Wood Deck Anchorage with High Snow Load

[eric294yz]

Hello all, looking for some input from my fellow engineers regarding wood deck anchorage in areas with high snow loads. My questions are arising as I have been asked by a client to look in detail to these requirements. Upon looking in detail, and applying the building code (as interpreted) by the letter of the law, I do not believe elevated wood framed decks in high snow load areas are "buildable" by standard practices that have been used by contractors and engineers in this area for decades!

I am an engineer who works in the Lake Tahoe, California area where we can have design ground snow loads above 400 psf. For elevated wood decks, above the "anticipated snow depth" per ASCE 7-10, I design my vertical deck framing members for the flat roof snow load + any drift loading requirements based on building geometry. The issue I'm now coming across is the design of the deck for lateral loads. CRC (California Residential Code = International Residential Code w/ Amendments) Section R5074.1 requires deck to be designed for vertical and lateral loads. There are "conventional" anchorage designs in this section for lateral loads; however, since we are located in area of ground snow greater than 50 psf, all residential design must be engineered and cannot follow conventional timber construction standards per the CRC. Therefore, I turn to ASCE 7-10 to generate by lateral design loads....

Per ACSE 7-10, 20% of the flat roof design snow load shall be considered in lateral seismic calculations. If I use ASCE 7-10 Sec. 13.3 with Table 13.5-1 using "Other Flexible Components of High Deformability" with Rp=3.5, I can generate very large seismic loads on the deck on the order of 50 psf or more.

Now lets take a 10'x20' deck for example, the total lateral load to design for is (50psf)(10')(20')=10,000#. Say the 20' length is the side adjacent the building, therefore the linear shear demand (assuming cantilevered design) is 10,000#/20' = 500 plf. Per the 2012 NDS SDPWS, horizontally sheathed 2x decking has a capacity of only 50 plf. Therefore, we need to provide diagonal strapping to create a "diaphragm" to transfer load back to the primary building structure. Our total diagonal strap load will be 1.41*10,000=14,100#. Using Simpson CMST16 strap (capacity = 4,545#), we would need four (4) straps in each direction, eight (8) total....Then comes the next problem, how do you anchor each one of these straps back into the building to transfer that load to the primary building frame????

I've attached a couple details I worked up (that I think are insane and would quickly put any rational engineer out of business in a hurry) but I would like to get other thoughts on how you would engineer an elevated wood framed deck, anchored to wood framed structure with wood shear walls.... Thoughts?

 

[BridgeSmith]

Wouldn't you also multiply by the seismic acceleration coefficient?

I would doubt that in most cases 400psf snow load is even possible on a raised deck, since it would be over 20 ft of the wettest snow. Is the snow load on decks actually the same as the ground snow load?

 

[eric294yz]

HotRod10, I'm using the "Flat Roof" snow load and not the "ground snow" load for the deck design. My post states I'm using ASCE 7-10 Sec. 13.3 to generate my seismic load which is the "seismic coefficient." FYI, wet saturated snow has a max density of about 30pcf, and yes I have personally seen 12-15 feet of snow accumulated on a deck.

 

[BridgeSmith]

Pardon my unfamiliarity with the ASCE. We use the AASHTO specs, which utilize location-specific ground acceleration coefficient spectra, which accounts for the seismic hazard and the period of the structure. According to the USGS Seismic Hazard app the peak ground acceleration is fairly high there in Tahoe, but the acceleration coefficient for long-period structures (those that are fairly flexible, such as a wooden deck) is less than 0.4. In other words, if the structure is flexible enough so that it doesn't move as far as the ground moves in an earthquake, the lateral forces are reduced.

I assumed the 20% value was accounting for the snow not being an internally rigid mass, and able to shear internally under the inertial forces.

 

[NorthCivil]

Not too familiar with American code, but have done similar designs up north.

Don't like your detail at all. You are right, where do those forces go once you have concentrated them like that? Better to keep the forces dispersed rather than concentrated. A good philosophy to keep in mind with wood structures is that if you can, its better to use ten thousand nails than a single big bolt.

From the numbers you are discussing it seems like you are mostly on track, save for your deck diaghram sheathing capacity. I can't comment in too much depth, but from experience using the canadian wood standards, assuming the right type of nails, nailing patterns, and diaghram sheathing (plywood), shear values of 500 lb/ft are easily acheivable.