Deck Connections

[dhoward26]

I'm looking at decks a little more in-depth than I have in the past and just thinking about framing options. Do most people use knee braces or do you use x-bracing or do you assume the decking itself acts like a cantilevered diaphragm? It seams that if you have the tension ties to the house to create a diaphragm that is 2-3:1 then you wouldn't need knee braces and could assume a cantilevered diaphragm via the decking connection to the deck joists.

Any thoughts?

 

[dhoward26]

Should have been a little more clear. I'm referring to the beam/column connections and whether people us knee braces or other methods.

 

[Archie264]

If it's attached to the building you don't have to have knee braces, though wouldn't hurt to have them.

The following site has some good guidance. According to the info on the first page it's based off Publication DCA6 by the American Wood Council.

http://www.fairfaxcounty.gov/dpwes/publications/decks/details.pdf

 

[SLTA]

The problem I have with knee bracing and X bracing is that architects, owners, etc hate that. I end up using a bunch of Simpson hardware, but I'm not in high wind or seismic areas.

 

[dhoward26]

Archie264: That document says that anything above 4'-0" off the ground requires knee bracing, cross bracing, or x-bracing. It is based off the APA prescriptive paper which also states that knee bracing shall be installed unless you orient the decking at a 45 degree angle to the deck joists. Makes me wonder what exactly is appropriate now because I have designed both ways, with and without knee bracing based on the architects requirements. I attached the APA portion, and it's under "Deck Stability".

 

[Archie264]

dhoward, I must have missed that requirement in the document. I'll look through it again.

 

[XR250]

Horizontal planking is not a very effective diaphragm. I usually put a planar, flat diagonal brace under the joists attached to 6x6 pt blocking at each end. At the house, the block is thru bolted into the house band. At the exterior end, it is thru bolted to the deck band and end nailed between joists. I space the braces to keep the loads reasonable.
You can also flagpole your columns out of big footings as well is the deck is not too tall.

 

[focuseng]

There has been quite a bit of research done on this exact issue by Virgina Tech U together with Washington State U and specifically Frank Woeste. DCA6 is becoming the go to code reference document based on their findings.
Decking placed diagonally is excellent for making the deck more rigid as opposed to perpendicular to the joists (no hidden fastening systems!) but the decks tested (great videos BTW) were pretty resilient to damage even if the deflections were out of this world, like in feet not inches. Composite decking vs. wood seems OK. PVC?
Sorry I don't have a good reference doc or link or anything but start googling those names mentioned and "deck" & "lateral". Also Simpson Strong-Tie has jumped on the bandwagon and has a few good references.

Personally I want to use diagonals braces and not little ones 2' down from the top either. Unfortunately as mentioned, nobody likes them and the bigger they are the worse it is. There have been times where I have insisted on full diagonals however and the connectors are not generally just a couple bolts for these. Depending on the client diagonal braces are a good sell in some form or another. Usually I am resorting to Tie-back method with Diagonal decking if I can get it. If the diagonal decking is a no-go then I put in more braces and/or tiebacks. I have tried something like what Excel mentioned but the size and # of diagonals placed on the underside gets to be interesting. I can never seem to get on board with the whole embedded post thing. Usually the wood is not treated to the correct rating for continuous ground contact due to purchasing ignorance. I sometimes make the treatment spec really noticeable so it gets purchased correctly but can't say if it is ever followed.

What is appropriate? Seems like a lot of inspectors are pretty clueless still and most decks get built by DIY'ers so appropriate is definitely a market thing right now from the owner/builder perspective. From an engineering perspective there is a building body of knowledge and evidence that it must be dealt with. Regardless of what you hear in the news, deck failures and injuries related are a serious problem. Most deck failures go unreported by the press unless they are spectacular. A lot of failures are either really incompetent installs or severe decay.

______________
MAP

 

[dhoward26]

focuseng: Thank you for the insight. I have that document in a folder and have used it for reference in the past.

That document refers to the IRC and has it's own conclusions as well. It says that the bracing requirements are required unless it is actually engineered. Where does one draw the line? How much difference in lateral stability is diagonal bracing going to create compared to perpendicular bracing when considering a deck that extends no more than 6 to 8ft from the house and is more than 20ft long? The proposed knee braces and connection seem kind of puny when considering the size of some decks.

I'm sure we will eventually have a better understanding in a few more years after more research. Seams like there are still a lot of unknowns and a lot of people designing in many different fashions. It somewhat appears that this is an issue that is client/contractor driven by aesthetics as well.

 

[RFreund]

I'll see if I can find some information from a presentation by Frank (mentioned above). Woodworks had a seminar that discussed this topic and the research mentioned above. See here: Advanced Design Topics in Wood Construction Engineering

Basically - you design the the deck as a horizontal shear wall (I also like the idea of combination cantilevered posts mentioned by Excel). The NDS or SDPWS gives values for allowable shear for horizontal and diagonal decking. Then you must have holddowns on each end that tie back into the structure as discussed by Excel above; due to the fact that the hangers are not rated for withdrawal loading.
The tests conducted by Virgina/Wash U were interesting in that diagonal or "V" diaphragm bracing below the deck did not help much as it was very stiff and created a 'brittle' failure. In regards to the horizontal diaphragm - The joist hangers contributed quite a bit of resistance to withdrawal/holddown forces when screws were used. Also the holddowns did not form a T/C couple rather each hold down was subject to tension due to a prying force created by the geometry of the holddown/joist connection. Ultimately neither the hold-downs nor the hangers failed rather large splitting of the joists occurred due to stress perpendicular to the grain where the deck was fastened to the joists (small T/C couples formed).

EIT

http://www.HowToEngineer.com

 

[IsaacStructural]

I'm not a fan of embedding the wooden posts in concrete. I guess it might depend on the climate, but up here you have two serious issues. First, it tends to lead to rotting as water collects at the base and doesn't escape, and it also serves as a place where water penetrates into the footing and freezes, damaging the post and footing. Instead I always use post bases, made by Simpson or other manufacturers.

In Chicago, I find the multi story decks are always tied into the main building structure and we rely on that for the later system. We do get some stability from having interior columns and stair stringers.

For short single story structures that are in the 3 to 6 feet off the ground category, I've used kickers pretty successfully. Architects don't mind in this case, since this is just a crawl space anyhow.

It also depends if the structure is clad or not. If it is an open structure there simply isn't too much area for wind to collect on.



M.S. Structural Engineering
Licensed Structural Engineer and Licensed Professional Engineer (Illinois)

 

[dhoward26]

RFreund:

Thank you for the information! I am going to see if I can get a copy of the seminar to watch.

What you spoke about with tension ties aligns with my thought that if you place tension ties to create a diaphragm length to width ratio between 2:1 and 3:1 then you have in effect created a "cantilevered diaphragm" or as you said a horizontal shear wall. It seams that in this instance, knee braces or kickers might not do much if the system is already tied into the structure for decks that are 6-8ft in width off the structure. The tension ties should limit the deflection of the diaphragm as well.

I added the following detail about a year ago to my work so that even if a contractor used nails for connecting the joists, I still had everything tied together at the exterior beam (although it does place the porch beam in cross grain bending). Now, that is just for the detail shown. It's different for joists floating over the beam.

 

[focuseng]

Interestingly, the tension tie design scenario applies if the cantilever diaphragm is assumed Rigid. What the testing showed is that the deck diaphragm is extremely flexible with perpendicular decking boards and the lateral tie-downs act counter-intuitive here because the pryout at the joist hanger is so extreme that in the typical install scenario, the "tension" side is in compression and the "compression" side is in tension due to local deformation at the joist sidesway. So in this perspective, knee braces do provide stiffeness/ bracing which may reduce the forces on the tiedowns. Generally the joist hanger can be expected to fail due to sidesway deformations using standard hanger nailing. The prescriptive tiedowns are an attempt to add another layer of redundancy to a crazy mess of fail possibilities known as "the wood deck".

In short:
1) decks without bracing of some kind will be quite flexible.
2) decks with high deformation are susceptible to second order failure modes so too much flexibility = bad (not to mention uncomfortable)
3) the tiedowns allow the deck to remain attached under high loads and distortion and the tension and compression side are not what you think under high deflections.
4) the lateral loading on deck due to occupancy is not well understood or codified so the tiedowns forces required are ???
5) decks are a frequent source of:
a) poor installation methodology
b) highly susceptible to corrosion of the fastening elements
c) highly susceptible to rotting due to poor moisture control techniques
*** all which make me want a redundant system

______________
MAP

 

[RFreund]

focus - good summary, thanks for clarifying.

Dhoward - it is a good seminar if ever in your area.

EIT

http://www.HowToEngineer.com

 

[dhoward26]

I can imagine it is a good seminar. The closest place to my area that it may come is Boise, ID...and I'm not too optimistic it will make it there.

I took my Timber Design class from Dr. Bender and it was the best class I took while in school. It's what really got me interested in wood structures. He is a great teacher and has a boat load of knowledge. Haven't had the opportunity to take any course work from Dr. Woeste though, but I recall Dr. Bender talking about him.

focuseng: thank you for the summary, things are making more and more sense to me and I'm adding lots of good ideas to my knowledge base. Thanks.