Allow Absurd Load Combination to Control? 7

[SandwichEngine]

Ran across this yesterday. I've simplified the numbers on this example to make the question cleaner.

Existing building is a 25,000 square foot low rise manufacturing facility in Texas. Ground snow is 5 psf.

End user wants to repurpose the building which would require hanging additional piping from the roof that weighs 5 psf over what the building was designed for.

Checking the existing building shows that it's slightly overstressed in some areas but only for the load combination DL+0.45W+0.75Lr, IBC Equation 16-13.

This equation checks to see how the roof members would fare if there was 75% of the roof live load on a completed roof during a 10 year wind event. For this building that's 500,000 375,000 lbs of men or (2500) (1875) 200 lb men on the roof during a 10 year MRI wind event. This amount of men of a roof would never be more than about 20 or 30 during construction, not to mention years later.

The worst case unity check for any other load combination is 0.50.

I have no problem allowing the new larger loading from a safety perspective but how do you navigate that it would require you to disregard this portion of the code.

This has me thinking I should call BS on the code when it doesn't make sense, like you would with software.

Haven't decided what I'll do yet but would like to hear other perspectives, especially ones that disagree with my initial take.

 

[MotorCity]

Maybe that many men on the roof at the same time does not make sense from an overall, global building perspective. However, the load combinations are also used for the design of local members, say a single beam for example. For the sake of argument say you have a 20' beam with a 5' tributary width. Thats 100 sq ft * 20 psf roof live load = 2000 lb/ 200 lb per man = 10 men standing (or equivalent equipment) standing over that beam.....thats very possible.

 

[SandwichEngine]

MotorCity,

That's a better way of approaching the problem than looking at it globally.

I still think it's extremely unlikely for any men to be on a roof in anything close to a 10 year MRI wind event. Maybe though, when they leave the roof to to increasing winds, they leave some equipment and material all concentrated on one roof member. Still just seems like an extremely remote possibility to me though.

I guess I'm thinking that if it were my building and I was given the choice to pay 10's of thousands of dollars to reinforce or gamble that the above scenario will never happen, I'd choose to gamble.

Now I'm talking myself into the position I didn't want to--that it's much more of a liability question than a straight-forward engineering question. Even if you get the owner to sign off on taking that gamble, if it somehow ever does happen, they're still going to sue. In fact, even if a tornado hits the building, there's a chance that they'd try to use that in a lawsuit. Jurors aren't going to understand this discussion nearly as well as this sub. They're just going to be hearing "non code-compliant" over and over.

Time to talk to the client about figuring out what the new loading really is.

 

[haynewp]

Look at the IEBC and the 5% additional gravity demand provisions. I wouldn’t push it beyond what the IEBC permits or you are on your own to argue your points in the case it fails.

 

[rb1957]

we'll sometimes get to equally ridiculous numbers ... pressure * area. A colleague invented the unit "one elephant" to try and get the point across ...
we have 50,000 lbs on this panel ... that 25 2000 lb elephants ...

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[ChorasDen]

Think the thread title is a bit disingenuous. Is DL+0.45WL+0.75Lr truly an absurd combo? What happens when you need to perform work on the roof, the morning is clear blue skies, but by 2:30 in the afternoon, that design wind event appears from nowhere, and everyone is trying to get off the roof by the only open exit stairwell at the center of the structure. Now you have all your men congregating around the doorway as they are trying to exit the roof, while the wind pressures are also developing.

How about you assign some criteria: owner replaces the members at and around egress routes, and then the rest of the roof is taped/lined/painted to not allow workers to occupy that painted space unless they are also carrying meteorological weather data indicating a stable wind pattern at the roof elevation?

 

[Brad805]

I am not sure all are familiar with a pre-eng roof. I for one would not be standing on a cold rolled girt spanning 25'-0" with a bunch of buddies, nor would I be standing on the tin cladding spanning 5'-0" or whatever it might be. That said, prudence is warranted.

 

[SandwichEngine]

ChorasDen,

The title is not disingenuous. The load combination is totally absurd when looking at the building globally. But I figured I might be missing something which is why I created the post.

The first reply pointed out what you are pointing out--that looking at individual members is where this load combo could become a more realistic possibility. Your adding egress adds a different dimension to the discussion that I also hadn't considered. Thanks for bringing that up.

 

[Tomfh]

I don’t think the title is correct either. You can argue that it’s unrealistic, but it’s not “absurd”. All you seem to be saying is the live load is higher than you think will ever occur. That is true for most all buildings.

 

[chris3eb]

Is it a flat roof? If so, it could also be a couple inches of water and debris on the roof during a big storm.

 

[phamENG]

Live load - and loading in general - is far from arbitrary. Or absurd.

Whether you're going to try to change them or not, but especially if you are, you need to understand how that load was decided. And why.

Take a look...

 

[ThomasH]

Hi

The loading is basically selfweight combined with a reduced wind load and a reduced live load. It may be unrealistic for global design of a large roof. But I would not ignore it for a single roof truss or sometning similair. But what part of the structure is designed for the load from 25 000 square foot? Where is the total load relevant to compare with?

What surprises me is that you make a load increase of 5 psf and have this problem. That means that the utilization was high to beging with. What type of building is this?

You say that it is "slightly overstressed"? What does that mean? What parts are involved and how much are the overstressed? If the loading should actually occur, what type of failure will be the result?

As for "gambling" with the safety. I can see the building owners incentive but I can't quite see yours.

Thomas

 

[Brad805]

Thomas, I believe he is talking about a roof like the image below. With a the loads he is talking I expect the roof girt spacing is quite large. Pre-eng companies do not waste pounds of steel. The utilization he speaks of is very typical.

 

[ThomasH]

Brad805,
Thank you for the info.
That was not the type of roof I would have guessed. But I have seen similar structures and I think I have also designed someting like that, but that was a while ago .

I agree that steel manufacturers usually don't "waste" steel. But since "additional piping" was mentioned I also thought about a possible industrial client. And those clients often want some margin in the design for future load upgrades. But that should of course be discussed during the design process.

But can one frame can support 25 000 sq foot?

Thomas

 

[WinelandV]

Quick question: does the roof pass DL + 0.45W + 0.75Lr if you don't include the 5psf from the new piping?

Please note that is a "v" (as in Violin) not a "y".

 

[Brad805]

Thomas, SandwichEngine works for a pre-engineered manufacturer. In my experience those buying a pre-engineered bldg do not always think about future allowances. Low bidder gets the work, and that is usually less steel. Given the building is being re-purposed, it could be a new owner. We have worked for roofing companies trying to re-roof poorly performing roofs with a torch on roof (5psf added +-) and it is a problem. I expect this roof would have frames at 20'-30' o/c.

 

[le99]

Checking the existing building shows that it's slightly overstressed in some areas...

Slightly overstressed, by how much, 5%, 10%? It is not wise to disregard code requirements, but write-off by "engineering judgment" was a frequently used strategy. I've seen the write-offs up to 15% with ASD as the design method. You should clearly state your "justification" in your report though.

 

[ThomasH]

Brad805,

Thank you for your reply. I didn't know what SandwichEngine worked with exactly. The result in itself is not surprising because a straightforward method to save money is by saving material and why build "better" than required? Saving material is usually part of what we do even if, sometimes, the material cost is not the most important parameter. I assume that the original design met the requirements at the time. And now there is a problem due to a small load increase.

But that does not necessarily make the governing load combination wrong. I think that it would be interesting to know how overstressed the roof is and what the failure mode is.

Thomas

 

[Settingsun]

I guess I'm thinking that if it were my building and I was given the choice to pay 10's of thousands of dollars to reinforce or gamble that the above scenario will never happen, I'd choose to gamble.

That's part of the reason codes exist.

 

[Scipio_Wright]

Have you checked 1607.12 in the IBC? The section is on roof loads, and includes reductions in roof live loads depending on the layout of the roof. That might get you low enough to not disregard that load combination you see as "absurd".