of the full cross section beyond the notch, the notch will create a stress concentration within the reduced cross section. So you're probably stuck with both.

As you know, wood failures are mostly stochastic things heavily influenced by local imperfections. In that regard, I'd be inclined to say that, if you're okay so far, you'll likely be okay for a good long time so long as no additional parameters change. If there was a local imperfection likely to set things off, I suspect that your beam would already be lying on the ground. And if there's no local imperfection, you're beam capacity is probably 400% of what you're calculating.

In reducing the section with the notch, you're also kind of messing with the original lumber grading by possibly taking imperfections what were formally in non-critical zones and moving them to critical zones. I see this as pretty minor given the scale of things with this though and would not be inclined to lose any sleep over this aspect.