Residential Deck Loading - Snow + Live? 2

[MCurry]

I've been having an ongoing disagreement with a fellow engineer. Allowable stress load combinations in the IBC (require Dead + Live + Snow be considered. My associate says that combination does not occur, so does not design for it. I say he is violating Code.
Do you typically combine snow and live loads when designing exterior residential decks?

Thanks

 

[JAE]

Well, not doing too many decks, but the same issue arises with the top level of parking garages.

Yes, per the code - the D+L+S combinations REQUIRE the addition of snow plus live loads....they DO occur simultaneously...its just that the snow gets crunched down when you walk or drive over it.

You just don't add Roof Live Load (Lr) and Snow (S).

 

[RockEngineer]

I do decks in some high snow load areas. Even though the code states do both, I don't when there are very high snow loads. If you have 40 - 240 psf snow loads you aren't going to get enough people on the deck to add another 40 psf. It isn't reasonable. Especially if you also engineer for sliding snow impact from the roof for decks along eves. To get that many people up there you have to shovel off the snow. That's where engineering judgement comes in. If you have snow loads below 40 psf then maybe you could design for both. It's still not very likely. My 2 cents worth. [reading}

 

[JAE]

Well, keep in mind that the IBC load combinations (for LRFD) use the following combos:

1.6S + 0.5L
1.6L + 0.5S
0.5L + 0.2S

For ASD they still have the straight up L+S in the combinations so the LRFD appears to be more logical (and consistent with RockEngineer) than the ASD combos.

 

[IceNine]

ASCE 7-05 has D+.75L+.75S for ASD

 

[MCurry]

Thanks JAE and RockEngineer,

I agree with both of you. We're in a 30 psf snow load area (Central Virginia). So while it may be unlikely to see both full snow and live loads, I'll design for both until the building code is changed.
I like RockEngineer's cut-off including live load above a certain snow load.

In fact, I'll send off an inquiry to the International Code Council for some feedback and let you all know what I find out.

 

[RockEngineer]

Since IBC 2006 defers to ASCE 7-05 for snow loading it appears that they have taken a little common sense into the latest code with D + 0.75L + 0.75S rather than 100% of both. Good find IceNine. Star for you.

 

[RARWOOD]

In Minnesota we are now designing roofs out state for 42 psf snow, where in the past they were designed for 30.

Now does IBC want us to design for 42 psf snow and another 20 psf live load for the guy on his roof shoveling off the snow.

This last winter the heaviest snow fall at my house was 18".
I weighed a 12' x 12' x 18" section. The total weight of the snow was about 15 psf. After weighing the snow I didn't worry about shoveling off my deck or roof. That was a good thing because I know my roof wasn't design for snow + live load.

 

[civilperson]

Yes, when the freak snow fall occurs, (and it will), my defense that I followed the prescribed code load combinations will make your defense that the loads do not co-exist seem weak! Assuming that both structures failed.

 

[MCurry]

Hello Eng-Tips engineers. As promised, I contacted the International Code Council for an interpretation of deck loading requirements. I'm pasting the email stream below (listed in reverse chronological order.


Here's my response:

Hello,

Thanks to all.

Regarding answer A1 – Table 1607.1 item 6 provides that deck live loads should be the same as the occupancy served – in the case of a residence, that would be a 40 psf live load. The load combinations to be considered according to 1605.3.1 don’t differentiate deck live loads from any other live load, so it’s a reasonable interpretation (in my view) to consider that the equation 16-9 (D+L+S) be applied, using the 25% load reduction per 1605.1.1.

The presence of 1607.11.2.2 (Special Purpose Roofs) suggest that if snow loads for the deck are developed using the code roof snow load approach, the total loading or load combinations would have to be directed or approved by the code official. I can tell you from personal experience, code officials in smaller jurisdictions are generally note willing or not able to provide “direction” and are hesitant to provide approval of anything not spelled out in the code.

The remaining option (per the code) for deck loads and load combinations would then be to apply the ground snow load to the deck, and per 16-9 to combine full D+L+S loads.

Given that deck collapses are a relatively frequent (and very public) type of structural failure, it seems appropriate to me that the code prescribed loading be quite clear (and reality-based of course), and subject to as little interpretation as possible. I encourage the ICC to prioritize this issue.

[red]By the way, I floated this idea at a structural engineering on-line discussion group (Eng-Tips.com) and found that engineers take very different approaches to this situation. I will share this correspondence with this group of practicing engineers, who may have their own input and/or suggestions.[/red]

Thanks again for your response, and all that you do to help us provide safe and serviceable structures.


--------------------------------------------------------------------------------
Here's the Response from ICC:

Sir,

Your paraphrased question and our reply are as follows:

Q1. The 2003 IBC requires that the load combination Dead + Live + Snow be considered [e.g. Equation 16-9]. Some engineers neglect this combination for exterior decks, saying it is not reasonable to combine snow and live load on a deck, particularly in areas with heavy snow loads. Is this combination required for exterior decks?

A1. The combination is required, but the more appropriate question might be whether there are any special considerations when applying this load combination to the design of an exterior deck. The code offers no specific guidance on applying the load combinations to decks. This means that where a design proposes any relaxation of this load combination [other than the 0.25 reduction for multiple variable loads permitted in Section 1605.3.1.1], it should only be done with the approval of the Building Official in accordance with Section 104.11. In considering an alternative to the load combination required by the code the following items could be considered.

First of all note that Section 1608.1 requires that the design roof load shall not be less than that determined by Section 1607. Typically this is referring to the roof live load of Section 1607.11. That is also the reflected by Equations 16-9 and 16-10 which indicate the use of rain load, R, or snow load, S, or roof live load, Lr, to determine the maximum load effects for the design of structural members. Again these provisions are written for the typical situation where it is the roof that is designed for the snow load, but they consistently state that it is the roof live load OR the snow load that applies rather than applying them concurrently.

Now consider special purpose roofs in Section 1607.11.2.2 which must be designed for the appropriate live load. The statement in Section 1608.1 does not distinguish between special purpose roofs and any other roof. So the literal application of the provision to the special purpose roof is no different. That is that the roof design considers the snow load but the design roof load is not less than the live load determined by Section 1607. There is no indication here that the live load applicable to the special purpose roof and the snow load should be applied concurrently. There could be some parallels drawn between treatment of a deck live load and the applicable special purpose roof live load.

Q2. Would the answer be any different for residential uses?

A2. No.

Code opinions issued by ICC staff are based on published codes and do not include local, state or federal codes, policies or amendments. This opinion is based on the information which you have provided. We have made no independent effort to verify the accuracy of this information nor have we conducted a review beyond the scope of your question. As this opinion is only advisory, the final decision is the responsibility of the designated authority charged with the administration and enforcement of this code.

Alan Carr, S.E.
Code and Standards
International Code Council
Washington Field Office
2122 112th Ave NE
Suite C
Bellevue, WA 98004

--------------------------------------------------------------------------------

Here's my initial email

Sent: Thursday, May 03, 2007 1:21 PM
To: Alan Carr
Subject: Load Combination Question

Hello Alan,

I’ve been having an ongoing debate with some fellow structural engineers regarding exterior deck loading.

The Code (IBC 2003) requires that the load combination Dead + Live + Snow be considered. Has there been any discussion in committees that you know of about the ‘reasonableness’ of this combination for exterior decks, particularly for residential use? I’ve talked with several engineers who ignore this combination for exterior decks, saying its not reasonable, particularly in areas with heavy snow loads.

I say we (engineers) can’t just ignore the load combination, but also question how reasonable the requirement is.

It would seem appropriate to have some cut off for requiring live plus snow load combinations when ground snow loads exceed some amount.

Thanks for your response, or for forwarding this to the appropriate person for a response.

 

[JAE]

My first inclination here is to question why we are even talking about ROOF loads? Yes, snow USUALLY falls on roofs, but it's still just snow and decks can occur either aligned with roofs, below roofs, or near the ground surface and the appropriate snow reduction factor should be used in any case.

But to include "roof" live loads in this discussion I think is off the mark. Decks receive loads from humans and other furniture as floors do - and do not represent the sort of live loads that are prescribed for roofs.

That is why I think that the L+S combination should be used, as opposed to the Lr vs. S which are never combined.

 

[MCurry]

JAE,

I the ROOF load the ICC folks are pointing to is under "Special Purpose Roofs" 1607.11.2.2. That section requires appropriate live loads be considered on a roof used for assembly, roof gardens, etc.

I guess those loads would be considered ROOF live loads, by the code (though I would consider them as floor live loads myself).

You said you use the "appropriate snow reduction factor" when designing decks? I only see that reduction used for Roof snow loads - in fact, I don't see that the Code specifically addresses snow loads on anything BUT roofs.

That being said, I've designing decks for snow loads, reduced as if a flat roof (just as you've done) and treat the 40psf live load as a roof live load, allowing the 25% reduction when considering D+L+S. It seems reasonable to me, but techncally, I may be in violation of the Code (unless I get approval from the building official per 1607.11.2.2).

I think the code needs to be more clear.

 

[JAE]

MCurry - I guess what I meant about "appropriate snow reduction factors" was that they depended on the height of the deck relative to the building. The reduction factor for snow is based upon the idea that generally wind blows a lot of snow off the roof if it is more exposed, high in the air, etc.

I would look at the deck and see whether it is up high where this might occur, or down low where it wouldn't and also include drifting on it as appropriate.

 

[NHstructural]

Are you aware of section 1605.3.1.1 Load Reduction in IBC 2003? It states:
it is permitted to multiply the combined effect of two or more variable loads by 0.75 and add to the effect of dead load. The combined load used in design shall not be less than the sum of the effects of dead load and any one of the variable loads.

So, if your snow load were high enough to exceed 0.75(L + S) then it seems just D+S would be appropriate. If the snow load were low, say 20psf and your LL is 40psf, then the design load of variable loads is 45 psf.

The variable load reduction seems similar to the ASCE 7 comment made by icenine.

 

[JAE]

I'm not sure snow is considered a "variable" load. We've discussed this topic here in the past.
I don't think we ever came to a conclusion on it.

 

[miecz]

Seeing that 1605.3.1.1 has the term "two or more variable loads", live load would have to be a variable load. If not, there could never be more than two variable loads.

 

[RockEngineer]

I always considered anything that wasn't there 100% of the time a variable load. So if it wasn't the material weights or machinery, it was variable. Just my two cents.

 

[HeberPE]

My understanding of the ICC answer posted above (answer A1) is that a "deck" can be considered to be a "special purpose roof" with a different live load Lr than a typical roof. This is equivalent to a special roof that is also a garden, assembly area, pedestrain walk-way, etc. Therefore, one only needs to design to the higher value of S or special Lr (floor load.) Quotes from ICC - "it is the roof live load OR the snow load that applies rather than applying them concurrently." and "There is no indication here that the live load applicable to the special purpose roof and the snow load should be applied concurrently." Yes, a deck is a special purpose roof, with a live load equal to the floor load.

This is the common practice in my area where the base snow load is 60 psf, and goes up from there. I have never seen calculations that use both S and L added together for design of roofs or decks.

The design snow load is based on the 2% probability of occuring in any given year. To add 40 psf live load to that value and assume that the deck will be fully loaded with people at the same time as the 50-year snowfall has occurred is quite unreasonable. I side firmly with MCurry's associate.

 

[MCurry]

HeberPE's approach makes absolute sense to me. I'm converted. We're in a relatively light snow load area (ground snow load varies from 25 to 40 psf, depending on the muncipality). No engineer I've spoken with in the area has ever referenced 1607.11.2.2 in the code, or considered a deck a 'special purpose roof'. But after this discussion, it seems clear that this is the correct code (and practical) approach.
It wouldn't hurt if the code explicitly stated to consider decks as special purpose roofs, with the live load considered as a Roof live load for load combination purposes.

 

[miecz]

I've had two very different experiences with "decks". One, the deck that I designed for my home. It had only occasional live load which was never present when it snowed. However, it was subject to serious snow drift from the roof. Our old building code (BOCA) specifically required snow drift to be considered on decks (good thing!). I see that our new code (IBC) does not. It requires snow drifts for roofs (1608.7), but decks aren't mentioned. Am I missing something here? Only one previous post mentionsed drifting. Does this imply that IBC considers a deck a roof, or that decks aren't subject to drifting?

The second experience had to do with a precast deck for a porch of a high rise residential building. The deck had failed and fallen on the deck below, which thankfully held. When inspecting 100 or so of the decks in this building, it became apparent that many of these decks were carrying permanent live load, i.e., refrigerators, freezers...This load was there when it snowed.

Based on the second experience, I'd have to say that a deck is not a roof.