Water cooled solar panels for significant output boost

[jflorey2]

Maybe, maybe not. More than likely possible, but not practical or cost effective.

Agreed. Given how cheap both PV and solar thermal panels are lately, there's no advantage to combining the two (outside of some very contrived situations.)

[J.P.M.]

From a previous post from someone who actually built one:
"Taking into account all pump usage, I'm getting a 4% (daily) boost."

One experiment is worth a thousand expert opinions.

Unless the experiment is not well done. How was the 4 % determined ? Kind of tough without a pyranometer.

[shocksofmighty]

Unless the experiment is not well done. How was the 4 % determined ? Kind of tough without a pyranometer.

See the original description. It was a comparison to two control panels that were not cooled, then extrapolated to the system as a whole.

I've slowly been modifying my design to go from a pop-up sprinkler to individual micro sprinkler heads for each column of panels. Once it is back in action I'll provide overall production and individual panel outputs for those that want to look at it.

Again, I'm just having fun ( and sunking is still wrong).

[J.P.M.]

See the original description. It was a comparison to two control panels that were not cooled, then extrapolated to the system as a whole.

I've slowly been modifying my design to go from a pop-up sprinkler to individual micro sprinkler heads for each column of panels. Once it is back in action I'll provide overall production and individual panel outputs for those that want to look at it.

Again, I'm just having fun ( and sunking is still wrong).

Is that the one described in your 05/26/2016, 1:43 P.M. post ?

[shocksofmighty]

Is that the one described in your 05/26/2016, 1:43 P.M. post ?

Yes.

[J.P.M.]

Yes.

Thank you. Without some way to monitor the input (from what's called a pyranometer) and the output, and what happens in between you will not get more than anecdotal information. Do not be under the mistaken impression that for example, solar irradiation is the same and unchanging, even over periods of a few minutes, in spite of how "sunny" it may look, or that, in the case of water cooling as you seem to be looking at, overlooking or knowing how to account for the atmospheric conditions of wind vector and, probably at least as important, atmospheric dew point to name but a few.

Here's the deal: Do as you wish and have a good time and learn stuff. I did much the same as you seem to be doing ~ 40 yrs. ago, but not w/ PV until later. You will learn a lot, but until you understand some of the fundamentals, you'll be less than efficient at the details of why you may be spinning your wheels. All the same to me. I was once where you now find yourself. I had a good time, but acquiring an ME degree was an eye opener to my ignorant hubris.

Once you get the basics of the engineering under your belt and/or into your mind, it will be easier for you to understand that others more experienced and learned in solar energy engineering and the thermal and fluid sciences have investigated precisely what you are on about, and have found it to be similar in principle to what are called economizer schemes, or energy scavenging or other terms. The idea is sound, established and mostly a no brainer from an engineering perspective. However, the devil is not only in the details, but the economics, not only because of the low thermodynamic availability of the recovered heat, but also the relative amount and type of equipment needed to make such recovery systems viable, workable, serviceable and most of all safe and reliable.

Right now, such things as what you are working on are probably more suited to the investigations of undergraduate students. If such systems were economically viable and practical, the proof would lie in a lot of PV solar farms that employed heat recovery systems. I don't think there are too many.

No rancor/ragging here - Good luck. Just know that your chances of discovering something new will get a lot better with more education as will your chances of most likely seeing you are reinventing the wheel. Just be careful you're ignorance doesn't harm you or someone else.

[foo1bar]

Thank you. Without some way to monitor the input (from what's called a pyranometer) and the output, and what happens in between you will not get more than anecdotal information.

I don't think that a pyranometer is really necessary.
Having one of the modules acting as a control is sufficient to provide clear data showing the benefit.

I think even with a pyranometer it'd still be anecdotal information - it would still be a one-off custom-built setup that one person is using in one climate/location.

Do not be under the mistaken impression that for example, solar irradiation is the same and unchanging, even over periods of a few minutes,

If the control module is right next to the modules being tested they'll experience the same solar irradiation, wind, etc.
And it should be easy to turn off the watering system at various times of the year and confirm that the control module consistently tracks with the other modules that are being experimented on.

The idea is sound, established and mostly a no brainer from an engineering perspective. However, the devil is not only in the details, but the economics, not only because of the low thermodynamic availability of the recovered heat, but also the relative amount and type of equipment needed to make such recovery systems viable, workable, serviceable and most of all safe and reliable.

I think the way he's approaching it makes it potentially economical.
He's using rainwater (no cost and should be very low dissolved minerals)
And it sounds like the material costs are pretty low (some cheap irrigation parts)
Since it's an open-loop system, I don't believe there is any recovered heat.

If such systems were economically viable and practical, the proof would lie in a lot of PV solar farms that employed heat recovery systems. I don't think there are too many.

It's not a heat recovery system that he's doing.
It's "use a little pump to intermittently spray some water from the rain barrel on the panels and see if gains from cooler modules outweigh electrical cost of pump and controller"

I would guess once you scale up and you pay for labor to install (and water) it becomes non-economical. Even just the cost of materials may make it non-economical. The additional production has to outweigh the sum of 1> additional labor to install 2> cost of low-mineral water 3> additional maintenance 4> additional materials and 5> cost to pump the water. (Oh - and also be cheaper than just adding N% more modules)

Oh - and the additional production should be over the life of the array - which if lifespan is shortened by the sprayed water could be a negative number. (How is it being sprayed, where is it being sprayed, how much additional minerals are deposited on top, etc? Easy to do something that shortens the life or causes additional maintenance)

[ButchDeal]

what are the tolerances of the modules being used? most come from the factory with a tolerance rating of something like -0+5w.
a 5w difference on a control module is quite a bit.

[foo1bar]

The performance difference between the control module and the experimental modules does need to be accounted for. But the performance difference will track, so it'll just be "OK, with water cooling I see +4.3% and without water cooling I see +0.2%, so I'm seeing a 4.1% increase."

[shocksofmighty]

No rancor/ragging here - Good luck. Just know that your chances of discovering something new will get a lot better with more education as will your chances of most likely seeing you are reinventing the wheel. Just be careful you're ignorance doesn't harm you or someone else.

I don't sense rancor, just condecension. Sadly that seems to be an all too common trait for engineers. As a scientist (Ph.D and all that) I enjoy performing experiments and getting empirical data. That's not to say that careful consideration and foresight are not warranted. I did a similar calculation to what you did on the previous page (though I would propose you don't need a 6 GPM pump at 80W to get a 5 C temperature decrease; I get it with a 4.2 L/min pump running at 35 W, which includes a solenoid valve to prevent siphoning). It seemed reasonable that an efficiency increase could be managed, and the question became the best way to approach that given the restrictions on my PV system, my access to materials, and my capabilities. My goal was never economic viability, and I wasn't out to discover something new, just solving a little problem I decided to explore.

I agree with foo1bar, control panels are a perfect acceptable way to ascertain if there is an improvement. I'll take some screen caps shortly for a case where I started testing microsprinklers. In this case I have 8 panels getting sprayed. The per panel output is clear, relative to all the other control panels that aren't getting sprayed.

[shocksofmighty]

Keep in mind, I only have 8 panels being sprayed in this case, at 5 minutes on, 5 minutes off when total array output is over 5kW. All the rest are "controls". First, here is showing when spray occurs (blue bars). Pink is when the secondary tank gets filled. cooling_overall.png

Now as per panel, you can clearly see which are being sprayed and which are not:

cooling_panels.png

And here are similar days when there is no spraying taking place (because I was out of rain water!).
no_cooling_overall.png

And the per panel view without cooling:

no_cooling_panels.png

[Sunking]

I see this SCAM has come up again.

[Mike90250]

Not so much a scam, as the cooler panels always produce more, but is the expense of the water, pump system and controller and shorter life of the PV panels (due to thermal cycles, hot cold hot cold) effective, vs adding another panel to the array to compensate for the reduction ?

[SunEagle]

Lets spend $1,000,000 to turn $500,000 of lead into gold. Makes sense to me.

[shocksofmighty]

Not so much a scam, as the cooler panels always produce more, but is the expense of the water, pump system and controller and shorter life of the PV panels (due to thermal cycles, hot cold hot cold) effective, vs adding another panel to the array to compensate for the reduction ?

I was worried about thermal cycles as well, but what I can tell from the literature, the type of temperature changes the cooling system would induce during constant operation would actually have little impact. Also operating at lower overall temperatures during the hottest portion of a day would actually decrease thermal degradation. If thermal cycling over normal operating temperature range (not the -45 to 80 C used in testing) was a huge issue, partly cloudy days would also kill your modules.

[J.P.M.]

I don't sense rancor, just condecension. Sadly that seems to be an all too common trait for engineers. As a scientist (Ph.D and all that) I enjoy performing experiments and getting empirical data. That's not to say that careful consideration and foresight are not warranted. I did a similar calculation to what you did on the previous page (though I would propose you don't need a 6 GPM pump at 80W to get a 5 C temperature decrease; I get it with a 4.2 L/min pump running at 35 W, which includes a solenoid valve to prevent siphoning). It seemed reasonable that an efficiency increase could be managed, and the question became the best way to approach that given the restrictions on my PV system, my access to materials, and my capabilities. My goal was never economic viability, and I wasn't out to discover something new, just solving a little problem I decided to explore.

I agree with foo1bar, control panels are a perfect acceptable way to ascertain if there is an improvement. I'll take some screen caps shortly for a case where I started testing microsprinklers. In this case I have 8 panels getting sprayed. The per panel output is clear, relative to all the other control panels that aren't getting sprayed.

Opinions vary. I was sort of formally educated in one of the sciences (Physics) back in the '60's before getting reeducated as a ME, in the '80's. Having been around both types, scientists and engineers, both academically and socially, my experience gives me the opinion that there may be what appears condescension among both groups, but the engineers usually (but by no means always) have more practical experience behind their opinions. Also, in these types of situations, my observation is that the accusation is usually either condescension or talking down to someone, either one of which, in the spirit of "take what you want and scrap the rest", says more to me about the accuser than the accused, but that's just opinion.

My #'s' as you describe from a previous page were meant as examples of a reasonable setup, not a mandate, nor were they of a working system. But, the #'s are based on best/reasonable estimates and somewhat lengthy experience with a lot of heat transfer equipment.

I was somewhat in error about how your setup was arranged. With my apologies, my error and sloppy reading. With that, I suppose a pyranometer is probably not absolutely essential for comparisons. I would however, retain my belief that there is more going on than you are either not aware of, or not considering for other reasons that will influence your results and how you interpret those results. In the interest of having fun and discovery for it's own enjoyment, I say Amen - enjoy.

It looks like we may at least agree somewhat on the idea that aux. cooling of a solar PV array is from an academic standpoint possible, but from a practical standpoint and also cost effective standpoint, at least at this time, more than likely less than cost effective for several reasons.

[Mike90250]

how about mounting bi-facial panels upside down in the shade ?

[J.P.M.]

how about mounting bi-facial panels upside down in the shade ?

I like it. A real entropy increaser for no benefit, theoretical or practical. A real consensus builder that.

[ButchDeal]

it would likely increase module output consistency.