Concerned about dead loads from the array

[Duxa]

Hi all, I know that the best thing to do is to hire a structural engineer to do this for me. The problem is that the few quotes I got quoted me at round $3000. I had calcs and stamp done through one of many online PV permit services. Obviously they didnt come out to my house and do calcs based on that, they looked at the roof on google maps, took info from me as far as trusses etc and then ran their calcs based on that. While this is good enough to get the permit, I dont 100% trust it. So Id like to do my own calculations so that I can sleep sound at night and not worry about my roof collapsing or flying away.

Hopefully it is one of those things that even if my calculations the error is by 30% or something its still good enough for the array. If its too close I guess Ill have to pony up half of my system's cost to have someone come out.

My roof has 2x6 rafters 24 inch on center with supports at every 3rd rafter (6ft). I am using Ironridge flashfoot 2 mounting system and Panasonic 330W panels. My roof is composite shingle at 19 degrees. Roof type Gable.

Panels are 18sq feet at 43lbs, I have 3 rows of 5 panels (15 panels total). So thats 270sq feet at 645lbs, which is 2.4 lbs per sq foot, plus the racking and we are at the around 3lb per foot. Which I believe is pretty standard for PV. My mounts are on 9 rafters, staggered (except the end points).

-0--0----0----0---0--0-
-0----0----0----0----0-

-0--0----0----0---0--0-
-0----0----0----0----0-

-0--0----0----0---0--0-
-0----0----0----0----0-

I am in Southern California, snow load is 0. 3s gusts of wind at 115mph.

Is that sufficient information to guesstimate how much load my roof can withstand? Even if with a huge margin of error and then take the lower end of that calc?

[bob-n]

Lots of good information here:

https://www.log-cabin-connection.com/roof-loads.html

[J.P.M.]

I am in Southern California,... Is that sufficient information to guesstimate how much load my roof can withstand?

Before I did that, I'd probably start with what the expected design wind load(s) in conjunction with other required external loading combinations might be before I did any stress analysis on a structure.

However, since you ask, as a retired P.E. who performed & signed off analyses and created designs with considerations for wind other external loadings combinations on structures, IMO, most of that is necessary but probably not sufficient information for a competent engineering analysis.

Since you have a permit, sounds like it comes down to a decision on your part as to how much uncertainty you want to tolerate.

Since, in a prior thread post you claim to be an engineer, I'd suggest you consult ASCE 7-16, "Minimum Design Loads and Associated Criteria for Buildings and Other Structures". There's a free PDF download if you snoop around some.

Another helpful publication that'll cost you 20 bucks to help satisfy your curiosity: From the Structural Engineering Association of California: PV2-2017, "SEAOC Wind Design for Solar Arrays". I've found that tome a pretty decent go by on using ASCE 7-16 for PV arrays.

Take what you want of the above. Scrap the rest.

[Duxa]

Before I did that, I'd probably start with what the expected design wind load(s) in conjunction with other required external loading combinations might be before I did any stress analysis on a structure.

However, since you ask, as a retired P.E. who performed & signed off analyses and created designs with considerations for wind other external loadings combinations on structures, IMO, most of that is necessary but probably not sufficient information for a competent engineering analysis.

Since you have a permit, sounds like it comes down to a decision on your part as to how much uncertainty you want to tolerate.

Since, in a prior thread post you claim to be an engineer, I'd suggest you consult ASCE 7-16, "Minimum Design Loads and Associated Criteria for Buildings and Other Structures". There's a free PDF download if you snoop around some.

Another helpful publication that'll cost you 20 bucks to help satisfy your curiosity: From the Structural Engineering Association of California: PV2-2017, "SEAOC Wind Design for Solar Arrays". I've found that tome a pretty decent go by on using ASCE 7-16 for PV arrays.

Take what you want of the above. Scrap the rest.

Thanks, Ill take a look. And as far as the wind loads, its at the bottom of the original post (115mph 3s gusts), unless you are referring to some other wind loads. But thats the measure the city specifies to be used in the calcs.

Overall I dont know that I necessarily need precise calculations to feel good about it. I just want to know in general, roofs with 2x6 trusses 24inch in centers and composite shingle, how much weight can they generally hold (assuming good condition of the roof). Ill be adding 3lb/sq ft. Is that approaching limits of a roof like that? Or is it not even close?

[sdold]

I think you'll find the uplift forces to be more significant than the dead loads. It's been a while but I think the hardest part of my install was figuring out the pull strength of the fasteners in the rafters.

[bob-n]

I wonder if Ironridge will do the engineering for you. It's worth a call. Their support team tries to help.

[J.P.M.]

I think you'll find the uplift forces to be more significant than the dead loads. It's been a while but I think the hardest part of my install was figuring out the pull strength of the fasteners in the rafters.

And one of the more uncertain.

See ANSI/AP&PA NDS-2017 et. seq. : National Design Specification for Wood Construction for some guidance on fasteners on wood.

FWIW, using the 2008 version of that spec for my design, I calc'ed the max. allowable withdrawal capacity of a 5/16" lag bolt in Douglas fir to be ~ +/- 270 lbf per in. of lag embedment using a wood sp. gravity of 0.60. For the design I used 5" lags, called it 4" of embedment/lag and then used a +/- 200 lbf total allowable for the max. pullout per lag in checking the vendor's design.
The calc'd/estimated wind only static withdrawal dead load pullout for my design using ASCE 7-16 methods for estimating wind loading for my site came to out to ~ 100 -110 lbf/bolt or so.
The allowable cyclic/vibrational loading for the joint as I did the calc was a lot less but the design had enough factor of safety to handle it.

[Ampster]

I think you'll find the uplift forces to be more significant than the dead loads. It's been a while but I think the hardest part of my install was figuring out the pull strength of the fasteners in the rafters.

Yes, I think that is the answer in a nutshell. A person walking in the roof will probably put more dead load on the roof than the panels if you space the connections appropriately. As an increase in margin of safety I ran the lags 1/2" deeper than suggested in the calcs. I have used both Iron Ridge and UniRac calculators.

[J.P.M.]

Yes, I think that is the answer in a nutshell. A person walking in the roof will probably put more dead load on the roof than the panels if you space the connections appropriately. As an increase in margin of safety I ran the lags 1/2" deeper than suggested in the calcs. I have used both Iron Ridge and UniRac calculators.

And for most arrays mounted on relatively small structures such as residences, what sdold writes will usually be the case.

However, I'd caution about implying any correspondence between a live point load and a distributed dead load, or a bunch of smaller dead point loads. The structural reactions will be entirely for each case and each needs to be considered on their own and in combinations as required by code.

You may think your comments are useful. I think not. I'd respectfully suggest you stick to what you know and stay away from making misleading, ignorant, simplistic and non helpful comments such as you bloviate about above. Add more to this thread and you'll do little more than further expose your ignorance to those who know something about the subject.

[Ampster]

And for most arrays mounted on relatively small structures such as residences, what sdold writes will usually be the case.

However, I'd caution about implying any correspondence between a live point load and a distributed dead load, or a bunch of smaller dead point loads. The structural reactions will be entirely for each case and each needs to be considered on their own and in combinations as required by code.

You may think your comments are useful. I think not. I'd respectfully suggest you stick to what you know and stay away from making misleading, ignorant, simplistic and non helpful comments such as you bloviate about above. Add more to this thread and you'll do little more than further expose your ignorance to those who know something about the subject.

No worries @J.P.M you probably did not understand that I was agreeing with you and @sdold. We all agree that the important issue is the uplift loads. If there was anything in my comments that was incorrect I would be happy to correct it.

[Ampster]

@Duxa, after looking at your diagram again, I thought I would mention another design consideration. If you are concerned about the number of roof penetrations, you can look at some options. Once you get on to the Iron Ridge or Unirac sites for the engineering you always have the option of using different size rails. Bigger rails can span longer distances and a different configuration may give you an acceptable trade off between heavier rails and less penetrations. The rail will be more expensive but the number of flashings could be reduced.

[Mike90250]

Roof penetrations contribute to wind resistance. If you want your array to stay put and not sail away, you need lots of bolts holding it on.
Or go for the longer spacing and have a tube of caulk on hand to fill in the few holes when it blows off.

[NorthRick]

Thanks, Ill take a look. And as far as the wind loads, its at the bottom of the original post (115mph 3s gusts), unless you are referring to some other wind loads. But thats the measure the city specifies to be used in the calcs.

Overall I dont know that I necessarily need precise calculations to feel good about it. I just want to know in general, roofs with 2x6 trusses 24inch in centers and composite shingle, how much weight can they generally hold (assuming good condition of the roof). Ill be adding 3lb/sq ft. Is that approaching limits of a roof like that? Or is it not even close?

Living in a city with a design snow load of 40 lbs/ft2, I myself wouldn't worry at all about adding 3 lb/ft2 dead load to a well constructed roof that you described. However, like the other posters I'd be worried about uplift from wind.

[Ampster]

Roof penetrations contribute to wind resistance. If you want your array to stay put and not sail away, you need lots of bolts holding it on.
Or go for the longer spacing and have a tube of caulk on hand to fill in the few holes when it blows off.

@Duxa will have to make that decision between under engineering or over engineering. I suspect the building code fits in there somewhere as well. Implicit in my suggestion was the use of Unirac or IronRidge engineering calculations and building department approval of those plans.

As long as we are talking about extremes, I will throw out some over engineering ideas. He could bolt the rails to every rafter, just to be safe. Bolting through the rafters would be even stronger than lags. He could also sister another rafter next to existing rafters to stiffen his roof even more. The list goes on.