air flow under solar panel; how much is enogh

[n8huntsman]

I have a clay tile spanish roof. Many have suggested the best way to attach the panels is to remove the tile, lay 90 pound roll roofing, place the panels on that, then put the tiles back on around the array. It would be similar to this:http://www.hardysolar.com/solar-san-...nish-tile.html
The main difference in my application is that I would have 7 or 8 courses of tile below the panels, not just two as in the provided link. The roofer says you can't nail through the roll roofing to add the tiles back on, so the tiles have to be glued on and he does not suggest gluing 7 courses of tile. He said if it were only two courses, he could do that. He suggested a custom flashing, from the new roof material, up on to the top of the first course of tile. This would cut my airflow under the panel, especially if I only used an L-foot. Air wouldnt be allowed to flow up the valley of the tile and underneath the panel because it would hit the flashing. Would it be worth the approxamately $800 to add standoffs which should get the array up above the tile? Or should I try to find tall L-feet, or just not worry about it?

[inetdog]

Would it be worth the approxamately $800 to add standoffs which should get the array up above the tile? Or should I try to find tall L-feet, or just not worry about it?

If you have a warm summer climate, then you need to provide at least four inches of clearance between the panels and the material under them. This will allow decent air circulation, since this is a sloped roof. Going down as low as one inch will not give you good air circulation and will cost you maybe 10% or more in panel output on hot days. Going above 4" to maybe 8" will be better, but the return in production per inch added goes down as you get past 4" to the point where you have to look at the additional cost.

[n8huntsman]

With the L-foot and ironridge XRL rail, it would be 4.146" off of the deck. The bigger issue to me is that the tile and flashing would be partially blocking the inlet. Air can't go up the valleys. A picture may be worth a thousand words.

[inetdog]

With the L-foot and ironridge XRL rail, it would be 4.146" off of the deck. The bigger issue to me is that the tile and flashing would be partially blocking the inlet. Air can't go up the valleys. A picture may be worth a thousand words.

Depends on what the exact geometry is between the point where the flashing is attached under the roll roofing to the edge of the solar panels and on to the point where the flashing comes over the top of the tiles.
If there is enough room for air to enter between the flashing and the panels, then it does not matter whether that air originated at the edge of the roof and came up the valleys or came in over the flashing.

That is, if the separation between the flashing and the bottom edge of the panel is 2", then it will be a significant restriction. If it is 3-1/2 inches, it will not be. IMHO.

[n8huntsman]

I'm imagining the interface to look something like this:

[inetdog]

I'm imagining the interface to look something like this:

If that turns out to be an accurate representation (and I think that the 5 degree slope on the transition portion of the flashing so that there is no pooling of water will be enough) then the effect on the air flow should be unnoticeable.