multi-ply continous wood beam fastening

[eng003]

For a typical continous multiply 2x wood beam I often see detailed lag or through bolts staggered. The required fastening to develope the plys on either side of splices at zero moment is not somehting I would typically try to quantify exactly ( I would opt to over do it) but I am questioning whether this fastening schedule is really needed. I wouldn't care but when you finishing out a beam that will be sean it would be nice to use just nails or smaller countersunk structural screws. Any comments and thoughts on how to easiliy check a fastening schedule.

 

[shobroco]

The fastening of plies has an awful lot to do with the way the beam is loaded. If a beam is loaded from one face, the fasteners must transfer the load to the other plies. If the beam is loaded from the top, the fasteners must simply hold the plies together, providing lateral restraint if an unrestrained beam or simply making one feel good if the beam is restrained by other elements. Generally speaking, I see way too much overdesign in the specs for fastening, but this is probably because that's easier than actually figuring out what is required and who is going to be responsible for ensuring that it's done correctly.

 

[TXStructural]

Fastener installation and wood quality are highly variable. You can certainly use the NDS to compute a nailing pattern that will be effective. Nails always ultimately fail in withdrawal, so restraining the plies can reduce this mechanism. This week, I spent time working on a building that had been blown apart, and the 100+ foot long multi-ply beam across the edge of the roof was nailed together and was holding up nicely as it hung from a balcony rail and down across a stairway at what was not mid-span of a catenary.

What I found when designing 100's of truss repairs on a project some years ago was that it is very difficult to develop full tension capacity of a 2x4 using nails over less than full length. There should be no trouble doing this where you need to bond the entire length of the members with a pattern like 8-10d common nails at 6-8" o.c. staggered. Use of truss plates or bolted connections was must less trouble and much quicker, not to mention more easily inspected and more certain to be done correctly. In your situation, it sounds like the use of other fasteners, or even adhesive, would be useful. A field-glued lamination could be designed. 3m makes a tube adhesive for structural use with wood, or even a PVA wood glue (like Titebond) could be used with proper construction. You might even use bolts or nails for the structural plies, and cap the exposed surfaces with appearance-grade plies glued in place.

For reference about wood glue strength, there is an article linked here:

http://www.titebond.com/news_article/12-01-18/How_Strong_Is_Your_Glue.aspx

 

[focuseng]

Nice little article about fastening with structural wood screws. Goes into the specifics of side loading fasteners % attributable and such. All the Multi-ply beam fastening tables by the LVL mfgs I have seen use values that correspond to this.

______________
MAP

 

[eng003]

Thank you all for the responses. I have done some calcs and put more thought into this since my original post. I am not looking at what it takes to transfer shear on a simply support beam, I have seen tables from fastner manufacturer's on doing this. I think I need to define my question a little better. The specific application I am looking at is when one trys to make a continous beam from multiple plys of 2x sawn lumber by positioning splices at the inflection points. I have seen other engineers use this technique frequently using 1/2" lags @ 12" o.c. staggered to fasten together.

First I would like to get others thoughts on the merits of doing this as opposed to just designing simple spans spliced at the bearing points, my thoughts are as follows:
1. Assuming a two ply beam in a two span condition you do NOT reduce the max moment by splicing at the inflection point.
2. In more than a two span condition you DO reduce your max moment.
3. Assuming a two ply beam in a two span condition you DO reduce your max deflection (you make a stiffer beam)
4. You need to check shear for one ply at the splice but this does not appear to control.
5. Assuming often you find deflection controls, even for a two span condition trying to make a continous beam may have its merits and for greater than two span it would have even more merit.
6. But what about constructibility, defining an exact location of a splice point and a fastening schedule is the exact kind of thing I imagine a contractor would often ignore.
7. Doing a continous beam has the added benefit of taking the splice away from usually an already congested connection point at a column where there may be limited wood (meat) to bolt and bear on.

Second I would like to get others thoughts on how to properly analyze a fastening schedule say for a two ply continous beam condition spliced. My thoughts are as follows:
1. Assuming a two ply beam held with two rows of fasteners (top and bottom) I applied the shear flow equation VQ/I with my Q calculated as half one ply to evaluate the required spacing for one of the rows, is this a correct application of VQ/I? most of the examples I have seen have applied this to say making an I beam out of a wood member... and using the base of the I for the Q, I am applying it a little differnt here. Based on my calcs for say a small double 2x8 beam I find that even 1/2" thru bolts at 12" o.c. do not even make it in the areas of high shear? You would need something on the order of 2" or 3" o.c. to make it work. Of course you could look at average shear flow in zones as the required spacing quickly increases and assume the loading is distributed among fasteners in the zone but is this accurate?
2. Thru bolts and lags take a lot of meat out of the beam, what about loss of cross section for bendingg?
3. Thru bolts and lags unless countersunk protude from the face of the beam which make it hard to finish so I would prefer to use structural screws but the thru-bolts barley seem to make it load wise?

Lastly, I came accross the following article from FEMA (see attached file link) where they suggest positioning the splice 12" from the pile?? I am all for doing this as I am well aware fo the congestion at these type of connections considering NDS edge distances etc. but what is theory behind this? are they assuming simple span and just moving the splice point? seems to me there would be a great amount of shear flow that would have to be resisted if the splice point was placed just 12" from the bearing point requiring a very close spacing of fasteners?

I appreciate any comments!

 

[gte447f]

If this is for a residential project, especially if it is for a small scale residential remodeling project, I would not even consider designing a spliced multi-span beam. Maybe a continuous beam with no splices, but preferably just simple spans. I say this because, based on my experience in this market, the contractor is likely to be a drywall mudder or other such unqualified person claiming to be an experienced framer or GC. Your 2 most important design considerations, in addition to a sound structural design, should be:

1) familiar, readily procurable materials (in other words, if they don't have it at Home Depot don't bother specifying it, and

2) ease of construction (in other words, make it as idiot proof as possible)

 

[focuseng]

I agree with gte447f. I don't honestly see the need to splice a wood beam off a support except in very specific and unusual situations.
Using your example of a 2-ply 2x8 double span: How long will each span be? I'll bet I can get the material length to make both spans easily or add another ply or make it 2x10 to get stiffness. What is more expensive - 2-2x10 or 3-2x8 simple spanning or a special spliced 2-2x8 with a #$@!# load of fasteners and the time to install them?

Also what is the fastener slip contribution to deflection at a splice joint?

With the availability of EWP up to the length of the truck you can get on site to deliver there is no real reason to specify a lapped splice in a EWP beam usually given the spans with which wood is likely to be use in (44' flatbed is not unreasonable). I can easily get 2x12 stock up to 24+ feet. Handling is another thing. Framers like to make shorter pieces and manually lift one ply at a time. If the job is that big and the beam is that important then they should have a crane or lift truck on site otherwise they may just not be a "qualified" contractor (IMO) or you should have designed a simple or shorter set of spans to accommodate their talents.

If a beam is required to be continuous or jointed over a support then I put the pertinent instruction on the plan about where these need to be. I have not specified splices at inflictions points -only at support locations for reasons above. I do joint beams over supports selectively to either reduce/control reaction loads or to make sure the overall length of the material is not too long or the beam is stiffened.

Other thoughts:
1) Shear or Moment usually does not control in long span. Deflection is the killer and I want that beam continuous and not "released" for this reason.
2) splicing at inflection points seems like an unnecessary complication for framers who still think a "carriage bolt" is equivalent to a standard hex with a shoulder (what is the drill hole size for a 1/2" bolt?) or couldn't tell you what the difference between a sinker vs a box nail vs that thing that comes out of the nail gun is. If I am lucky I will see maybe 30% of my projects shop drawings for framing layouts in residential wood design, much less get a conversation with the contractor about tricky spots (before he/she messes them up).
3) I don't have time to design splice joints. Gimme 2 ply's continuous - done.

Is this getting to your questions?

______________
MAP