New Rooftop Solar Panels 1

[chamokinawan]

I wanted to check my thought process on adding solar panels. The code allows a 5% change in gravity loads to avoid retrofitting the structural elements. The solar panels are fairly light and if you really think about it, wherever you place the solar panels, you're removing the area for roof live loads. So is it safe to assume that you're actually reducing the gravity load, because you are removing your roof live loads (assuming the solar panels are less than your live load)? Thanks in advance.

 

[KootK]

Sounds reasonable if the space beneath the panels is not accessible, as I would expect. Do you have snow?

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[chamokinawan]

KootK: Thanks. Nope, no snow. And the solar panels are low so there is no room for crawl space. Thanks, again.

 

[jdgengineer]

I agree with the thought process in a non snow area but most jurisdictions in my area do not allow that argument. You also have to watch 10% seismic limit as well.

 

[RHTPE]

I have examined this for a conventional stick-framed gable roof in a 35 psf ground snow load. I would expect wind loads would also apply.

Panel installers (and the homeowner) want the fewest penetrations/attachments to the existing roof. If the rails that the panels attach to are connected to every 3rd roof rafter, what is the load path to where it no longer matters, and what is the tributary area for that load?

Both wind and snow loads on the panels are not permanent loads. But both loads can be applied to the panels and subsequently from the panels to the roof framing - the rafters supporting the panels must be checked. Could the panel installation actually increase the loading on select rafters? While the NDS might recognize load-sharing between multiple repetitive members, my observations of roof sheathing quality causes me to not use the permissible stress multiplier (for wood).

While I can agree that roof live load might never be realized after the panel installation, it will exist during panel installation or panel maintenance - should you really discount it?



Ralph
Structures Consulting
Northeast USA

 

[XR250]

Panel installers (and the homeowner) want the fewest penetrations/attachments to the existing roof. If the rails that the panels attach to are connected to every 3rd roof rafter, what is the load path to where it no longer matters, and what is the tributary area for that load?

I usually add strongbacks in the attic to distribute the load to all the rafters from the 4' O.C. point load.

 

[RHTPE]

XR250 - I have done the same. I have also required that the rail-to-roof connection be offset from one rail to another so that every rafter receives a portion of the load that the panels apply to the roof.


Ralph
Structures Consulting
Northeast USA

 

[WARose]

That's been my approach as well: as far as the gravity load goes, you are actually reducing it because there cannot be much roof live load where the panels are going. However, you will be adding to the seismic dead load. I looked at a commercial building the other day and he wanted to completely cover the roof in solar panels......and by my numbers that meant a complete run through on the lateral force resisting system. (And the icing on the cake: he didn't have any drawings for this building.)

 

[RHTPE]

WARose - I agree that there cannot be significant live load AFTER the panels are in place. And, for the most part, live load may not be a controlling load. But there WILL be a live load on the roof during installation and any future maintenance of either the panels or the roof itself.

I keep getting telemarketers calling about installing "Free" solar panels on my roof. I usually tell them that if they'll put a new roof on my house, then they can install whatever solar panels they want. But until I have a roofing system with a 20 year+ life span, no go on the solar panels.


Ralph
Structures Consulting
Northeast USA

 

[chamokinawan]

Thanks, everyone. I already checked seismic which was easy. The rafter tributary is interesting; I asked for cut sheets on the panels, so I haven't checked it yet. But it's a commercial building so I doubt the 2x4@24"o.c. will work.
@WARose: I have the same thing, but it's tilt-up so adding solar panels didn't make a huge dent on the seismic load.
@RHTPE: San Francisco has already mandated PV panels. Just a heads up as everyone in California will probably follow.

 

[BadgerPE]

I would concur that the loading of the roof (gravity only) is in fact reduced so long as the space below the panel is inaccessible. I recently attended a code change for the 2015 IBC and they discussed this specifically. See the third slide in the attachment.

 

[WARose]

While we are on this subject......let me ask everyone: as far as wind goes, at what point off of the top of the roof does the panel become an independent object to be loaded like a sign. Only a few inches off, and most people feel we are talking something loaded like C&C.

 

[MrHershey]

Check out 2015 IBC, they added language specifically to address this condition. Check out 1607.12.5.1

 

[Once20036]

I've heard that ASCE 7-16 is going to address wind loads on solar panels, but it's not due out until autumn.

 

[RHTPE]

This exchange is what I really enjoy about this forum. We all can't know everything about everything but each of us has a unique perspective on our craft, and the particular niche we evolved to (or fell into). The exchange of knowledge can't underestimated.


Ralph
Structures Consulting
Northeast USA

 

[EngDYIGuy]

I have a project where the PV panels are rated to be walked on. So I wouldn't throw-out the live load. Also, on Pre-Engineered Metal Buildings, even though the dead load from PV and the racking system is about 4 psf, it's still kicking off the IBC requirements. So be aware of this when dealing with PEMB's.

 

[sandman21]

SEAOC PV2-2012, has design methods for PV panels on roofs for both wind and solar.

 

[humanengr]

Did I miss something or was wind load not mentioned in the above thread?
I would have thought the panels may increase the lateral wind load (possibly downward and uplift as well).