I have seen a lot of discussion on the forum regarding the loss of power PV panels have at higher temperatures, ie, above 25 degrees up to 70 degrees.
Are there any simple ways to help cool the panels during the heat of summer that won't damage them?
I know that it is recommended to raise the panel off the roof by about 150mm for ventilation, but what about some form of forced air cooling, or reflective roof paints etc, thoughts please...
Cooling PV panels
(48 posts) (19 voices)-
Russell Moore
MemberPosted Thursday 3 Dec 2009 @ 10:55:56 am from IP # -
Russell Moore
MemberSo I take it that no one has any ideas about this?
The only site I found on the net was of a large system in California that used misting sprays between the panels, but not directly onto the panels.
Would misting sprays damage the cells if they were used directly onto the panels in heatwave conditions?
Posted Thursday 17 Dec 2009 @ 10:29:02 am from IP # -
I doubt that misting would cause enough thermal stress to damage the panels and as they are weather proof I don't think it should affect their longevity. Would you get enough extra power to run the pressure pump required to make a fine mist though?
One idea I'm looking at is using panels as eaves for my otherwise eavless house. By using them as eaves the panels should give good shade and have excellent ventilation from behind. I'd try to look for passive measures rather than active systems to keep everything as simple as possible.
Posted Thursday 17 Dec 2009 @ 10:46:38 am from IP # -
Sunshine
MemberRussell Moore,
Reflective roof paint won't work as the panels themselves are in the sun, and the roof is being shaded by the panels.
I wonder if they can panels white in the future rather than blue/black colours. That would help alot.Posted Thursday 17 Dec 2009 @ 1:07:58 pm from IP # -
Russell Moore
MemberAnother site I visited had water circulating in copper pipes underneath the panels, with a copper heat exchanger attached to the panels, the water was fed into a solar hot water heater, and thus acted as a preheater for the SHW system.
Posted Saturday 19 Dec 2009 @ 7:40:07 am from IP # -
Peter R
MemberSunshine: PV panels are designed to absorb as much light as possible. The blacker they appear the better they do their job. White panels would reflect all visible light and not produce much power at all (maybe some from UV and IR).
Russel: one reason for misters trying to stay away directly from the panels would be to avoid reducing the insolation. Mist, like clouds would reduce the amount of light that gets to the panels. Running water over panels is not useful too for the same reason, it also increases reflectivity of the surface.
Having your panels off the roof to allow a sea breeze to get underneath is very helpful. The idea of using panels as a pre-heater for a SHW system sounds appealing, but water and electricity don't mix well, if you can't find a professional solution I would not do that as a DIY project.
Posted Saturday 19 Dec 2009 @ 10:26:01 am from IP # -
Russell Moore
MemberThe idea behind the mist sprays would be that the mist would need to be very fine, around 5 microns, and this would cool the panels without wetting them too much, if at all.
The possibility would be to have a thermostat under the panels and a solenoid valve to operate the mist system.
Seabreezes are not all that reliable here in western Sydney, and on the really hot days, 35 degrees plus, we usually get a searing NWer.
The idea of the SHW preheater was just that, an idea that I saw on a web site, I don't have any plans to try it!
Posted Saturday 19 Dec 2009 @ 11:52:52 am from IP # -
Russell Moore
MemberThere is very little information on the net about ways to cool PV panels, and the increase in power that you may get from doing so.
One idea I have contemplated is a simple and cheap method. An oscillating sprinkler could be set up to a tap timer set to run, say one minute every fifteen minutes on days with temperatures in excess of 32 degrees. I could run this off a water tank and pump I have.
I did run down a youtube site where the ambient temperature was 38 degrees, and the panel temperature was 66 degrees, and after hosing down the panels there was an increase in power, but it was difficult to see exactly how much increase there was. I have seen quotes of anything from 1/2% to 1% drop off in power for every degree increase in panel temperature. Hopefully I will have my system installed soon, and I can try something along these lines myself.
Posted Wednesday 6 Jan 2010 @ 10:00:34 pm from IP # -
Sunshine
MemberPeter R,
I am talking about the material below the silicon wafer being white in colour, this shouldn't reduce the amount of light hitting the silicon.
I have found that water on solar PV panels doesn't seem to cause a decrease in power generation due to the reflectivity of light by the water.
I have found that after a rain storm in the middle of the day, as soon as the sun comes out again the amount of power being generated really goes up, around 3600w from a 3000w system.. this is when the panels are still wet from rain, so the water on them doesn't appear to affect them. the coolness of the panel though causes a temporary increase in power generation.
The bottom line though, is it worthwhile to cool solar panels by keeping it simple and reliable without using more energy than what the gain will give in returnPosted Wednesday 6 Jan 2010 @ 10:20:55 pm from IP # -
alfresco24
MemberPV cell performance depends on type (Amorphous or crystalline), the available spectrum, irradiance and angle.
Efficiency decreases with lower irradiance, eg Reduction of efficiency by 6% from an irrandiance of 1000W/m2 to 200W/m2 (cell temperature 25'C)
Cell voltage decreases as temperature rises Coefficient of Voc -1.20x10-1 V/'C (1.2V for every 10'C)
Cell current can increase as temperature rises Coefficient of Isc 5.15x10-3 A/'C (5.15mA for every 'C)
Common multicrystalline panels have 40 to 50 cells in series parallel arrangement.
Spectrum and angle changes from summer to winter.
During Summer cells are more efficient crystalline by 7-9% and amorphous by 14-16% shown in one study.Posted Monday 11 Jan 2010 @ 12:14:12 pm from IP # -
Michael L.
MemberThe instructions with my set up advise against using water on hot panels, even to clean them, as it can break the glass cover, so I think it would be wise not to try using water at all.
Posted Monday 1 Feb 2010 @ 11:08:03 pm from IP # -
swanning_it
Memberalfresco24, Are you sure you're right there? The temp. vrs IR curves I've seen show a marginal voltage rise with increase in temp, but a drop off in current giving an overall dropoff in power (this has been the same for amorphous and crystalline....just that amorphous drops off far less per degree rise than crystalline).
Munter, the eaves idea sounds great. I hadn't thought about it like that, in fact I was contemplating increasing eave overhang (Ritek panels) to increase available roof space for panels. Perhaps I need to think more about overhanging bracketing for the panels alone. Have you looked into wind loading of the panels?
Russell, I was concidering an obverse side cooled panel system, incorporated into a hot water circ. system (cooling wth the return to the tank). Since reading some of the comments on this forum, I've decided I need to look closer at the economics of hot water circ. I also thought about using a Peltier ceramic cooler or perhaps I should more correctly say the Seebeck effect cooler which could in fact use the temp difference to power the coolant pump. My water tanks are semi-underground, so a very cool temp, even here on the Gold Coast (mecca of the South pacific....hub of the universe :-))
I will try one (or both) as I don't see this as being hard nor expensive....just unworkable perhaps
Posted Tuesday 2 Feb 2010 @ 9:09:47 am from IP # -
Russell Moore
MemberI will still experiment with some form of water cooling when I FINALLY get my panels, despite what Michael L thinks about it. The cells may actually last longer if they are not baking at 70+ degrees on heatwave days.
Another thing I will try is to paint the cell frames white, and the roof slope area above the panels black, and the roof slope area below the panels white, I hope this will induce some form of convective air currents under the panels as well.
The cooler the panels the better the power output.Posted Tuesday 2 Feb 2010 @ 11:36:40 am from IP # -
Russell Moore
MemberAnother thought that I might try is a form of solar chimney that would attach to the top end of the panels, and draw air under the panels.
It would consist of metal sheeting formed into a rectangular cube that would be painted black and angled at say 45 degrees at the top of the panels, say 600mm high, as the air in the chimney heats up it will rise and draw air from the lower end of the panels, beneath the panels and out the top of the chimney. Painting the lower part of the roof white may also assist here, as would mist sprays in that spot.
Posted Tuesday 23 Feb 2010 @ 1:06:52 am from IP # -
Russell Moore
MemberInterestingly I was talking to my PV installer today, and they have seen people use sprinklers to cool PV panels, and they were not against the idea, except that they thought that the energy used to pump the water may not be recouped by increased output from the panels. As I only intend to run the sprinkler for two minutes every fifteen minutes, it may not consume that much power, and some of the water will recirculate back into my tank.
I will have to do some tests on the subject and will keep you posted.
When I talked to them about my solar chimney idea, they just thought I had too much time on my hands!!
Posted Saturday 27 Feb 2010 @ 8:28:00 am from IP # -
Russell Moore
MemberOK, I am in the middle of having my panels installed, and I have started to paint a one metre wide strip around the panels, black across the top and white on the sides and bottom. This is to (hopefully) induce some extra convective air currents under the panels.
The roof is metal, and is coloured bronze olive, a dull bush green colour. What I have noticed is that the black is very hot, the green a bit cooler, and the white amazingly cooler. Such that, on a warmish day at 25 degrees C, I can't place my hand on the black sections without burning them, the green is just tolerable, and the white actually feels cool to the touch and there is no discomfort!
I have tried an oscillating sprinkler on the panels at 2 mins on 13 mins off, over 3 cycles and the panel temperatures fell from 55C to 35C in 45 min, which should in theory increase the output of the panels by +-10%. The panels are due to be connected today.....and then I can start some real tests on output under different conditions.
Posted Wednesday 3 Mar 2010 @ 2:33:55 am from IP # -
Which brings you into the position of needing to do a tradeoff of one resource against another. What is the water consumption of the sprinklers and at what point would you consider the consumption too high to warrant it's use?
Looking forward to your test results.
Posted Wednesday 3 Mar 2010 @ 2:45:07 am from IP # -
Russell Moore
MemberMunter, the pump will use about 500 litres a day, pumping from my 10,000 litre tank, using a 300 watt pump, for 40 mins, 2mins each 15min x 20. From 10am 'til 3pm EST.
I am expecting about 200 litres to return to the tank, the rest will be lost to evaporation, this is to be confirmed. A 10% power increase will be an extra +-$1.50 a day from my system (5.04kw). The pump should consume around 200 watts all up, which is about 12c worth of buyback electrcity. Tell me if I am wrong, math is not my strong point...hahaha....
Posted Wednesday 3 Mar 2010 @ 6:41:05 am from IP # -
Russell Moore
MemberOK, I have some rough early facts and figures from my attempts to keep my new panels cool(er).
I completed the paint job around the panels, and even if it doesn't help keep the panels cooler, it looks cool, well to me it does! I am in the process of adding some polycarbonate 'wings' to guide more breeze under the panels, these are 3mm thick sheets about 1000mm long, 150mm high attached at a 45 degree angle to the external array angles. This will hopefully scoop more air under the panels.
By far the biggest effect has come from the oscillating sprinkler, it turns on at 10am standard time and runs at 2mins on, 13mins off for 20 cycles until 2.45pm. It has just enough pressure from my tank pump to cover all 28 panels.
The effect is to reduce the panel surface temperature from roughly 55c to 35c on days when we have ambient temps of 30-33c, the effect is most noticable after the first three cycles, when it stabilizes around the 35c mark.
Watching the inverter output, you can see the output rise by around 7-8% or more. In my case it climbs from 4,000w to around 4300w +-.
Today (9/3/'10) in Western Sydney at 11.30am standard time the output under a crystal clear sky was 4300w+- with an ambient temp of around 33c, this is from a 5.04kw system, consisting of 28 Suntech 180 panels and a Fronius IG40 inverter.
Comments please, good or bad!
Posted Tuesday 9 Mar 2010 @ 1:35:58 am from IP # -
rsigmund
MemberInteresting post - thanks Russell
Looks like your getting around 300W extra .. good job
Like you and others have mentioned this is at least part consumed by your pumping costs...
And in addition - you now have extra capital costs and moving parts...
It would be interesting to do a cost/benefit analysis against an extra panel for example (assuming that will not exceed the inverter limits).
Posted Tuesday 9 Mar 2010 @ 5:30:49 am from IP # -
Russell Moore
Memberrsigmund, I would need two extra panels to make up for the water cooling effect with the sprinkler, and strange as it may seem, I already have the railing in place if I was to go that way!
So how much is a panel? $800? x 2 = $1,600?, plus I would need to have the panels fitted and restrung. My pump is already in place doing very little, and I bought a 25mx12mm hose for $60, plus a sprinkler for $30,and a timer for $25, so $115 all up, plus 12c a day running costs and wear on the pump.
Will the lower temperatures of the panel extend their life?, or will the water damage them? I think the former, and I will probably only use the sprinkler on 26c+ days anyway. So I guess thats about three months of the year.
I just love to tinker as you can see! Thanks for your comments....Russell
Posted Tuesday 9 Mar 2010 @ 12:18:47 pm from IP # -
Hi Russell. 4,000 W from 5,000 W rated panels is approximately 20% derating. Have you done a breakdown of the causes of that derating such as inverter/temperature/partial shading etc? If 300 W of the derating is temperature, what is the remaining 700 W?
It sounds like you're making all efforts to extract the maximum from the panels. Have you got a picture of the sprinkler in operation?
Posted Wednesday 10 Mar 2010 @ 5:38:27 am from IP # -
Benny
MemberGood work Russell. I washed my panels a few weeks back and could see a blip in the output of about +5% but it didn't last 13mins. Maybe as you say it takes about 3 cycles to get the whole frame etc cooled down.
Sounds like another interesting article for the ATA magazine. I'm sure Lance would be willing to help with de-rating calcs etc if you gave an article + pics.
Posted Wednesday 10 Mar 2010 @ 6:59:05 am from IP # -
Russell Moore
MemberMunter, the output is AC not DC, the panels are rated in DC,(5040w) but the output from the inverter is AC.
So the max AC output at 93% inverter efficency would be 4687w AC, the max I have seen is 4,500w AC, but it could have been more, I will have to check the inverter, and see what the peak output has been, now where did I put that manual....
Also we are not at peak insolation time now, as we are almost three months past Summer solstice.
Yesterday I got 28kwh out of the system, under almost full sun all day, there was some cloud in the latter part of the day, and I miss out on about one and a half hours of sunlight before the all panels are in full sun at 9am standard time. So 30kw may be possible on a complete no cloud day, which appears to be rare now that the panels are up!
Going by my rough calculations the large tree I have to the NE of the house is costing me about 1.5kwh of output per day, but I really like that tree...oh well...
Posted Wednesday 10 Mar 2010 @ 8:49:37 am from IP # -
Russell Moore
MemberMaximum power output according to the inverter display has been 4,629w.
Still can't seem to crack it for a 100% cloud free day though!Posted Sunday 14 Mar 2010 @ 10:24:07 pm from IP # -
Russell Moore
MemberBenny, they are fixed at my roof angle of 16 degrees, with a 8 degree azimuth offset from true north to the east. This offset was done to assist the house to gain an earlier face to the sun to assist with winter warming in the morning. At the time (28 years ago) we had no intention to add PVs.
Posted Monday 15 Mar 2010 @ 2:30:09 am from IP # -
Daniel Boon
MemberIdeally, the optimum setting would be around 24 - 28 degrees wouldn't it ?
At 16 degrees, a proportion of the radiation would be skimming / bouncing across the panel ...
Afterall the ideal solar hot water system placement is 40 degress (facing true north)Posted Tuesday 16 Mar 2010 @ 6:47:46 am from IP # -
Russell Moore
MemberMost of the newer PV panels have a tempered glass face that has been optimised to reduce reflections.
You are right, the best fixed angle for Sydney's latitude is 28 degrees, that would have meant tilt frames, with more spacing between them to prevent winter shading. It was simpler just to go with the roof angle. There is not much loss of power output from 28 to 16 degrees anyway. As to the 40 degrees for SHW, yes that is a better angle for SHW in winter, but PVs that are grid feeding, work better in summer at lower angles when there is more daylight available.
Also I have painted some vertical parts of the house to the south of the panels using one of the new reflective paints, white in this case, and the lower angles benefit from this better than higher angles.
Posted Tuesday 16 Mar 2010 @ 12:12:25 pm from IP #
