Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations LittleInch on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Upgrading Existing Building Joists?

ChiEngr

Structural
Joined
Oct 19, 2021
Messages
78
Location
US
Hello,

I am working with an existing building comprised of metal decking steel roof joists and joist girders. On one side of the joist girder line, the owner wants to add a couple of rootop mechanical units. The original design code was BOCA. The original design load (D+S) was 40 psf. Under current design code for the village, the snow load is 10 psf larger. I have the following questions:

1. Do I have to check all the joist girders and joists for this increased load, even at areas away from the RTU's?
2. For the joist girders supporting the joists directly supporting the RTUs, do I have to use the current snow load provisions even at the joist spans not supporting the RTUs?

The IEBC is not very clear on this circumstance, so any help or guidance would be greatly appreciated. Thank you!
 
1) I would say no, though the building official would have the final say. I think you would be wise, however, to inform the owner that loads have changed, why they've changed, and what impact it could have on the structure. If the BO isn't going to require it, let the owner make the decision not to upgrade.

2) Yes. And also the joists impacted by the snow drift against the units.
 
2. For the joist girders supporting the joists directly supporting the RTUs, do I have to use the current snow load provisions even at the joist spans not supporting the RTUs?

If I understand the question accurately, I feel that the logical path for this is simply consistency.

Determine which roof areas you will apply the additional snow load too and then carry only those additional snow load through the rest of the load path. Certainly, I would not be applying the extra load to only the joists on one side of the girder but then designing the girders for that same load having been applied to the joists on both sides of the girder. I would consider such an approach to be both irrational and needlessly punitive.
 
The IEBC is not very clear on this circumstance, so any help or guidance would be greatly appreciated.

Unless someone ponies up with a silver bullet code clause to resolve the ambiguity, my approach would be:

1) Tell the owner there is a fork in the road here. You can go with a very expensive solution that will for sure receive AHJ approval. Or, you can go with a much cheaper solution that you feel is entirely reasonable and has maybe a 1:5 chance of getting torpedoed by the AHJ. Their choice as to how to manage that risk. I'll assume that they choose the low cost option.

2) Propose that the only new loads to be considered be:

a) The RTUs themselves obviously.

b) The drift loads that will be caused by the installation of the RTUs. I would seek to not even apply the extra snow load to the portions of the directly affected joists outside of the drift zones. Honestly, I might also propose that the drift be calculated based on new code but the ground snow load that goes into that drift calculation remain the original snow load.

The basic philosophy would be to only use higher loads where the installation of the RTUs is actually the cause of the higher load. Where the cause of the higher load is simply improvement or change in the meteorological record, I would try to dodge that.

Obviously, this is a pretty aggressive strategy. That said, I feel that the root aim of all of the retrofit codes is to encourage engineers to to responsible things for thoughtful reasons. And I feel that my approach would be exactly that.

The mechanical upgrades on this building will surely increase the utility and usable life span of the existing structure. And that is good, both for society and for the environment. You'll be doing the Lord's work, so to speak by facilitating the mechanical upgrade. And I've seen cases where the AHJ is receptive to such arguments. An example is using external post-tensioning to upgrade the capacity of concrete floor systems. Often, it is cost prohibitive to fully fire rate the PT in such situations and the AHJ will let that slide because the upgrade to the structure is considered to be a net benefit to society even with the fire rating compromised.

Another way to approach the situation is to ask the question: where the additional load is simply of meteorological thing, why do areas of your building not directly impacted by the presence of the RTUs have to be upgraded when the roof of a building down the street does not?

As I understand it, the two fundamental things that trigger an upgrade are:

1) Alterations that would significantly increase the expected load, even without the meteorological business.

2) Situations where the occupancy of the building changes and society deems that a wholesale upgrading will be a reasonable cost relative to the other costs associated with the occupancy / ownership change.

Judged by that litmus test, I feel that my approach qualifies as doing a reasonable thing for thoughtful reasons.
 
Dear lord.

I suspect most rational engineers would apply this analysis for the increased snow loads to the bays and the affected girders. And not cheat the analysis by using the "old" (known to now be incorrect and unconservative, i.e. unsafe) ground snow load. Remember you are on the hook for a complete load path AND life-safety per a design standard and code. The remainder of the structure, hypothetically, could be reinforced or analyzed and proven acceptable for the increased load with the next re-roof, particularly if you put that on a drawing... and a letter. You should also look at roof slope and susceptible bay....

If this is old enough it's a cantilever system and that's a much woolier creature and analysis would extend beyond just the bay. Effects (even without cantilever framing) could go as far as reinforcing footings and columns, but one would hope not. And if it's old enough you've got the whole tdes or not to tdes for the TS/HSS columns circa 1990 one would tend to expect in a cantilever roof system. You said BOCA, so I'm tempted to think it's a 1980s structure. When exactly the "thickness cheating" on Tube Steel and Hollow Structural Systems began, but if it's in the same decade you might want to check it for tdes, and hope it works that way.

When the loads increase AND the structure is being modified, don't play.

When the structure is NOT being modified, it would be allowed to remain "as-is". You might have some ethical/liability involved if there is a collapse and you don't inform them of the changed snow load affecting the whole structure.

I'd probably check it without the drifts to see what's needed at a minimum, and if the drifts require too much reinforcing, you can elevate the RTU, that's in O'Rourke's book on Snow loads from ASCE. Drift loads aren't all that common except on the roof mounted chillers, most of the normal HVAC is smaller (in plan) and it's not required.

open web steel joists are finicky creatures.... proceed with the standard of care.
 
Under the prescriptive method, the IEBC states for Existing Structural Elements Carrying Gravity Load (502.4):

Any existing gravity load-carrying structural element for which an addition and its related alterations cause an increase in design dead, live or snow load, including snow drift effects, of more than 5 percent shall be replaced or altered as needed to carry the gravity loads required by the International Building Code for new structures.
The Work Area Method has substantially similar language as well.

You would need to evaluate the joists and joist girders at the area of the RTU (and any RTU caused snow drift) to see if you exceed that 5% allowed increase. Presuming you do, you need to design the elements with the increased loads for new snow load as required by the IBC, plus the new RTU load. You may have a situation where the joists require reinforcing but the joist girders do not, and often columns won't need reinforcing. You only need to address the areas affected by the change, so it's pretty straightforward to state that the areas that are not loaded by the RTU are not affected by the new RTU load, so would not be included in the alteration.

It's worth noting that in the 2015 IEBC the text of the code read

Any existing gravity load-carrying structural element for which an addition and its related alterations cause an increase in design gravity load of more than 5 percent shall be strengthened, supplemented, replaced or otherwise altered as needed to carry the increased gravity load required by the International Building Code for new structures.
My office has always interpreted this to mean that you can look at the total load in combination (DL + Original Design SL typically) before and compare to the load after (DL + Original Design SL + New RTU) to determine the 5% limit. The new language appears to indicate that the comparison should be DL to DL + RTU only, not including the snow load, but I'm not entirely sure if that was the actual intent of the change.
 
I agree in general, Bey, but the IEBC would more rationally be applied to load combinations, and dead load alone is a load "combination" therefore it always read that way. You wouldn't just check for "total load". You check for changed dead loads and you check for the other combinations, if they are all below 5% further analysis is not required.

2015 language isn't current, by the way.

1753204074199.png
2024.

Unless we're not comparing the same code sections.
 

Part and Inventory Search

Sponsor

Back
Top