array temperature question

A place to post images of your OutBack Power System

Moderators: OutBack, OutBack Moderator Team

Post Reply
User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

array temperature question

Post by Heimhenge » Sun May 17, 2020 2:07 pm

The CSUN panels in my array have a temperature coefficient of -0.42% / °C (according to the spec sheet). This is the power coefficient, which I believe is the proper one to use for de-rating array output. I'm interested in how they actually perform because I'm in Arizona. I get how to do the actual calculation for comparison but have a question about array temperature.

Obviously the air temperature is meaningless. I need to measure the temperature of the panel itself which will usually be way above air temperature. So I just ordered a remote IR thermometer (not surprisingly back-ordered right now) with enough range to be useful in this application. If you're curious, it's this one:

http://www.extech.com/products/42510A

Here's my question: To measure the temperature of the panels should I measure the front side of the array where the cells face the Sun? Seems like I might get excess reflected IR if I do that. Or should I measure the back side of the array, which looks like this:
new-array-under.jpg
I'm sure the back will be cooler than the front, but don't know if it's more representative of the cell temperatures. When the thermometer arrives I'll do some tests both ways and see which better reflects the measured array output compared to rated output and post the results. Just looking for some advice here in case anyone else has done this. Thanks.

provo
Forum Guru
Posts: 391
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Eight Rolls S-550 (2 strings, total ~800Ah @ 24V)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC and Trimetric
Location: Sierra foothills

Re: array temperature question

Post by provo » Sun May 17, 2020 2:26 pm

Heimhenge wrote:
Sun May 17, 2020 2:07 pm

Here's my question: To measure the temperature of the panels should I measure the front side of the array where the cells face the Sun? Seems like I might get excess reflected IR if I do that. Or should I measure the back side of the array....
With the glass and the underlying silicon, the emissivity could be weird ... for tricky surfaces (shiny copper is another one) I like to put a 1" square of masking tape down, and keep the laser dots on that when I take a measurement. I'd do that on the top side rather than the underside....

User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

Re: array temperature question

Post by Heimhenge » Sun May 17, 2020 3:02 pm

Great idea. Shouldn't take long for the masking tape to reach the same temperature as the glass. Would it make much difference what color tape I use? I mean, as long as it isn't aluminized or something like that.

provo
Forum Guru
Posts: 391
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Eight Rolls S-550 (2 strings, total ~800Ah @ 24V)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC and Trimetric
Location: Sierra foothills

Re: array temperature question

Post by provo » Sun May 17, 2020 3:14 pm

Heimhenge wrote:
Sun May 17, 2020 3:02 pm
Would it make much difference what color tape I use? I mean, as long as it isn't aluminized or something like that.
I think the temp. of the glass under it won't be very affected by the tape -- tan might be better than the dark blue stuff :grin: . Do some experiments and let us know!

User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

Re: array temperature question

Post by Heimhenge » Sun May 17, 2020 3:34 pm

provo wrote:
Sun May 17, 2020 3:14 pm
Heimhenge wrote:
Sun May 17, 2020 3:02 pm
Would it make much difference what color tape I use? I mean, as long as it isn't aluminized or something like that.
I think the temp. of the glass under it won't be very affected by the tape -- tan might be better than the dark blue stuff :grin: . Do some experiments and let us know!
Count on it. I have the tan tape (which I'll use). Also have some black and blue and white, which I'll run for comparison.

provo
Forum Guru
Posts: 391
Joined: Thu Jul 27, 2006 4:34 pm
My RE system: Sixteen Evergreen EC-120
(4 strings, total 1920W)
Eight Rolls S-550 (2 strings, total ~800Ah @ 24V)
One FM80
One VFXR3524A
Hub 10
Mate3s
FNDC and Trimetric
Location: Sierra foothills

Re: array temperature question

Post by provo » Sun May 17, 2020 3:42 pm

Heimhenge wrote:
Sun May 17, 2020 3:34 pm

Count on it. I have the tan tape (which I'll use). Also have some black and blue and white, which I'll run for comparison.
Said like a true scientist!

pss
Forum Czar
Posts: 500
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 8330 watts in three strings, Flexmax 60 x 3, Radian 8048A, GSLC load center, Mate 3S, Hub 10.3, FN-DC and 900 Amp, 48V Trojan T105-RE battery bank.

Re: array temperature question

Post by pss » Mon May 18, 2020 8:33 am

Whoa, whoa, whoa. I've already posted significant info on panel temperatures and power output. Search the forum.
Forget tape.
Know the ambient temperature, use a laser thermometer and point it at the panel surface, then the nearby ground. You will see how much hotter the panel surface becomes. I'm is So Cal. I have recorded panel surfaces close to 150 degrees F.

Panel output is significantly degraded in these conditions and there is really no way to cool them safely and efficiently in extreme heat conditions. The possibility of putting a circulating cooling tube system on the back side like an ice rink uses with a compressor will use up all your PV benefit.

Heat kills panels just when your AC systems are working their hardest.

User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

Re: array temperature question

Post by Heimhenge » Mon May 18, 2020 9:02 am

pss wrote:
Mon May 18, 2020 8:33 am
Whoa, whoa, whoa. I've already posted significant info on panel temperatures and power output. Search the forum.
Forget tape.
Know the ambient temperature, use a laser thermometer and point it at the panel surface, then the nearby ground. You will see how much hotter the panel surface becomes. I'm is So Cal. I have recorded panel surfaces close to 150 degrees F.

Panel output is significantly degraded in these conditions and there is really no way to cool them safely and efficiently in extreme heat conditions. The possibility of putting a circulating cooling tube system on the back side like an ice rink uses with a compressor will use up all your PV benefit.

Heat kills panels just when your AC systems are working their hardest.
I did search for "panel temperature" and got this:
search.php?keywords=panel+temperature
... none of which answered my question.

I know the panels will be hotter than air temperature. The "nearby ground" is a sealed/white painted wood roof. I'm not interested in adding cooling, but you can see from the photo that the backs of the panels are exposed to great airflow. Much more so than panels mounted flat on a tilted roof.

All I'm interested in doing is comparing the calculated de-rated output to the actual output. I need an accurate measurement of cell temperature to do that.

So you are saying that just pointing the IR thermometer at the business side of a panel gives an accurate reading? Is there no additional IR being reflected from the panels? I need the IR emitted by the panels alone, which is why I speculated about measuring the backside of the panel.

fcwlp
Forum Guru
Posts: 345
Joined: Wed Jan 22, 2020 9:40 am
My RE system: GS8048A, FM80 w/3,600 W PV Fixed, FM80 w/2,700W on Zomeworks tracker, Mate3, 24 Trojan 2V L16 1100AH @ C20, Grid-Tied with Kohler 14RESA LPG Generator and MEP-803 Diesel if needed
I also install and maintain grid-tied and off-grid systems, details will be given for these system if required
Location: 80 miles NE of Phoenix at 5500'

Re: array temperature question

Post by fcwlp » Mon May 18, 2020 9:23 am

Heimhenge wrote:
Mon May 18, 2020 9:02 am
I know the panels will be hotter than air temperature. The "nearby ground" is a sealed/white painted wood roof. I'm not interested in adding cooling, but you can see from the photo that the backs of the panels are exposed to great airflow. Much more so than panels mounted flat on a tilted roof.

All I'm interested in doing is comparing the calculated de-rated output to the actual output. I need an accurate measurement of cell temperature to do that.
I would point it at the back side EVA polymer. As this is a relatively thin polymer compared to the topside glass, you will get a better temperature of the silicon. You can try the tape, but my hypothesis is that it won't make much of a difference as compared to the EVA.

It is amazing how much cooler a white low emissivity roof is in AZ as compared to a black roof. At our previous house in Scottsdale we had a black roof on the garage that could barely be stood on in the summer and after coating with the white low emissivity coating it was almost cool.

User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

Re: array temperature question

Post by Heimhenge » Mon May 18, 2020 9:57 am

For sure. White is the only sensible color for a roof in Arizona. Here's what mine looks like:
foam.jpg
That's two inches of poly foam with an elastomeric sealant. You can walk on it just fine. As long as you ain't wearing cleats. :)

pss
Forum Czar
Posts: 500
Joined: Mon Jul 24, 2017 8:40 am
My RE system: 8330 watts in three strings, Flexmax 60 x 3, Radian 8048A, GSLC load center, Mate 3S, Hub 10.3, FN-DC and 900 Amp, 48V Trojan T105-RE battery bank.

Re: array temperature question

Post by pss » Mon May 18, 2020 1:27 pm

viewtopic.php?f=15&t=14204

viewtopic.php?f=1&t=13243

viewtopic.php?f=17&t=9055

My arrays are ground mounted and the panels have anywhere from 2 feet to 8 feet of space underneath them. I have measured temperatures on the panel surface at 1 PM on the hottest days as well as ground temperatures in the sunlight and shade. Quite interestingly I have found that the hottest surface temperatures I can record are on my turf in out back yard. Hot enough to burn a dog's paw.

I have learned about the degradation of my system due to heat and so far I don't see anyway around it. Water is SoCal cannot be used to cool the panels, not even mist them. The air is so dry it would evaporate right away.

LAST QUESTION: Why do people drive around in black cars in Arizona? In So Cal and even in Vegas, may roofs are not white because heavy insulation in the construction neutralizes the benefit of a white roof. In your case, the roof underneath the solar panels is in the shade because of the solar panels, so maybe try measuring the temperature of a piece of flat black painted wood on the shite roof under the solar panels and see how much warmer that is when shaded by the panels. Maybe white not much help under a panel.

User avatar
Heimhenge
Forum Guru
Posts: 119
Joined: Sun Jan 26, 2020 10:28 am
My RE system: 9 CSUN 390W panels for 3.51 kW
UniStrut PV rack on a flat roof
Midnight Solar MNPV3 combiner box w/ breakers
Outback Radian GS4048A-01 inverter
Outback FM100 charge controller
Outback FLEXnet DC monitor
Outback MATE3s system controller
Outback Hub
Trace DC breaker and shunt box
Lifeline GPL-8DL x4 = 12.2 kWh @ 48 volts
Running OpticsRE since March 2020
Location: 30 miles north of Phoenix @ elevation 2200 ft

Re: array temperature question

Post by Heimhenge » Mon May 18, 2020 3:50 pm

Those were all informative links, thanks. Curious they didn't show up in my search.

Regarding cooling the panels ... I wonder why they don't use a back panel colored high emissivity black? Sure, you'll get some IR from the ground during the day, but if the panel temperature is higher than the ground (white roof in my case) temperature, the net is radiative loss from the panels. But I think if I painted the backs of the panels black it would void the warranty. :)

No idea why people here in AZ drive dark colored cars. They pay for it in AC load on the mileage.

Post Reply