Residential Deck Engineering 3

[medeek]

I have found plenty of design guides for decks specifically aimed at contractors and builders but not a whole lot for engineers. Given the recent deck failures I've seen in the news I am specifically wondering about lateral loads due to occupants moving about on a deck and how to best address that.

I have a free standing deck with a large hot tub. The deck is rather high (9' from grade) so as to bring it in level with another two story deck positioned on the residence. The deck supports a large 3,000 lbs hot tub (water + tub) on six 6' diameter poles. The owner purposely designed the deck to be independent from the house so in the case of a seismic event it could break away and not take down the house if it were to fail catastrophically. Apparently this was a similar approach she had from a previous residence in where a structural engineer had suggested this method given the high seismicity of the area.

The posts on the concrete bases are essentially pin jointed and with a couple of 5/8" bolts into the beams above, not offering any significant lateral or moment resistance. I will design some knee bracing that will provide the required lateral strength, in both directions.

I am not too worried about the typical lateral wind and seismic loads or even the vertical loads. What I am wondering about is how best to quantify a bunch of rowdy teenagers running and jumping about on this deck, I am thinking about 10-12 people could theoretically be on this deck surrounding the hot tub at any given time. Does anyone have any past experience with a a similar situation or references to papers on this subject.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[Michel60]

Take a look at the AWC publication "Wood Design Focus". There was a series of articles by a University research group specifically dealing with deck design, anchorage, and rhythmic excitation. I believe its was in the last 12-18 months if I recall correctly.

 

[medeek]

On a related note, does anyone know why the AWC DCA6 in Figure 10 prohibits the use of knee bracing on center posts?

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[bowlingdanish]

Can you just specify some steel cross bracing rods? I'd make sure you can take 5% of your total ULS vertical load if it was applied laterally.

 

[medeek]

Here we go:

http://www.deckmagazine.com/Images/MeasuringLateralLoads_tcm122-2103359.pdf

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

These two papers also cover wind and seismic quite thoroughly:

http://www.deckmagazine.com/images/windloadsanddecks_tcm122-2103128.pdf

http://www.deckmagazine.com/Images/SeismicLoadDeterminationforResidentialDecks_tcm122-2140592.pdf

However, I am noticing that all of these decks are attached to the primary structure and DCA6 states that decks should be attached to the primary structure. What is the thinking on decks that are free standing?

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[boo1]

Two X braces under the joists?

 

[KootK]

Yeah... decks. None of them are really up to snuff laterally in the Malone & Rice sense. Some thoughts:

1) When decks have diaphragms with some stiffness, I suspect that they mostly cantilever laterally from the house proper and the knee braces at the front don't do much.

2) When deck diaphragms are soft, the knee braces at the front become much more important. I believe that deck failures occur most often in this scenario due to P-delta effects. These stability issues are exacerbated by tall posts, effectively pinned bases, and the flexibility inherent in the knee brace connections.

3) I don't see a deck delivering enough load to a house proper in a seismic event to materially affect whether or not the house collapses. For the reasons listed above, however, I think that a typical deck is much more likely to collapse if not attached to the house proper as the deck then lacks that one stiff shear resisting support provided by the house. Is the concept to accept that the deck will collapse in a seismic event and design it to a lower standard that would be objectionable for the design of the house proper?

I'm fine with designing decks as free standing so long as consideration is given to stability and connection flexibility. I feel that it is inherently riskier than attaching to the house however.

My wife built us a little bench for the end of our bed. See the photo below. The diaphragm was much stiffer than a free standing deck diaphragm and the moment connections at the top of the legs was probably about as stiff, relatively speaking. Yesterday I sat on it to adjust one of my shoelaces. The bench P-Delta'd over slow enough for me to perceive it but too fast for me to react and prevent my other foot from getting squished.

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.

 

[Lomarandil]

Well, we typically design working platforms etc., for 50 lbf horizontal per construction worker (ASCE 37, among others). That might be a start.

 

[XR250]

I always use full length wood X-bracing on decks. The numbers on kneebraces never work out.
Here's something that will throw a wrench into your works - None of the Simpson or USP bases are rated for any shear load except the ones that are cast into the footing (the ones the contractors hate)
So, technically, none of our lateral bracing works anyway. (I generally ignore the lack of shear testing on the bases as they seem to work fine)

 

[medeek]

I sure they have some rating. Its probably that adequate testing has not be done so there is an absence of a rating which does't mean an absence of capacity.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

Most decks I have looked at so far I ususally go with a 40 psf live load and a 10 psf dead load. To be conservative I think I will add this distributed load to the 3000 lbs of the hot tub. Also this problem gets more complicated when one has to consider how the concentrated load of the hot tub is going to get distributed to the deck joists and then into the two beams and six posts supporting the deck.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[jayrod12]

The hot tub will likely load the deck joists as a reasonably uniform load. I tend to provide double or triple joists under the two edges running parallel to the joists anyways.

 

[Archie264]

>>>I sure they have some rating. Its probably that adequate testing has not be done so there is an absence of a rating which does't mean an absence of capacity.<<<

+1 to that. I consider it atypical of Simpson's otherwise excellent product information and customer service. If the bases really had no capacity then it would be rare that they could be used. While we model statically determinate structures with roller connections we don't usually use them to model 3-D space frames; pinned connections are more the norm. In any case, if they they truly had no lateral capacity then we certainly couldn't justify using them on a free-standing deck, at least, not theoretically.

 

[Archie264]

Medeek,

I don't know why AWC prohibits use knee bracing on the center posts. The only guess I can come up with is that since the center posts carry twice the vertical load of the corner posts perhaps they want to make sure no undue bending moment is applied to them as well.

A related question I have is if anyone knows why they prohibit through-bolting lumber to the sides of posts? Are they trying to keep the beam plies adjacent to each other for better lateral bracing?

 

[medeek]

Archie264, explain the prohibition on bolting through the posts, I've never heard of this before. I'm actually thinking about lag bolting or through bolting two 1/2" Dia bolts into the knee braces (or X-braces) and then into the posts, probably with some 2x6 or 2x8 pressure treated lumber. I may have to up the braces to 4x6 if the compression numbers don't work out. I think this would be the simplest solution.

The steel rod bracing does sound attractive but I have no idea how to implement that solution or details showing its construction.

However, before I get that far I am still working on my spreadsheet which examines the lateral loads due to wind, seismic and occupancy.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

Kootk I agree with you that the deck should be attached to the primary structure, that is how I would construct it. The client though has other ideas and is quite stubborn in the matter, so I think I can make it work but the bracing is probably going to have to rather substantial in both directions. That last thing I want is a call back letting me know the deck has racked and is now 6' lower than originally designed and cousin Joe was underneath it at the time.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[dhengr]

Medeek:
At the very least, you (someone) would/might cut or drill half the post way, right at a max. load and moment point on the post, in an effort to attach two knee braces, without a second thought. Some builders claim to be good enough with a chain saw, that they work fine for them for this kind of joinery. If you could use some hardware which did less damage to the post and you designed for all of the various loads and moments, I’ll bet there isn’t any code prohibition against knee braces on the center post. Remember, that “AWC DCA6 in Figure 10 “ is written for builders, DIY’ers. and homeowners who aren’t qualified engineers and usually don’t know how to follow the IRC, or understand it, let alone understanding or following the IBC. If the deck builder follows DCA6 to the letter, they will usually stay out of serious trouble, and probably won’t need an engineer’s stamp when they go into the AHJ for a permit.

As for various post bases by Simpson or USP, they just don’t know how some duffus will install them, so they are much safer saying nothing. The bolt or bolts are usually cantilevered up a few inches out of an unknown blob of concrete, called a pier or footing, with unknown embedment, it is just too difficult to pin down a lateral cap’y. I suspect that in an engineering discussion both of the manufacturers would admit that there is some lateral cap’y. in many of those bases, when installed properly. But, then they would immediately launch into a discussion on their liability when they start putting numbers in a catalog.

Free standing decks are dangerous because they don’t have the considerable lateral stability of the house to help support them laterally. When properly attached to the house, and properly tied back into the floor diaphragm in the house, decks of a reasonable proportion work pretty well. When not properly attached they can peel/unzip, right off the bldg. and maybe pull the rim joist/board off with them. The free standing deck, and particularly one with 3000-4000lbs. of sloshing water up at 8-10' above the foundation, and in a seismic region, needs to provide all its own lateral stability within its legs and foundations. Most homeowners don’t like to see knee braces or x-bracing on their decks, they like the cleaner lines, and think nothing of lateral stability until the deck folds over and is on the ground.

 

[medeek]

Thank-you for your input Dhengr, especially the last paragraph. After some thought about the self standing deck I called the client earlier this afternoon and did explain the problems with bracing such a beast and there probably being the need to install significant unsightly bracing to the underside of the deck, basically aesthetics were out the window.

Does anyone have any bracing ideas or details that would work well with 6' round peeler poles. I'm pretty familiar with knee braces using conventional methods with square/rectangualr timbers, lag bolts and/or simpson brackets (KBS1Z). Perhaps there are better methods to apply bracing to round poles that I am yet unaware of.

I'm looking at the Simpson CJT right now (CJT3 w/ 4x8 knee brace), see sloped view:

http://www.strongtie.com/ftp/catalogs/c-c-2015/c-c-2015-p229.pdf

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

After reading some more online and finding some interesting (horror) bracing stories on various construction and DIY sites I'm firmly convinced that knee bracing will not be sufficient for this design. Without the bracing from the primary structure side sway is going to be a major issue, with or without the hot tub.

Full on lumber x-bracing or steel rod bracing is the direction I'm taking now.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com