Roof Ballast

[SteelPE]

I have a potential client who needs to have some roof work done (Current = IBC 2015, Vult = 130 mph) on a building that was designed and built in 1986/1987. I am involved with infilling some existing roof exhaust fans, verifying additional capacities in the roofing system for increased roof insulation and replacement of the existing RTU's. Part of the project scope consists of removal of existing roof/roof ballast and replacement with a new adhered roofing system (something that I wasn't asked to address specifically).

I have some reservations in regards to the removal of the roof ballast as it is going to end up creating an uplift situation on the roof joists. At a bare minimum I would like to install joist uplift bridging at the BC panel points. I suspect I am going to face significant pushback from the client in regards to any roof modifications due to what the client perceives as a decrease in the loads on the roof.

Has anyone else faced a similar situation in the past?

I have been to plenty of projects in the past where no uplift bridging was installed after removal of stone ballast.

 

[GC_Hopi]

Dont envy your position since you will have to be the bad guy. Hopefully the record drawings are in good condition and you can get field photos. Like you said you will have to check the roof system for the new loads. I am assuming you have open web joists so I think the minimum is to have BC bridging on the first panel point then after that the extent is based on the numbers. SJI started to produce better information regarding uplift design.

 

[phamENG]

Is there an architect on the project? If so, is this getting permitted as an alteration? Reroofing falls under level 1. And level 1 is silent on what you have to do if net uplift increases. If anything that puts you in a worse position - because the right thing to do is to reinforce the roof for the new net loading but now you can't tell them that you must do it.

At the very least I'd insist on bracing to ensure stability of the roof system. I'd also compare the new loading to contemporary capacity tables and see if you're at least close.

 

[SteelPE]

Yes, open-web steel joists on the roof.

I have worked at this facility a few different times. I have some of the structural drawings but not all of them. They have been enough for me to pop in a few equipment cupola's in the past... but I have never measured the existing joist sizes. I have an idea where the bridging is, but I wil

The good news is, the owner is not an idiot, so hopefully he will buy what I am selling without much pushback.

 

[KootK]

At a bare minimum I would like to install joist uplift bridging at the BC panel points.

1) Obviously sharpen your pencil enough to prove to yourself that the uplift is real but, otherwise, this seems like a prudent approach.

2) As you know, the bridging does not need to be at all of the BC panel points. You can probably make some pretty good hay with BC bracing at only the ends and the middle.

3) A joist system that was properly designed to begin with should already have lateral BC restraint near the ends whether or not the original condition produced uplift. This much is needed simply to laterally brace the joist compression webs. If this is true of your project, then perhaps you're only adding a central bridging line for the most part.

 

[KootK]

By the book, I suppose that one also has to consider shear reversal in the webs. I've not found this to be a big deal in most uplift cases however. You can pretty much tell by inspection if you've got gravity tension webs that will be noodley under compression.

 

[bones206]

I had the same issue on a re-roof job for a laboratory. Gravel ballast was being removed and the roof joists did not have any uplift bridging, but inside work was strictly off the table.

I ended up specifying a concrete decking product to partially replace the ballast weight:

https://www.usg.com/content/usgcom/...r-panels/structo-crete-structural-panels.html

The concrete decking was installed on top of the decking before the insulation and EPDM roofing was installed.

 

[RWW0002]

I have encountered ballasted roofs where the ballast is being removed as part of re-roof or has been removed sometime in the past. I have recommended the addition of bottom chord bridging on several of these based on the existing bridging and net uplift magnitude. At a minimum I would recommend adding bridging the the first bottom chord panel point if there is any net uplift at all per SJI specs section 5.11 (assuming K joists)

3)A joist system that was properly designed to begin with should already have lateral BC restraint near the ends whether or not the original condition produced uplift. This much is needed simply to laterally brace the joist compression webs. If this is true of your project, then perhaps you're only adding a central bridging line for the most part.

I have run across several ballasted roofs with no bridging at the first panel point (or anywhere near the end). Is there a specific requirement in old specs to bridge this panel point if there is not net uplift? If so, it was not followed too closely in my experience back in the days where these roof systems were common. When ballast is removed I generally recommend adding this bottom chord restraint at the end at a bare minimum, and then additional rows as analysis dictates..

 

[KootK]

Is there a specific requirement in old specs to bridge this panel point if there is not net uplift?

Unless one is willing to count on the self weight of the bottom chord to do the stabilizing as we sometimes do with big architectural trusses.

The YMCA that I frequent is fairly modern and has a run of joists where there's only one line of bridging, placed dead center. The joists are very short but, still... equilibrium and such. Buckle and spin. Drives me nuts. I have to keep my eyes down while I'm getting swole lest my indignation break my intense focus.

 

[RWW0002]

I am not overly familiar with the specification either, since # rows of bridging is generally left up to the joist supplier. I have seen several ballasted roofs or floor joists with only one row of bridging down the middle, and in glancing briefly at SJI specs more than 1 may not be required for short span joists with gravity load only.. I have never had to dig into this more deeply and maybe I should self-educate when I have some spare time.

I have several buildings/conditions on my "Don't Look Up" list, but joist and bridging hasn't made the list yet. Maybe it should..

 

[KootK]

Non uplift bottom chord bucking is definitely one of those "paper" failures that never seems to actually come to pass. Hybrid timber truss designs often omit this bracing for aesthetic reasons. It seems to work fine and I don't get the impression that anybody is checking the fancy resistance mechanisms.

 

[bones206]

maybe I should self-educate when I have some spare time.

I believe SJI has published/will be publishing Tech Digest 2 on the subject of bridging.