My biggest disappointment about my 200KW (commercial scale) solar system.

[mbcijim]

What is it you might ask?

They don't work in the heat.

My system went live Sept 9, 2010. In today's technology age I can check second by second how much electricity it's producing. In Pennsylvania, we've had a remarkable string of sunny, almost cloudless days going back to 3-4 days before memorial day. My thought was my system is racking up the Kilowatts!

As I checked my graphs on daily production with the guy who installed the system (and several other commercial systems), I couldn't help but notice that about 3 days of the production was down 20-40%, which I didn't understand. So of course, my installer is sitting by my side. What happened these 3 days I ask? Oh those were the days it was 90-95. Me: So? Installer They don't work so well in the heat!

My installer never told me this, and I've read a fair amount about solar panels. Up until 2 weeks ago I never read about it. It was in the local newspaper about a week previously when another local (who built a 1MW system) complained about it.

So there you have it folks. They don't shut down, but they are definitely not as efficient in the heat as they are in the cool. Sort of makes the whole 'put them in the desert thing' not such a good idea. Our memorial day weekend temperatures were by far the highest we've had since it went live 9/9.

(broken link removed to http://www.learnonline.com/pdf/The) Effect of Heat on Solar Panel Output.pdf

http://www.reuk.co.uk/Effect-of-Temperature-on-Solar-Panels.htm
"The total power loss due to the increase in temperature was from around 750 mW down to just 458 mW - a fall of almost 40%! Had the solar panel been pre-cooled in the fridge before the experiment was started the peak power output would certainly have been even higher."

 

[madrone]

Interesting. Definitely something to consider when working out your payback timeline.

Steam generation via solar concentration might be better in the desert? Obviously not for a household...

 

[jimbom]

Thanks for the information. I knew efficiency dropped off with heat, but didn't know the magnitude of the drop. If solar paid, systems would be flying off the shelf.

 

[mbcijim]

Oh, it pays, and pay it does.

But it's the government and the electric utilities paying me for my solar credits.

 

[mbcijim]

sorry on the first link, I can't seem to fix it.

 

[begreen]

The first link works. You just need to copy and paste it into the address bar because there is a space in it.

This is something I am hoping we don't run into at all. It's rare for us to get in the 80's here. And our panel design allows more free-air flow which should help it cool better.

 

[DaveH9]

I am a little surprised at the amount of your systems loss by heat, especially on a ground mount system. Roof mounts will heat up more than ground mounts because of heat radiating from the attic and from less air flow around the modules. If you compare 90 F with 20 F PV will perform much better in the cold. Trouble is the cold days are shorter! But still the production on a daily basis should be much greater, due to the hours of the day. On an hourly basis, yes, Jan beats Aug. I find it best to compare the equinoxes with the solstices, if you compare solstice with solstice on kW output your results may be disheartening.

 

[woodgeek]

Can you paint the back of the panels, in the field, with a high emissivity paint?

 

[mayhem]

Electrical effiiciency in alot of products and devices falls off inversely with a rise in temperature. I bet in full sun and in high amient air temperature, the heat inside those panels is hundreds of degrees and if there is little or no wind the air around the panels is probably heat soaked and can't mop up much of the heat generated. Can you hit the various sufaces of the panels with an IR temp sensor at different times and air temperatures and see just how hot they're getting?

Makes the gears in my head turn though...if this is as big an issue for efficientcy as it sounds like, why not cover the back of the panels with rows of heatsinks and run small electric fans to keep air moving over them? Seems awfully obvious to me so somebody else must have thought of it and done the math and found out its not viable, but it seems to me that some heat sink paste and a bunch of computer CPU style heatsinks stuck together (already mass produced and pretty cheap) and some small electric fans to move air across them (agan, already mass produced and very cheap) might lower the temp of the system enough that the increase in efficiency would offset the electrical load of running the fans.

Just a thought.

 

[peakbagger]

Anyone who designs a system has to take into account the efficiency change due to temperature to make sure that the inverter stays within the upper and lower voltage limits. Generally the panel set up is biased so that the maximum voltage which is produced during the coldest weather is just below the maximum rated voltage. Every panel spec sheet will have a set of thermal coeficient numbers for calculating the curve. Some folks have advocated misting the arrays to get evaporative cooling to drop the surface temp, but considering the string voltage is probably 400 volts, that may not be a great idea. Depdending on the state rebate program, many states pay a subsidy per namplate watt installed, so they dont care about the output. Other states pay a feed in tarifff which is depdendent upon output. I expect the installer has someone do their design and therefore the installer may not be aware.

In far northern climes where the daylight varies significantly from winter to summer, trackers can be installed which substanitally increases the summer production well over the thermal derate amount. Most of the large proposed thermal arrays in the desert are thermal systems rather than PV based so the impact due to higher temps is less. Arco Solar learned the hard way about what happens to PV cells when they are overheated. They used concentrators to send more sun to each panel and "cooked" them

(broken link removed)

 

[mbcijim]

This is something I am hoping we don't run into at all. It's rare for us to get in the 80's here. And our panel design allows more free-air flow which should help it cool better.

If you only get into the 80's you should be fine. My problems were on 90*+ days.

My system sits in the middle of a parking lot/field. Plenty of air, IF the wind is blowing.

When you stand next to/or under these things you can feel the heat radiating off of them. It's HOT. Hot enough to worry about them long term. I knew they loose efficiency each year, but we all know what happens when something sits in the sun day after day - year after year. And these are installed with the intention of getting as much sun as possible.

 

[semipro]

It seems there's a opportunity here for combining PV with solar hot water supply. Heat or preheat your DMH while cooling the PV panels. You still need DHW when the weather it hot.

 

[woodgeek]

It seems there's a opportunity here for combining PV with solar hot water supply. Heat or preheat your DMH while cooling the PV panels. You still need DHW when the weather it hot.

Actually that's why you see very few hybrid systems--the PV doesn't work when its too hot, the thermal panel only works when its very hot.

 

[begreen]

This is something I am hoping we don't run into at all. It's rare for us to get in the 80's here. And our panel design allows more free-air flow which should help it cool better.

If you only get into the 80's you should be fine. My problems were on 90*+ days.

My system sits in the middle of a parking lot/field. Plenty of air, IF the wind is blowing.

When you stand next to/or under these things you can feel the heat radiating off of them. It's HOT. Hot enough to worry about them long term. I knew they loose efficiency each year, but we all know what happens when something sits in the sun day after day - year after year. And these are installed with the intention of getting as much sun as possible.

I meant the actual design of our panels are inherently cooler. The cells are laminated between layers of tempered glass. I'll watch for output changes on a hot day and will take some backside of the panel temp readings to check this out. So far in aging tests they have passed the 60yr test which apparently is an industry first. For sure they will outlast me.

In a similar vein, there is research being done on applying a quantum dot, dye-sensitized surface layer to solar panels to capture the waste energy (heat) of the sun and convert is to usable power, similar to the way a plant does. Initial testing is promising. And there are experimental cells built with quantum dot material that show promise. If it works, it will revolutionize the solar panel industry.

http://en.wikipedia.org/wiki/Quantum_dot_solar_cell

 

[DaveH9]

... I knew they loose efficiency each year, but we all know what happens when something sits in the sun day after day - year after year. And these are installed with the intention of getting as much sun as possible.

PV modules are exposed to a lot hotter temps than what you are seeing, like in CA and AZ. They were made to with stand it. I wouldn't worry about loosing anything more than .5% a year in degradation.

 

[mbcijim]

I meant the actual design of our panels are inherently cooler. The cells are laminated between layers of tempered glass. I'll watch for output changes on a hot day and will take some backside of the panel temp readings to check this out. So far in aging tests they have passed the 60yr test which apparently is an industry first. For sure they will outlast me.

In a similar vein, there is research being done on applying a quantum dot, dye-sensitized surface layer to solar panels to capture the waste energy (heat) of the sun and convert is to usable power, similar to the way a plant does. Initial testing is promising. And there are experimental cells built with quantum dot material that show promise. If it works, it will revolutionize the solar panel industry.

http://en.wikipedia.org/wiki/Quantum_dot_solar_cell

I missed it, what do you do for a living? Are you working for one of these companies? Your info is interesting.

 

[begreen]

I work in a very unrelated field (computers) and have no relatives working for them either . Green technologies have always been a passion of mine. I like it when we humans get together and work intelligently for a better future. It gives me hope and gets me out of bed every day.

 

[mbcijim]

I work in a very unrelated field (computers) and have no relatives working for them either . Green technologies have always been a passion of mine. I like it when we humans get together and work intelligently for a better future. It gives me hope and gets me out of bed every day.

So you're installing a system on your house then? I'm referring to your comment about your cells being between two pieces of glass.

You'll be interested in the fact that I spent all day today working on numbers for a 1MW solar field. It seems construction costs are down to around $3.00/watt which equates to a 3.7% return on $.10/KW electric. It's amazing when you consider that they were 1.8% ROI just 2-3 years ago. My guess is it needs to get down to around $1.25-$1.75/KW to be financially sustainable without subsidies.

However, it doesn't appear that the price drop is due to technological advances so much as material fluctuations although some of the price drop was due to the industrialization of it. Now I could have 10 competent bidders, whereas 3 years ago the installer and manufacturer were sitting on some very, very fat margins.

The problem with getting to $1.25/KW is the racking system can easily cost $.50-$.75/KW which has nothing to do with power production.

 

[ssupercoolss]

a friend of mine installed a huge system at his shop. i think he said the best electricity making weather is 32 F and sunny.

 

[begreen]

Now that our system has been up and running I see the same thing. We are running at about 85% of capacity in full sun. Interestingly enough, when there are high cirrus clouds and the sun is on their edge, we spike up to full capacity. I think this is called the rim effect. The cloud edge scatters the light rays from a single point source (the sun). Unfortunately clouds don't stand still too often, so it doesn't last. It would be nice to optically duplicate this effect for that extra power.

 

[mbcijim]

Due to some price haggling by myself, I got the system cost down to $3.21 per watt. I can not believe how hard prices have crashed. My 200KW system last year was right around $5.50/watt. What's that work out to 41% price reduction?

 

[Adios Pantalones]

Now that our system has been up and running I see the same thing. We are running at about 85% of capacity in full sun. Interestingly enough, when there are high cirrus clouds and the sun is on their edge, we spike up to full capacity. I think this is called the rim effect. The cloud edge scatters the light rays from a single point source (the sun). Unfortunately clouds don't stand still too often, so it doesn't last. It would be nice to optically duplicate this effect for that extra power.

Ya, light travels in straight lines- most of those lines miss your panels. Any rays that can be bent or redirected have another shot at interacting with the panel. I'd guess that parabolic reflectors are one easy way to boost the light interacting with a panel.

Do they use an anti-reflective coating on the glass?

 

[pring7]

What brand collectors are you getting estimates for?

 

[woodchip]

light travels in straight lines- most of those lines miss your panels. Any rays that can be bent or redirected have another shot at interacting with the panel. I'd guess that parabolic reflectors are one easy way to boost the light interacting with a panel.

Interesting.

I had considered pv cells here, but we have an SE and an SW facing roof.

I had two thoughts, use mirrors to bend the suns rays, or alternately, have the pv panels on a turntable independant of the roof which tracks the sun somehow.

Mirrors sound simpler............

 

[pring7]

Woodchip, couldn’t you just catch early sun on one side and late day sun on the other?