IEBC Allowable 5% Gravity Increase on Total Load or Load Type? 2

[waytsh]

My understanding based on previous editions of the IEBC is that the 5% increase was based on the total gravity load. The wording in the more current versions, however, could be read as 5% of the load type, i.e. dead, live, or snow.

“Any existing gravity load-carrying structural element for which an alteration causes an increase in design dead, live or snow load, including snow drift effects, of more than 5 percent shall be replaced or altered…”

How do you guys read this?

 

[Nick6781]

Note that it's the effect of the load. Think in axial, shear, moment, and overturning, etc.

 

[lexpatrie]

I've always viewed the "origin" as being intended for a lightweight roof overlay. That would lend to the "solo" 5% increase that would net, in some load combinations, 15%. I would NOT read it that way without signoff from the building official and disclosure on the drawings/letter that that's what is being done. (a la Florida 'violating' the code or standards).

Were it my project, logically, the 5% overstress would apply to a load case or a load combination under the appropriate code, and I suspect there've been a few code change proposals (that I also presume failed) that would have clarified that.

Further, the 'overlay' in my mind, would typically be shingles, not all that typical with the "current" vogue for flat roofs. This would technically be a dead load, but unless there's some kind of offsetting effect, a change of occupancy (i.e. live load increase), adding weight (like solar), and changing insulation (higher snow load), these all "eat" from the same trough of allowable 5% increase.

I'd also add something about roof slope. We've seen enough roof collapses in the media, better to try to avoid them on the design/repair side, if possible.

Mike Mike - code change proposals are "free". No need to go the drastic step of joining the organization... I've gotten one approved. That or just call that Reid Middleton guy who writes a hundred code change proposals a week, he can squeeze yours in, I'm sure.

EZ building - I've referenced that provision in several damage reports calling for engineering to satisfy those requirements. One was a residential, the other was a poured gypsum roof from 1970. It shows up in Hurricane damage projects with structural and roof damage, there's a dollar limit in the FBC/FEBC for residential and a precedence lists, and I often see public adjuster estimates that include renailing of the diaphragm (for uplift), as a precautionary measure. T.L. Smith discussed that some years ago. And What's SID, anyway?

 

[RWW0002]

I've always viewed the "origin" as being intended for a lightweight roof overlay. That would lend to the "solo" 5% increase that would net, in some load combinations, 15%.

I am not following the math here. Can you give me an example of how 5% increases in loads can lead to a load combination of anything over 5%??

My opinion is that the letter of the "new" wording is more restrictive and more conservative than previously. It seems that you are implying that it is less conservative than previous codes from your statement.

I agree with you though that the 5% has always, in my mind, applied to the total gravity load effect in the past.

 

[lexpatrie]

5% increase in dead load (hypothetically insulation and a 1 psf EPDM, say)
5% increase in snow load (due to R value crossing from Ct = 1.0 to Ct = 1.1, say)

D+S load case 10% increase?

Or a snow or rain load increase due to the revision in the definition of susceptible bay?

You're perhaps thinking of the D + 0.75 (W+S or W+Lr) cases? Or did the load combinations change while I wasn't paying attention again.

 

[Mike Mike]

Lex, in your case D+S would increase 5%, not 10%
1.05D + 1.05S = 1.05(D + S)

Thanks for the tips, I am way too busy with work to get into the committees right now, but I'm writing all this down on my to do list for the next housing bubble / virus leak. I assume all Reid Middleton's stuff goes thru a committee of experienced structural engineers, right? do you know if those committee positions are paid? do the majority of changes come from outside ICC staff?

 

[lexpatrie]

Uh, Reid Middleton is a company but the guy who submits a ton of code change proposals (or did) was like one guy "representing self" as I recall, and I've forgotten the name, it got a write error and has Phil Brazil in that spot now in my storage.

Ok, maybe I got that partially correct, except the proposals (of late) are for WABO.

Redirect Notice

I meant a 5% increase of dead load, and a 5% increase of snow load, 1.05D/D + 1.05S/S = 1.05 + 1.05 = 1.10.

One code change proposal isn't too bad, there's a lot of overhead in setting up an account and formatting things to submit them, and it's not intended for "gee this provision is unclear please clarify this" kind of commentary. What you do in that case is provide a language change and then all the more involved people can step in and public comment on a "better" version and then they can all fight it out. You don't have to be all that involved and the proposals come in three year spikes anyway. They aren't doing the 18 month revisions anymore. If you want to delay it, I'd write down some notes before sticking it at the bottom of the "to do" pile so it isn't so "what the heck was I thinking" when you pick it up in two or three years.

Influencing I-Codes.... Lorenz, Interface Magazine, June 2021.

As far as the "latest" anything on the I-codes, it's surprisingly hard to find, but the votes on the 2026 CAH from say, March, is here. (Almost useless without the proposals themselves as they just list D for denied, AS for As submitted, etc. There will be another hearing in October....

This might be more useful, but still lacks the proposal text. But it is almost transparent about what was done and why. Feels pretty disorienting)

2025 Group B ROCAH to CAH #1 Now Available​

I think this is the actual proposals (and I also think it's for the right year...).

https://www.iccsafe.org/wp-content/uploads/2025-Group-B-IBC-Structural.pdf

S172 is my personal favorite this time around.

S157 gets the dumbo award for being the most dangerously stupid idea. Went down, HAPPILY, in FLAMES 14-0 disapproving.

 

[waytsh]

Thanks for all the input, and the references to the commentary, which I don't currently have. It seems like the commentary is still written using the language from older versions of the code. Did they simply forget to update it?

Another thing I find frustrating is the inconsistency within Section 806.2, where the first half of the paragraph discusses loading, and the second half shifts to capacity. Like others have mentioned, why not just reference capacity throughout? For example, if I’m reviewing a building that was originally at 80% capacity and the new load takes it to 90%, the actual increase in capacity utilization is 10%. But if the added load represents a greater than 5% increase in the load itself, we’re forced to comply with the new code? That feels unnecessarily rigid in situations where we know the capacity.


Mike Mike, I fully support your involvement in the code process. Your input could bring some much-needed clarity.

 

[phamENG]

we’re forced to comply with the new code?

If you know the capacity, and the new demand is under that capacity, then you're good.

The 5% (whatever that 5% actually refers to - clearly it's, well....unclear) is a threshold for doing anything to it. If you don't exceed it, then you're good. If you do exceed it, then you need to go deeper. First stop is determining the capacity of the system as is and comparing it to the new load. If it works, you don't have to do anything. If it doesn't, then you reinforce, replace, etc.

So if you increase by 4% and it's now 103%, you're good - a part of that built in factor of safety is whether or not it survives construction, and it clearly did that, so we give it a pass. If you increase by 8% and now it's at 94%, you're still good since your demand is less than capacity. If it increases by 7% and now you're at 106%, then you need to reinforce it because we can't reliably depend on it to support design loads.

 

[Mike Mike]

Thanks lex, read thru all your stuff and it looks like I would need to set myself a reminder to submit my proposal between October 15, 2027 and January 10th, 2028. Then there would be a bunch of rounds of stuff that I would probably need to partake in, culminating in maybe 10 hours of my time? I would rather get paid for my time, but I guess I'll give this a shot, just this once.

Your formula to determine the combined increase resulting from individual increases is incorrect. It should be as written in the screenshot below. Not sure we're on the same page here.

waytsh, if you were at 90% utilization under previous codes, I would think you would be under 100% under current codes. Codes don't generally change that much.

 

[lexpatrie]

But if the added load represents a greater than 5% increase in the load itself, we’re forced to comply with the new code? That feels unnecessarily rigid in situations where we know the capacity.

I think that's thematically correct but potentially overkill. If it's over the 5% you're required to analyze it, isn't that the wording? That isn't exactly binding it to the new code, in terms of language?

The one issue I could see is there have been a LOT of changes in punching shear in concrete, as well as the pre 1971 moment loading on slabs which wasn't correct in the statics, or perhaps changes in shear stirrups in beams, things like that, where the elements aren't exactly isotropic. Or open web steel joists where they don't have uniform strength properties along the length, unlike, say, a steel beam. That's actually come up, I'll see if I can find the article.

The Basics of the 5% Rule, Wagner, Structure Magazine, March 2021.

Where the code isn't clear, you want to meet the Standard of Care, that wooly little sprite.

Mike, that's literally not the formula for the load combination. Sorry. It's maybe the right result, but it's not the right formula.

 

[phamENG]

Mike, that's literally not the formula for the load combination. Sorry. It's maybe the right result, but it's not the right formula.

Right. D+S is. At least, for that one load combination. So if you're comparing loading before and after the alteration, and both dead load and snow load increased by 5%, the new loading could be written as 1.05Do+1.05So, where Do and So are your original dead and snow loads.

Say the dead load was 100plf, and the snow load was 50plf. Originally, you had 150plf. Now, you have 1.05*100plf+1.05*50plf=157.5plf. 157.5plf/150plf = 1.05. The total load went up by 5%. Not 10%. Mike isn't writing the load combination, he's writing the algebraic expression comparing the new load to the original load to determine the percentage increase.

The more I look at this, the more I think the new wording makes a little more sense. Before, it was a question of a 5% stress increase. To prove that, you not only have to determine the loading before and after, but you have to analyze it, too. So a real stickler of a reviewer could demand to see calculations proving that the stress increase was only 4.2% (or whatever). Now, it's a simple loading comparison. Dead load was 100plf, now it's 102plf...good to go. It actually makes it a lot easier to comply with it. Unless, of course, you're in an area where effective snow load has gone crazy. But, then, you'd probably have to do it anyway because the stress will have changed.

 

[lexpatrie]

As a side note, it looks like the actual text hasn't changed between 2018 and 2024. Here's the 2024 for reference, I notice the bit about current IBC in there now, I was working off brute recall.


Source: 2018-2024 IEBC

What I'm getting at is the basic question was about what the 5% applied to, one would imagine it's load combinations, not outright loads themselves, meaning.... a 5% increase in dead load alone is "okay", because D is a load combination, [ hypothetically a roof designed for 3 psf original dead load under this exception could be allowed to take another 3 psf for a 100% increase, but let's skip that for now].

But just a 5% increase in snow or live loads would be okay because it's not a solo load combination, so D + 1.05L or D + 1.05 S, is, hypothetically, less than 5% as a resultant, (provided the dead load doesn't increase).

The situation I was considering as a counter point, where Mike Mike and I got hung up, would be a recover that changed say, the thermal coefficient resulting in an increased snow load (not the most likely scenarios, I'll grant, but where I live SLM - Snow loads matter).

Anyway, say DL = 60 psf and SL = 20 psf. (I suppose that's a Risk Category II structure where the minimum snow load governs). Anyway, 3 psf additional roofing for a recover (say 1 psf membrane and 2 psf insulation), that's 63 psf [ it LOOKS like exception 2 applies here but it really doesn't, due to the additional insulation which isn't "roof covering"]. [Brings me back to the position that the 3 psf is intended for steep slope shingles...... as no membrane itself is 3 psf that springs to mind. Not that shingle manufacturers allow this, but again, let's skip that.

If the snow load coefficient increases to 1.05 (which it really CAN'T, 1.1 to 1.2 or 1.0 to 1.1 are actual coefficients but let's skip that, too), if snow load goes from 20 psf to 21 psf, that's a 5% increase.

63/60 + 21/20 = 1.05 + 1.05 = 2 ( 1.05) = 2*2 * (1.05/2) = 1.10. Doesn't it? That's what I was getting at.

The original load combo (D+S) is 80 psf. The final load combo (D+S) is 84 psf. Which is a five percent total increase.....

What was the question again?