Portrait vs Landscape layout...does it matter?

[HX_Guy]

I'm getting different things told to me about this, what is your guys' input?

I have one company that says the only right way to do install panels on the roof is in portrait mode and it should be avoid "at all costs" (as he put it) to put the panels in landscape as "that is not good for the roof and it's really not the correct way to do it".

I brought this up with another installer, who suggested landscape because we could fit more panels if needed, and he said it doesn't matter one bit aside from installing in landscape tends to be about 20% more track, therefore more expensive for the installer.

What's your guy's take? Does it matter at all? I don't see why it would.

[russ]

No difference - the installer telling you one way is bad has shown himself to be a fool!

[Mb190e]

I wasn't necessarily told it was wrong but i was told, if you put them in portrait layout it spreads the weight of the panels out over more of the roof rafters. And if you get snow the panel will start working quicker in landscape layout. I was told that it starts working after the first third of the panel is uncovered in landscape and portrait the whole panel has to be uncovered.

[russ]

Quote from your post, "I have one company that says the only right way to do install panels on the roof is in portrait mode and it should be avoid "at all costs" (as he put it) to put the panels in landscape as "that is not good for the roof and it's really not the correct way to do it"."

The entire panel has to be clear of snow - yes

[HX_Guy]

I spoke with this company again and here is their claim...

"Installing panels in landscape mode means you have to run the rails parallel to the trusses which is not good for the roof.
The correct way to do it is run the rails perpendicular to the trusses which means the panels have to be mounted portrait."

I asked why can't you mount the rails perpendicular to the trusses and also mount the panels in landscape and they said that voids the warranty from the manufacturer (SunPower in this case) as the rails have to be mounted parallel with the short side (perpendicular to the panel).

Any truth to this or are these guys full of crap?

[Volusiano]

I spoke with this company again and here is their claim...

"Installing panels in landscape mode means you have to run the rails parallel to the trusses which is not good for the roof.
The correct way to do it is run the rails perpendicular to the trusses which means the panels have to be mounted portrait."

I asked why can't you mount the rails perpendicular to the trusses and also mount the panels in landscape and they said that voids the warranty from the manufacturer (SunPower in this case) as the rails have to be mounted parallel with the short side (perpendicular to the panel).

Any truth to this or are these guys full of crap?

Actually what they said makes a lot of sense to me.

The first part that makes sense is that if you have the rail perpendicular to the trusses, you can control the exact spacing of the rails from each other. If you run the rails parallel to the trusses, your rails must be right on the trusses or else there's no mounting point, and this doesn't allow flexibility on where the spacing between the rails can be. The only way to solve this is to have 2 layers of rails. The bottom layer of rails is perpendicular to the trusses, and the top layer of rail can sit on the bottom rails and be parallel to the trusses and still be placed for correct spacing. This will of course cost more money, not to mention having the panels placed a bit higher than normal. And this is assuming that somebody offers a rail layer on top of another rail layer solution.

The bit about needing to have the rails parallel to the short side of the panel may make sense if the panel is designed to be clamped on the long side only and the short side is not designed with clamping in mind.

If either side can be clamped just the same as long as 4 clamping points are required, then it would take a lot more rails if they run parallel to the long side -> more $.

Overall, I can see a lot of good logics in favor of mounting the panels in portrait mode.

[J.P.M.]

I spoke with this company again and here is their claim...

"Installing panels in landscape mode means you have to run the rails parallel to the trusses which is not good for the roof.
The correct way to do it is run the rails perpendicular to the trusses which means the panels have to be mounted portrait."

I asked why can't you mount the rails perpendicular to the trusses and also mount the panels in landscape and they said that voids the warranty from the manufacturer (SunPower in this case) as the rails have to be mounted parallel with the short side (perpendicular to the panel).

Any truth to this or are these guys full of crap?

Assuming for the moment a rectangular layout, and at least 4 attachment points between each panel and the rail system, with one panel "touching" two rails, I think the company is mistaken due to what seems to be their resulting implication that the rails must be perpendicular to the short side of the panel. Snow loads will be the same no matter the orientation between panel and rails. For most applications were the plane of the array is parallel to the roof, wind loads will probably be similar or very close in either orientation, maybe even a bit less for landscape vs. portrait if the array is not parallel to the roof, tilted or sawtooth. I agree it is usually and probably almost always far better design, depending on the layout and mounting surface to have the rails normal to the trusses in a roof system to spread the load among as many trusses as possible and not just a few. This however, has little to do with the orientation of the instrument of the induced loads, i.e., the panels.

My observation is that most arrays use portrait for the panels but the overall array is usually landscape. Looking at the roof loading, it may or may not be necessary to adjust the array layout with respect to orientation or shape as conditions warrant and/or as the designer decides, but I doubt this has little to do with the orientation of the rails with respect to the individual panel orientation - long or short side. There MAY be some advantage to being able to control the max. dimension between clips in portrait, but in the end that has the tradeoff of possible increase in individual panel bending moment and "using" fewer trusses for the same # of panels. There ain't no free lunch and each design is different.

I designed my array before I got any bids, including structural w/wind & seismic per UBC/ASCE etc. The chosen vendor got very close to my racking layout without knowing it. My array is 4 X 4 landscape with 8 rails perpendicular to the trusses, each panel supported by 2 rails.

I looked at the racking websites this A.M. Maybe I missed something but I can' find any reference to landscape vs. portrait with respect to panel orientation in any of their design info or literature.

[russ]

Totally agreed - Too many sales people have no idea at all but love to chatter.

Assuming for the moment a rectangular layout, and at least 4 attachment points between each panel and the rail system, with one panel "touching" two rails, I think the company is mistaken due to what seems to be their resulting implication that the rails must be perpendicular to the short side of the panel. Snow loads will be the same no matter the orientation between panel and rails. For most applications were the plane of the array is parallel to the roof, wind loads will probably be similar or very close in either orientation, maybe even a bit less for landscape vs. portrait if the array is not parallel to the roof, tilted or sawtooth. I agree it is usually and probably almost always far better design, depending on the layout and mounting surface to have the rails normal to the trusses in a roof system to spread the load among as many trusses as possible and not just a few. This however, has little to do with the orientation of the instrument of the induced loads, i.e., the panels.

My observation is that most arrays use portrait for the panels but the overall array is usually landscape. Looking at the roof loading, it may or may not be necessary to adjust the array layout with respect to orientation or shape as conditions warrant and/or as the designer decides, but I doubt this has little to do with the orientation of the rails with respect to the individual panel orientation - long or short side. There MAY be some advantage to being able to control the max. dimension between clips in portrait, but in the end that has the tradeoff of possible increase in individual panel bending moment and "using" fewer trusses for the same # of panels. There ain't no free lunch and each design is different.

I designed my array before I got any bids, including structural w/wind & seismic per UBC/ASCE etc. The chosen vendor got very close to my racking layout without knowing it. My array is 4 X 4 landscape with 8 rails perpendicular to the trusses, each panel supported by 2 rails.

I looked at the racking websites this A.M. Maybe I missed something but I can' find any reference to landscape vs. portrait with respect to panel orientation in any of their design info or literature.

[Volusiano]

I think the bottom line is that in most cases, the overriding reason to do portrait is that it'll take less railing in portrait mode and therefore lower cost. That's why portrait layout is more popular than landscape layout. And this is dictated by the direction of the trusses. Had the trussed been going horizontal on the roof for some reason, I would bet that the landscape layout would become more popular. It's all about the $.

[J.P.M.]

I think the bottom line is that in most cases, the overriding reason to do portrait is that it'll take less railing in portrait mode and therefore lower cost. That's why portrait layout is more popular than landscape layout. And this is dictated by the direction of the trusses. Had the trussed been going horizontal on the roof for some reason, I would bet that the landscape layout would become more popular. It's all about the $.

I'd agree that a lot, if not most of the reason for portrait layout is rail length, thus cost. I'm not sure however, that I've ever seen a horizontal rafter (truss member) on a pitched roof - purlins maybe. I suppose it could be done, but would look rather funky and I bet take more material and labor.

[skanmyth]

I see this house on link most of the panels installed landscape and rails are attached to long side of the panel http://solarpanelinstallations.ca/po...prettyPhoto/0/
I hope I am not in trouble adding this link Moderator - please remove the link if I not suppose to add any solar sellers link I looked up google satellite view its not updated yet with resent view.

I am working on my roof lay out 27 of may panels out of 40 going to be landscape I am doing it my self so i can take the 20 or 30% whatever extra cost for rail to add extra 10 panels by doing landscape orientation.

[Mike90250]

It depends on what the PV panel Mfg recommends in the installation instructions.

[Raul]

I really can't believe what I'm reading . I agree with post 6 . OP the company that was telling you the landscape to be avoided if possible was right. But it can be done.
If you have a row of landscape panels going up the rails will be attached to the same 2 rafters in paralel . All the brackets on one rail will hit the same rafter . That is a lot of weight and stress from one row of panels on 2 rafters . What is required in this situation is to double up the loaded rafters in the attic so it eliminates the flex . Same goes for prefabricated trusses. There is no such thing horizontal trusses .
When your panels are in portrait you don't use the same rafter for the brackets , the top rail can be alternate / staggered ; if that's not posible for eny reason then the next row can , so you don't load the same rafter all in line up. Think of it you got a decking with the joists exposed , wich way are you going to sheet it up ? Paralel or across? Every survey I came across the structural engineer allowed for loaded rafters reinforcements on landscape and staggered loaded points on portrait.
Same goes for clamping on the short side , unless the frame is strong enough and the manufacturer specifies in they spec that short sides clamping is allowed. Otherwise you will leave the panel unsupported on its length and will flex more than permitted tolerances , hence snail trail and cell failure in the short future. Take a Lg 285 mono and compare it to a trina 240 silver C Chanel 40mm framed . The lg will flex like paper while the trina is solid. Not all panel frames are the same , to keep cost down some frames are less beefy than it used to.

[radareclipse]

Very nice skanmyth. I wouldn't want to get on tbat top roof level. Did you photograph that with a drone?

[bcroe]

Landscape Panels

My comments. A landscape panel in theory can start producing when 1/3 of the snow
slides off, which generally will uncover the first section (running the long way) with the
other 2/3 bypassed. However the available voltage will be only 1/3, so this will be
usable only if the inverter system can operate in that mode.

Second, I'm leaving a big gap between panels in the downhill direction, so the snow
only has to clear the short way across one panel instead of several. That never gets
mentioned, but will get built here eventually. I suppose that is only practical with a
ground mount.

This costs more? Yea, so does having panels covered with snow. Bruce Roe