Thermal heat reduction impact of solar panels on roof vs. shading optimization

[huge]

This may be a question for the pros, but might help everyone to some extent with all the expert knowledge that is given in these forums.

My main goal is to save money and second is doing it with the least amount of time I have to waste in the future on these. I live in a part of southern California that gets very hot in the summer, over 100F for several weeks and over 90F for several months (total per year). My cement tiles get very hot and continue to heat the house well into the night and our somewhat inefficient AC has to run a long time, causing a lot of electricity use. I am hoping that getting some solar panels will make the roof cooler and decrease our energy needs for the summer

So question #1: how much do the panels help decrease the roof tempratures in the hot months? Is it enough to justify getting panels that are less efficient and spread them out, instead of just focusing on the South?

The second question has to do with the shading in this situation. I have a limited space in the South side. The East side has some minor shading issues (Direct sun starts after first 2-3 hours after sunrise then the shading hits again in the later parts of the afternoon. The west side has even more shading issues (partial then full shading starts 1-2 hours after mid day).

#2. Given these heat and shading issues, is it better to focus on putting panels on the least shaded areas for optimal efficiency (East and South), or spread them out for optimal heat control (East West, South)?

On a side note, the quotes I got have the Solaredge inverter with optimizers at $1000 more expensive than the old Sunny Boy inverter. I will probably go with the Sunny Boy if I have to have only 1 row on the South and 1 row on the east, but is it worth the extra $1000 if I have to spread them out to the West as well, given the shading issues?

Thanks a lot for any input

[DanKegel]

Just painting the roof white may be better at cooling the house, see http://link.springer.com/article/10....546-016-0160-y

So, no, don't spread them out, paint your tiles white

[J.P.M.]

Since you ask, the short answer is do not worry too much about panel or array placement for the sake of the heat that it will keep from the interior - that's a minor advantage, and mostly a minor concern if adequate insulation is present - which is the no brainer solution if insulation is missing. So, the short answer to your first question is I wouldn't worry about it - add insulation.

The answer to your second question is a bit more complicated. Simply put, the panels should be placed in locations that result in the safest, most serviceable operation, and produce the most annual output. Aesthetics and other things are also considerations. Often, it's a bunch of tradeoffs.

FWIW, here's some numbers from my roof, winter,10/09/2014 - 02/09/2015 and summer, 06/14/2015 to 08/15/2015. The roof is med./dark brown lightweight concrete tile. All measurements taken at min. solar incidence angle for the days when measurements were taken. All reading taken and recorded with a hand held IR thermometer. All roof deck temps from under the array are the average of 32 readings for each day's readings - once under each of 16 panels on either side of min. incidence angle time. The exposed roof deck temps were only 4 - one on either side of the array at beginning/end of the 16 min. data gathering. All days when measurements were taken were "clear". All readings are instantaneous (well, not really, but sort of close), that is, they are a "snap shot" of conditions at the 16 min. - 8 min. on either side of the time of minimum incidence angle on the roof, and hence the array that's on it, for the days when measurements were taken. Results are applicable to my roof only, and only as examples of relative roof temps. Your conditions may be similar, or not. My guess is my #'s are somewhat representative of others in So. CA with similar roof materials, colors and conditions, bit no guarantees are expressed or implied.

Winter, 34 measurement days.

Average roof surface temp. under the solar array: 28.6 C.
Average temp. of the exposed roof: 60.3 C. (not under the array)
Average ambient air temp. on the roof: 23.9 C.

Summer, 26 measurement days.

Average roof surface temp. under the array: 38.6 C.
Average roof temp. on the exposed roof: 67.9 C.
Average ambient air temp. on roof: 33.1 C.

Ave. wind speed, winter & summer, ~ 1.9 m/sec., continuously recorded 1X/minute and averaged for the 16 min. period on event days.

A part of the array is over my living room and that ceiling has no crawl space over it - just wall board, rafters and insulation, then roof. During the summer, the ceiling where the array is runs about 2 or so deg.C. warmer than the rest of the ceiling in that room that is not under the array.

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

[huge]

Great replies guys. That article and the reply by JPM were excellent. You guys are so good at this that I kind of wish someone like JPM could do my panels

[J.P.M.]

You guys are so good at this that I kind of wish someone like JPM could do my panels

I'm retired. Ask Dan, Willaby and 55ncs what they think of my opinions.

[DanKegel]

JPM's helpful when describing how to do something useful, or how solar systems work, as he was here. His data's great.

But there are other data points, e.g.

http://www.sciencedirect.com/science/article/pii/S0038092X11002131

which says
"daytime ceiling temperatures under the PV arrays were up to 2.5 K cooler than under the exposed roof. Heat flux modeling showed a significant reduction in daytime roof heat flux under the PV array. At night the conditions reversed and the ceiling under the PV arrays was warmer than for the exposed roof indicating insulating properties of PV. Simulations showed no benefit (but also no disadvantage) of the PV covered roof for the annual heating load, but a 5.9 kWh m−2 (or 38%) reduction in annual cooling load. "

Another paper (can't find it now) says the effect varies strongly with height above the roof, something about convection.

Anyway, it's complicated. You might want to read up on Cool Roofs in general. Good luck!

[Willaby]

Well, I just gave JPM a thumps-up-like because he used an emoji, a rare event.

[Mike90250]

When designing the system you must allow for voltage droop when panels are hot.

Many schemes have been tried to cool panels, but they consume more power than is recovered.

[FFE]

I used my IR temp sensor on the ceiling inside my house the first week of November. My panels are on the vaulted ceiling part of my house. Asphalt shingles and probably R19 insulation then sheetrock. The ceiling is white so the IR sensor isn't really designed for this. Anyway, the ceiling under the array was 2F cooler than the area that wasn't. I measured at the same "height" since my ceiling is hotter the higher one measures. Yes I realize that it is not a true representation since the hot air in the room also heats the ceiling and other heat transfer issues... But, I could outline my panels using this method. I wish I had a FLIR to get better info. If you want I can take readings tomorrow if you are curious. I will probably end up with the same numbers as JPM.

[J.P.M.]

[QUOTE=FFE;n315644] The ceiling is white so the IR sensor isn't really designed for this. QUOTE]

The thermal emissivity of white paint is about the same as most paint colors in the infrared and something of the order of 0.92 -0.95 or so. The color is a perception of the human (and other species) eye. The IR sensor will see it the same way it sees any other color of the same product. If the sensor has an adjustment for emissivity, use it.

Black paint will register and be warmer than some color paint or white paint of the same formulation (except for the pigment) because of its absorptivity in the solar spectrum. If of the same material, and at the same temperature, the color won't make much, if any difference. The IR thermometer will read about the same for either sample. The IR thermometer is "seeing" thermal radiation, not "color" as we know it. The black surface may be warmer than the white one because it's absorptivity to solar radiation is higher than other some color or white.

[J.P.M.]

I used my IR temp sensor on the ceiling inside my house the first week of November. My panels are on the vaulted ceiling part of my house. Asphalt shingles and probably R19 insulation then sheetrock. The ceiling is white so the IR sensor isn't really designed for this. Anyway, the ceiling under the array was 2F cooler than the area that wasn't. I measured at the same "height" since my ceiling is hotter the higher one measures. Yes I realize that it is not a true representation since the hot air in the room also heats the ceiling and other heat transfer issues... But, I could outline my panels using this method. I wish I had a FLIR to get better info. If you want I can take readings tomorrow if you are curious. I will probably end up with the same numbers as JPM.

While there is some benefit to the shade provided by an array, it's not the biggest benefit.

My solar magnum opus residence out in Borrego Springs (designed through prelim. plan check but not built) was designed to have two roofs, one on top of the other. The HERS auditor wouldn't sign off on lower roof insulation required from lower roof skin temps, but I left the double roof anyway. I even had a slick way to keep the kangaroo rats and other critters out of the sandwich area. The top section was removable in sections for service.

[solarix]

I say - If you live in an area with high roof temperatures, don't put electronics (microinverters or optimizers) up there.

[SunEagle]

[QUOTE=J.P.M.;n315649]

The ceiling is white so the IR sensor isn't really designed for this. QUOTE]

The thermal emissivity of white paint is about the same as most paint colors in the infrared and something of the order of 0.92 -0.95 or so. The color is a perception of the human (and other species) eye. The IR sensor will see it the same way it sees any other color of the same product. If the sensor has an adjustment for emissivity, use it.

Black paint will register and be warmer than some color paint or white paint of the same formulation (except for the pigment) because of its absorptivity in the solar spectrum. If of the same material, and at the same temperature, the color won't make much, if any difference. The IR thermometer will read about the same for either sample. The IR thermometer is "seeing" thermal radiation, not "color" as we know it. The black surface may be warmer than the white one because it's absorptivity to solar radiation is higher than other some color or white.

+1. Good info on IR and emissivity of paint when using an IR camera.

[rsilvers]

I don't see why the shade from an array would not be a giant benefit. The sun is putting 1000 watts per square meter into your roof. Think of your roof as a solar thermal heater.

I have a FLIR and can do some testing later. I just tested a $30 Duck brand attic stair cover and decided it was not good enough.

[SunEagle]

I don't see why the shade from an array would not be a giant benefit. The sun is putting 1000 watts per square meter into your roof. Think of your roof as a solar thermal heater.

I have a FLIR and can do some testing later. I just tested a $30 Duck brand attic stair cover and decided it was not good enough.

I agree that you should see lower temperatures inside a home if you install a pv array on your roof. Although a white roof surface may reflect more heat than what may be absorbed by the black surface of your pv panels.

To make sure the heat the panels absorb does not go into your attic you need to make sure there is enough air circulation between the panels and the roof surface. That will also keep the panels happy.

Also venting the attic and using a thermal barrier near the top inside of the roof frame can lower temperatures inside a home.

[DanKegel]

Cool, please do!

I bet tilted panels on a flat roof would have a larger cooling effect - the hot panels would be well up off the roof, with plenty of ventilation.

[ButchDeal]

Cool, please do!

I bet tilted panels on a flat roof would have a larger cooling effect - the hot panels would be well up off the roof, with plenty of ventilation.

It will all be site specific

higher roofpitch will cause more convection between roof and modules,
most racking has hight adjustments, install a few inches higher will help as well.
effect on the home is also site specific. For example I have energy start shingles which reflect a great deal of the energy, so the PV has less thermal impact than if I didn't have them.
Better insulated homes, better ventilated attics etc, all will decrease the impact of the PV on the home.

The point is that it is a good idea to do these other things before solar as they are usually cheaper than solar and have a larger impact.

Where PV thermal effect on cooling the home would have the greatest impact would be on a home with vaulted ceiling. Insulation is limited, and no attic to ventilate.