Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Affect of heat & cold on Solar Glass ?

This site may earn commission on affiliate links.
A study a couple of years ago on solar production potential noted that panels in northern climates will often produce more electricity than those in southern climates because the panels are less efficient when they heat up.

Any idea how Solar Glass will perform on this? Given the design I'd guess that the difference would be more pronounced as Solar Glass seems to have less capability for cooling so may have a greater decline in output when hot than a standard panel.
 
Interesting question. A SolarGlass roof tile sits 3/4" or so over the roof decking and underlayment. And there are vents/drains at the base when the gutters are, and vents and caps at ridgeline. So it does have airflow underneath unlike a composite roof or ceramic roof. In additional to ventilation, in this space is where they route the power cables from the tiles to the inverters.
 
Interesting question. A SolarGlass roof tile sits 3/4" or so over the roof decking and underlayment. And there are vents/drains at the base when the gutters are, and vents and caps at ridgeline. So it does have airflow underneath unlike a composite roof or ceramic roof. In additional to ventilation, in this space is where they route the power cables from the tiles to the inverters.

Yes, I think these gaps are intended at least in part to help with air flow/cooling, though Tesla also suggests they help with keeping the house cool (not necessarily the panels):

Exceptional Performance
Keep your home cool and comfortable by allowing hot air to vent with high-quality underlayment and integrated air gaps.
About Solar Roof | Tesla Support

Also, Tesla does provide certain performance guarantees, so while I do believe that heat reduces production, hopefully Tesla is taking that into account and their estimated production reflects that - otherwise they could potentially run into issues with the guarantees.
 
I don't know where the conclusion that there's less capability for cooling comes from, they are vented underneath. Certainly it would be expected as with any solar cell that it will be less efficient the hotter it gets. I don't know if anyone has any data on what temperature a typical racked panel gets to on a hot/sunny day, I'd be able to take surface thermocouple, contactless IR thermometer, or IR camera shot of my roof (as well as either of those two latter things work on a reflective surface), though getting good correlation to a panel would still be an issue unless it was right down the street, even for my town I'm in an unusually hot area, temps in my neighborhood are generally 5-15 degrees hotter than the 'official' San Jose temperature on any typical summer day. There does appear to be some degradation on hotter days, though of course there are many other possible variables such as haziness, inverter temperature, etc.

Last week here there were a few "clear" but increasing temperature days, if you look at my PVOutput data (link in signature):
5/23 (73.7 kWh) the peak temperature in my back yard was 93.9.
5/24 (71.0 kWh) the peak temperature in my back yard was 100.9.
5/25 (68.5 kWh) the peak temperature in my back yard was 108.5.
5/26 (67.4 kWh) the peak temperature in my back yard was 112.8.
5/27 (68.9 kWh) the peak temperature in my back yard was 110.3.
5/28 (69.7 kWh) the peak temperature in my back yard was 100.9.
[Yeah, last week was not fun, considering that my house has no AC and uninsulated walls. :p Good news is the attic doesn't get as hot since the Solar Roof was installed (presumably due to that vent layer), so that has seemed to help the peak house temps a bit]

So 5/24 and 5/28 both had the same locally-measured high temperature (and same average wind according to my weather station also), and both curves on PVOutput are smooth (no clouds, etc), but the production is still ~2% different (and the sun should still be getting higher/longer each day, so 5/28 should have been the better day from that perspective). Clearly temperature isn't the only thing having an impact here, so I find it difficult to draw any real conclusion from the above data, the impact could have more been because the hotter days were hazier, for example.
 
Yes, I think these gaps are intended at least in part to help with air flow/cooling, though Tesla also suggests they help with keeping the house cool (not necessarily the panels):


About Solar Roof | Tesla Support

Also, Tesla does provide certain performance guarantees, so while I do believe that heat reduces production, hopefully Tesla is taking that into account and their estimated production reflects that - otherwise they could potentially run into issues with the guarantees.

There's a bunch of research related to the effect on building cooling and heating requirements.
I started from reference to a 2011 paper.

https://scholar.google.com/scholar?q=paper+j.solener.2011.06.010&hl=en&as_sdt=0&as_vis=1&oi=scholart
 
Yes, I think these gaps are intended at least in part to help with air flow/cooling, though Tesla also suggests they help with keeping the house cool (not necessarily the panels):


About Solar Roof | Tesla Support

Also, Tesla does provide certain performance guarantees, so while I do believe that heat reduces production, hopefully Tesla is taking that into account and their estimated production reflects that - otherwise they could potentially run into issues with the guarantees.

I imagine any time you can put a shade over your roof it is going to be a little cooler. Also, the SolarGlass roof install has a Firestone Clad-gard SA-FR underlayment which as a foam insulation material against the decking boards. And then above this there is the 3/4 inch or so air gap to the solarglass tiles. So the layers are deck board, SA-FR underlayment, air gap, SolarGlass.
 
I imagine any time you can put a shade over your roof it is going to be a little cooler. Also, the SolarGlass roof install has a Firestone Clad-gard SA-FR underlayment which as a foam insulation material against the decking boards. And then above this there is the 3/4 inch or so air gap to the solarglass tiles. So the layers are deck board, SA-FR underlayment, air gap, SolarGlass.
Yeah - this all makes sense, and it is actually one of the extra benefits we are hoping to see over the summer is a slightly cooler attic. I was mostly noting that Tesla's website is advertising the air gap for purposes of helping cooling the house, as opposed to cooling the roof panels. Not to say it doesn't do both - just what they are plugging.

Interestingly, when we were first working with them on v2 panels, their plans included installing a ridge vent to help with circulation/cooling. That was dropped when v3 came along.
 
I just checked the specs on my REC N-peak PV panels and found they have a 0.35%/ºC derating with a base temperature of 44ºC. Some of the newer cell technology have lower derating for temperature. Tesla did not do my solar system, but I have the Tesla batteries.
 
Interestingly, when we were first working with them on v2 panels, their plans included installing a ridge vent to help with circulation/cooling. That was dropped when v3 came along.
That’s interesting since for my v3 they said a ridge vent was required and they installed it. The airflow and coolness of my attic since the v3 was installed has been unexpectedly fantastic.
 
I don't know where the conclusion that there's less capability for cooling comes from, they are vented underneath. Certainly it would be expected as with any solar cell that it will be less efficient the hotter it gets. I don't know if anyone has any data on what temperature a typical racked panel gets to on a hot/sunny day, I'd be able to take surface thermocouple, contactless IR thermometer, or IR camera shot of my roof (as well as either of those two latter things work on a reflective surface), though getting good correlation to a panel would still be an issue unless it was right down the street, even for my town I'm in an unusually hot area, temps in my neighborhood are generally 5-15 degrees hotter than the 'official' San Jose temperature on any typical summer day. There does appear to be some degradation on hotter days, though of course there are many other possible variables such as haziness, inverter temperature, etc.

Last week here there were a few "clear" but increasing temperature days, if you look at my PVOutput data (link in signature):
5/23 (73.7 kWh) the peak temperature in my back yard was 93.9.
5/24 (71.0 kWh) the peak temperature in my back yard was 100.9.
5/25 (68.5 kWh) the peak temperature in my back yard was 108.5.
5/26 (67.4 kWh) the peak temperature in my back yard was 112.8.
5/27 (68.9 kWh) the peak temperature in my back yard was 110.3.
5/28 (69.7 kWh) the peak temperature in my back yard was 100.9.
[Yeah, last week was not fun, considering that my house has no AC and uninsulated walls. :p Good news is the attic doesn't get as hot since the Solar Roof was installed (presumably due to that vent layer), so that has seemed to help the peak house temps a bit]

So 5/24 and 5/28 both had the same locally-measured high temperature (and same average wind according to my weather station also), and both curves on PVOutput are smooth (no clouds, etc), but the production is still ~2% different (and the sun should still be getting higher/longer each day, so 5/28 should have been the better day from that perspective). Clearly temperature isn't the only thing having an impact here, so I find it difficult to draw any real conclusion from the above data, the impact could have more been because the hotter days were hazier, for example.
With my Enphase inverters I am able to read on board temps for each inverter. I am just not sure if that is panel temp(there is no small wires from panel to inverter just power) or the inverter temp.
But, let me assure all readers, the temp readout on my rack installation gets very hot on hot days.
At this time, last temp reading at 11AM was 70F at the house and 98F on the inverter/panel.
I was able to look at historical data, Yesterday, the 3rd it reached 122F panel/inverter peak at 2 PM. Don't know what outside temp was.
May 25th it was 123F up there.
I just love this about Enphase inverter data, 6 parameters recorded and it seems stored from day 1 of operation.

ps. a question on that solar glass vent, does it vent the attic or just under the panels..

Either way the area under panel doesn't get radiant energy to conduct into the attic, hence cooler attic but if other areas do get radiant energy on surface to conduct into attic, the whole attic will have air mixing, but perhaps at a slightly lower temperature.
 
Last edited:
Yeah - this all makes sense, and it is actually one of the extra benefits we are hoping to see over the summer is a slightly cooler attic. I was mostly noting that Tesla's website is advertising the air gap for purposes of helping cooling the house, as opposed to cooling the roof panels. Not to say it doesn't do both - just what they are plugging.

Interestingly, when we were first working with them on v2 panels, their plans included installing a ridge vent to help with circulation/cooling. That was dropped when v3 came along.

Are you talking about a vent for the attic? We already had some 2' X 3' vents in each end of our attic and I don't think they added anything extra. It has never gotten too hot up there. Probably no more than a few degrees about the ambient air temps even in our 100+ summers. Also helps that it is tall, at the highest it is 8+ feet above the top of the ceiling joists.
 
Are you talking about a vent for the attic? We already had some 2' X 3' vents in each end of our attic and I don't think they added anything extra. It has never gotten too hot up there. Probably no more than a few degrees about the ambient air temps even in our 100+ summers. Also helps that it is tall, at the highest it is 8+ feet above the top of the ceiling joists.
Yes, in the attic - a ridge vent is at the top of the roof (along, or really just below the ridge line.) It looks almost as if the top of the roof was raised up a few inches to allow a gap for ventilation.

Our attic is about the same height at the peak, but it does tend to get pretty hot/humid in the summer. We are hoping this will help a bit. We don't actually use our attic as living space, so it is not crucial, but the temp/humidity can in extreme heat cause condensation.
 
Yes, in the attic - a ridge vent is at the top of the roof (along, or really just below the ridge line.) It looks almost as if the top of the roof was raised up a few inches to allow a gap for ventilation.

Our attic is about the same height at the peak, but it does tend to get pretty hot/humid in the summer. We are hoping this will help a bit. We don't actually use our attic as living space, so it is not crucial, but the temp/humidity can in extreme heat cause condensation.
From my time living in Virginia I remember that humidity was the big thing.

Here, when the temps go up, the humidity drops like a stone. My weather station in the yard shows 9% humidity and temp 95F. Was 60% humidity and temp 62F this morning. No condensation worries here.
 
Last edited:
Based on the conversation I decided to climb out on the roof and check the roof cap. There is an opening and if I stick my face down there I could see the same metal strip with slots that is used at the base of the roof to lead water into the gutter. So it looks like the Solarglass roof does come with vents on top and bottom. See picture below of cap and part of the vent slots. The vents run an inch portion visible in picture. (Sorry, bad angle)
20200605_092257.jpg


20200605_091841.jpg
 
I don't know where the conclusion that there's less capability for cooling comes from, they are vented underneath. Certainly it would be expected as with any solar cell that it will be less efficient the hotter it gets. I don't know if anyone has any data on what temperature a typical racked panel gets to on a hot/sunny day, I'd be able to take surface thermocouple, contactless IR thermometer, or IR camera shot of my roof (as well as either of those two latter things work on a reflective surface), though getting good correlation to a panel would still be an issue unless it was right down the street, even for my town I'm in an unusually hot area, temps in my neighborhood are generally 5-15 degrees hotter than the 'official' San Jose temperature on any typical summer day. There does appear to be some degradation on hotter days, though of course there are many other possible variables such as haziness, inverter temperature, etc.

Last week here there were a few "clear" but increasing temperature days, if you look at my PVOutput data (link in signature):
5/23 (73.7 kWh) the peak temperature in my back yard was 93.9.
5/24 (71.0 kWh) the peak temperature in my back yard was 100.9.
5/25 (68.5 kWh) the peak temperature in my back yard was 108.5.
5/26 (67.4 kWh) the peak temperature in my back yard was 112.8.
5/27 (68.9 kWh) the peak temperature in my back yard was 110.3.
5/28 (69.7 kWh) the peak temperature in my back yard was 100.9.
[Yeah, last week was not fun, considering that my house has no AC and uninsulated walls. :p Good news is the attic doesn't get as hot since the Solar Roof was installed (presumably due to that vent layer), so that has seemed to help the peak house temps a bit]

So 5/24 and 5/28 both had the same locally-measured high temperature (and same average wind according to my weather station also), and both curves on PVOutput are smooth (no clouds, etc), but the production is still ~2% different (and the sun should still be getting higher/longer each day, so 5/28 should have been the better day from that perspective). Clearly temperature isn't the only thing having an impact here, so I find it difficult to draw any real conclusion from the above data, the impact could have more been because the hotter days were hazier, for example.
How big is your installation? We have a 8.6kw solar glass v3 installation in Santa Clara and I get only 17kwh per day max.. :-(
 
How big is your installation? We have a 8.6kw solar glass v3 installation in Santa Clara and I get only 17kwh per day max.. :-(
Sounds like something is wrong unless you have significant shading or no south facing PV tiles. Might want to check the inverter read outs during the day. I have 10.52kW v3 in San Jose and got 65-69kWh the past 3 days. About 50% south, 35% east, 15% west facing PV tiles.