Is there any hard data out there or any personal experience about reduced cooling load as a result of solar panels installed on a south facing roof? So far I have found 1 study showing a 5 degree roof temperature difference on 1 roof over 3 days...

In my mind you would be able to see a marked difference in summer heat gain from an exposed roof compared with a roof with solar panels.. The home I currently am investigating is in New England, has a dark colored shingles and I'm looking in particular into a finished attic space with a directly south facing roof plane with a low pitch. The room below has a cathedral ceiling and it appears to have R-19 below the roof. We are designing to use about 58% of the area. The panels will be offset 4 inches from the roof, so I'm assuming that they will provide more shade from the sun's radiation than the heat that they will radiate into the roof.

So am I right to say that this room below will stay cooler, would there be a way to put a number to that? And therefore reduce the amount of air conditioning power draw and bring that many kilowatt-hours to using net-zero annual energy.

Tags: j, manual, modeling, net-zero, solar

Views: 1860

Reply to This

Replies to This Discussion

Dannie said: " As the temperature of the solar panel increases, its output current increases exponentially,"  No.   As the temperature of the panel increases there is a tendency for increase recombination (leakage) of the PV cells.  You're using a diode current example... in which heat helps the mobility of the electrons...  PV cells are very poor diodes.   (FWIW - PV cells can also act as light emitters when voltage is applied on them... but very poor emitters).

See: "The Performance of solar cells is dependent on environmental conditions and their output parameters such as output voltage, current, power, and fill factor vary by temperature. Experimental results showed that the most significant changed by temperature is voltage which decreases with increasing temperature while output current slightly increase by temperature. Reduction in the open-circuit voltage for silicon solar cells is about 2mV/°C. As well as the effect of temperature on the maximum power output is menus 0.005 mw/°C. The best performance of solar panels in sunny and cold day has been suggested. Reference link:

Think of it this way... you have a large tub with a sump pump in it that is supposed to move the water up ten feet and out the basement.   But there is a crack in the pipe.. that grows in width as the temperature increases.   When the pump is running (light on it) the out flow of the pipe will depend on how much water leaks out of the crack in the pipe and recombines back in the tub.  The output of the pump may seem to increase - because the water is pumping up as far before it leaks out... but the net change is not going to be significant.

FWIW,  I'm an electrical engineer....

The reason while the cold extreme temps are used for PV array simply that for nearly ALL silicon based modules there is a leakage path that increases as the temperature of the cell goes up.   The voltage remains essential the same - but the current "production" drops.   In the cold winters in Colorado with a clear sky - its possible that the current output will increase far above the STC rated conditions... and the attached loads may pull all that they can from the panel.   The part that gets over stressed (once again) in those conditions is the interconnect between the cells... and the connection between the module and the wire leads that connect a chain together.

As the current goes up... down at the inverter with its fixed impedance - the inverter may actually see a voltage rise....across its DC side.   You get into problems if the designer and installer tried to maximize the system voltage without considering the winter time cooling effects.

After re-reading this I realized that I forgot to explain that in colder temps the voltage across each PV cell increases by about 0.1 Volt (roughly),  in most common roof top applications the module will have 60 cells,  so the cold temps could be 6V higher, if the system is configured in a string - that(perhaps 15 modules in string) that could be a 90V difference between normal day and cold day - which is why the cold temp design for the strings is so important.  (More likely strings are 10 moudles or less with a combiner to put strings in parallel).   Microinverters don't have this problem.

Basic principles to always follow:

Keep the solar PV panels as cool as possible. 

Insulate the roof as much as possible R 40? or better, with venting above the insulation, not through it,  if at all possible to keep heat in winter and solar gain heat out. 

Avoid any thought of sky lights on a pitched roof. Figure out another way to get light in the space. 

PV modules are going to be in full sunlight (if possible),  worrying about the heat on panels is secondary to the good sunlight and secure mounting on the roof.  

The original post is now nearly three years ago, at the time he was looking at a house and looking to confirm if the solar panels MIGHT make a difference.  I had previously measured the effect... and as noted it is mostly insignificant.

Insulation has a bigger impact.

I have used IR equipment in thousands of houses, both new and existing, since 1981, and my observation is that it does not matter what the temperature in the attic is, it matters what the temperature of the drywall surface inside is.  

For the last 20 years or so we have seen a minimum of R-30 and now R-49 by code in new houses in our mid-Atlantic region, and this has lowered the drywall surface temp over those attics with less.  We used to see a temp about 3-4° hotter than the air temp in the room with a blown attic at R-30 or less, and now we are seeing closer to a 1-2° difference with an R-49 with a blown attic, less with more.  

Once the drywall surface is within a degree or so of the air temp, then there can't be much effect for lowering the attic temperature further.  Of course in our mixed region, this has hurt in the winter as well as helping in the summer.


Correct,  but drywall is a essentially a thermal bridge... with lots of surface area.  You can model it easily with Therm.  I believe they have a dry wall material layer already including in Therm.  I did not want to put thermal sensors attached to the underside of the ceiling so I put the sensor above the sheet rock and under the insulation.   

FWIW I also have thermal cameras... and I do have sensors that are strong up along the interior wall in several rooms where we've been measuring the temps near the ceiling all the way down to the floor... for about the last seven years.

As I noted... its the insulation that makes the most impact.  But I can see that the roof area under the panels are indeed at a lower temp and also exposed to less UV.  Not discussed is would that increase or decrease the life of the roof under the solar array.   My guess is the it well "depends"  -- in a NW environment like Seattle - moss on the roof can be more damaging than UV on the roof membrane.  I haven't seen any long term study research that provides any guidance one way or the other on the aging of the roofing.

Therm is available from the Lawrence Berkeley National Labs  at:

Not sure the drywall being a thermal bridge pertains in this discussion - it is on the inside of a bunch of insulation and serves to add a bit of continuous R-value and provides airtightness.  What are the readings you are getting on the attic side surface of the drywall?

ALL building materials are thermal bridges which allow for the flow of heat in either direction. (Some materials may have a directional effect).  Concrete foundations will act as a thermal bridge and provide a good path for heat loss (or gain) unless the interior floor and space is isolated (insulated) from concrete.

The temperature inside the living space was within perhaps 0.5C  or less.  On other charts (not uploaded) I monitored the temps in the garage under the first floor (my shop area).  The temp there was about 1.0C less than the second floor of the house temps.  All that shows is the temperature stacking in a house.

If the attic temps get HOT (Phoenix, Tucson, etc)... that sheet rock can act as an effective radiator into the room below.   Heat normally rises away (convection), but radiation/conduction still occurs (albeit with losses from convection along the path).  During the heating season drywall conducts heat quite nicely into colder spaces.  It is a bidirectional thermal material.

The original question was do panels make a difference... and the answer is a qualified yes (as I noted in my explanations above).  If you've put panels on a roof in AZ or TX on a house that little or no attic insulation... you would see a slight difference in the ceiling temps..... but as I already noted ... by far the biggest improvement is to INSULATE and insulated beyond the minimal prescriptive code.  When you insulated the difference that is seen down at the sheet rock layer... is essentially the same and becomes then dependent on other things happening in the room... one of which can be seen on the chart in which windows in the room on the south side were open and sheet rock temps dropped further at night... where as the temps on the northside where the room temps remained about the same didn't drop much.  That drop was about 1F.... 0.5C roughly... not enough to base a house design on... it was/is mostly insignificant.

The original question was about roof temps...  and indeed I can see cooling on the bottom side of the roofing plywood.  But again as noted - what's the value in that during the lifetime of a roof... probably not enough info to really determine.


I get what you are saying, and that definition makes R-13 batt insulation and an inch of styrofoam thermal bridges also - it's all just a matter of degree.  

And room temp is effected by surface temps.  The closer to the interior air temp the surfaces are, the less time the heater/AC spends running.

The surface temperature of a material determines how much of a radiator it is.  If the attic has no insulation making the interior surface temperature of the drywall  100° on a hot day, then you have a very good radiator.  If the attic is well insulated and sealed, and the surface temperature of the drywall is 76° with a 75° thermostat/air temperature, the radiator is not so good.

Same argument as radiant barriers - yes they work, but cheaper and better to just insulate.

I would think that in a well insulated and sealed building, attic temperature fades from importance.  However, the room in your original question is one of the most difficult to properly seal and insulate, but also one of the most important.


The radiation "effectiveness" doesn't change with temp... its the heat flow that changes with the delta-T.   

Q = U x A x T   where Q is the heat flow (either direction), U=1/R with R being the insulation R value (or the wall assembly R value),  A being the area that has the  value of U and exposed to the delta T (temperature).  If delta T = 0, the flow of heat is zero.  

The radiation "effectiveness" is encapsulated in the U.  For most materials the U values stay relatively stable, there are big exceptions of course - phase change materials,  insulation that gets wet... and some specific insulation that may change U values when looking at wider temperature ranges.  But at a range of a few degrees... the radiate effect of the surface is stable... radiation can occur from a surface even when the room air temperature and that surface temp are the same.   Conduction requires contact between materials (air/ceiling),  convection is the air movement... but a ceiling can radiate heat down to objects on the floor...  We feel cold when standing to a single or double pane window during the winter time... not because of direct contact with the window... but because our skin radiates energy out through the window to the dark space.   If the ceiling drywall is slightly warmer (1F) than the a person sitting on the floor... even though the temperature around the person may be in equilibrian... they may still "sense" the ceiling is warmer because of the heat radiation toward them.

Remember the roof question was posted by another person (Tim) asking: "Is there any hard data out there or any personal experience about reduced cooling load as a result of solar panels installed on a south facing roof? So far I have found 1 study showing a 5 degree roof temperature difference on 1 roof over 3 days..."    my response was I've previously instrumented roof, attic and house and have measured the effects.  And as I noted multiple times in the thread... that the most effective solution to cooling the occupied space below the roof... is insulation - more than the prescriptive minimum.

In Alabama the duct work is usually in the hot attic. So it is not just the drywall temperature. IMO ducts should only be allowed inside the thermal envelope.

When solar pool panels are removed from an asphalt shingle roof the roof looks almost new. I wonder if it would be cost effective to only replace the exposed shingles?

Generally when PV modules are added to a roof, the installers are supposed to ensure that AT LEAST fifteen years remain on the roofs life span (more likely they'd look for 20 years).

It would be difficult to do an adequate job that a roofer would certify for another twenty+ years if they just replace single shingles.  I don't believe the manufacturers would approve of that... and in some cities and states  roofing jobs need building permits and inspections.  I doubt that an inspection would approve of anything that doesn't match the manufacturers recommended steps.

DOE has capture a lot of reference material for ducts in attics - particularly in the south.  They have investigated and do have recommended steps for "burying the ducts"  in combination of closed cell foam and blown insulation.

There isn't any question that ducting AND HVAC systems should not be in the attic -- or in a crawl space... poor building design - but sometimes that's what the owners are stuck with.

FWIW, TX. LA, MS, AZ also have ducting (and HVAC) in attics... as do some other southern temperate locations like CA.   Mostly these are all designs and builds from decades ago.... at a time that the energy consumption was not on the top of the list... but lowest costs builds were...

I was in New Orleans a few weeks ago... looking at some of the architecture of homes... those older 50's and 60's generation homes didn't have air conditioning... just low efficiency furnaces (when they had them).   Air conditioning was often added in late 60's and 70's.

I took the Amtrak train (City of New Orleans) to Chicago,  along the tracks north of New Orleans... along Lake Ponchartrain past houses built on pilings driven into the lake and swampy areas.  In those areas its the latent heat (humidity) that is being addressed with air conditioning... and its extremely hard to solve a RH problem in a house built above a lake.... in the south.

In general we have a lot of residential buildings out there that are extremely energy inefficient... because of old codes, history,  and going for lowest cost only....

Solar panels on the roof... based on the data I have do not significantly alter the year around heat/loss through the roof... but the energy produced can significantly reduce the net consumption of the building.   PV in cooling environments (like the south) more frequently coincides with the hot humid days on which air conditioning is likely to be used.


Latest Activity

Emily Ambrose's blog post was featured

Focused Strategy: The Rising Tide Effect of Precision Marketing

Among the many challenges we face as digital marketers, some of the strongest and most consistent…See More
47 minutes ago
Profile IconGreg Hawthorne, Christine Brill and JES joined Home Energy Pros Forum
2 hours ago
Franco Oyuela replied to Steve Nations's discussion high rise humidity
"Why Is My House So Humid? imply put, high humidity in a home occurs when moisture comes into a home…"
19 hours ago
Franco Oyuela commented on John White's blog post HVAC Spring Maintenance Checklist
"Spring HVAC Maintenance Checklist. Inspect the indoor coil for cleanliness and clean if…"
20 hours ago
Brett Little posted an event

How to use the Home Water Score - Free CE Webinar at Webinar Online

March 20, 2019 from 12pm to 1:15pm
The home water score (v1) is here and can be used now to make more informed and strategic decisions…See More
Don Fugler replied to Don Fugler's discussion Difficult retrofits in the group Kitchen Ventilation
"So here is my first report (after which I will wait until I have multiple trials). I am trying to…"
Tommie John replied to Tommie John's discussion Blower Door and Duct Blaster! in the group Energy Auditing Equipment for Sale, Trade or to Purchase
"You can reach me at 281-310-0258 directly "

© 2019   Created by Home Performance Coalition (HPC)   Powered by

Badges  |  Report an Issue  |  Terms of Service