# a stupid question

Discussion in 'Photovoltaics' started by jacjentait, Feb 1, 2004.

1. ### jacjentaitGuest

super conductors work at cold temps
so
would water cooling / refridgerating the back of the PV cell generate more
current ?
would backing the cell with a cold magnetic material to help electron jump ?

some simple thought logic . william james tait

2. ### Anthony MatonakGuest

....

I don't think super conductors or magnets are particularly related
to PV's and probably wouldn't help here on Earth.

PV cells do lose efficiency as their temperature rises. This means
that you can get more power out by simply cooling them. The question
is how much does this cooling equipment cost in materials, maintenance
and energy to run vs. how much extra energy it helps to produce.

I.e. If it takes a kilowatt to run the pumps to cool the panels and
you only get half a kilowatt extra out of the panels then why do it?

Anthony

3. ### Guest

I can picture something like this on the ground, made from 12 18' pieces
of Unistrut, viewed in a fixed font:

|
. |
F| At dawn, in full sun, the panels might
. r| see about 0.9x0.9x0.9x800W/m^2x16'/6' =
e| 1555 W/m^2... 6 62"x32.5" 165 W Sharp
Dynaglas . s| panels might produce 1539 watts, if
n| they are cooled by a few inches of water
. e| flowing through plastic film ducts on top.
18' l| 16'
. | The panels would dissipate about 1539/0.15
r| 10.2K Btu/h of heat. A couple of Attwood
. e| V1250 24V 35W bilge pumps could move 42 gpm
f| of pool water through the ducts, keeping the
. | panels about 10.2K/(42x8x60) = 2 F above the
duct duct | 24-hour shaded-pool wet bulb temp.
. panel panel |
-----------------|------evap pool-----
| ~6' | ~8' | ~1'
......................................................

--------------------------------------
| | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| P P | 10' wide |
|-----------------| EPDM rubber | 16'
| V V | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| | |
--------------------------------------

With a basement heat store, this might also provide some
wintertime space heating or hot water for showers.

It could have a building underneath.

Nick

4. ### Guest

I can picture something like this on the ground, made from 17 18' pieces
of Unistrut, viewed in a fixed font:

|
. |
F| At dawn, in full sun, the panels might
Dynaglas . r| see about 0.9x0.9x0.9x800W/m^2x16'/6' =
corrugated e| 1555 W/m^2... 6 62"x32.5" 165 W Sharp
polycarb. . s| panels might produce 1539 watts, if
n| they are cooled by a few inches of water
. e| flowing through plastic film ducts on top.
18'purlin l| 16'
r| -1539 = 12.3kW of net heat. Two Attwood V1250
. e| 24V 35W bilge pumps could move 42 gpm of
purlin f| water through the ducts, keeping the panels
. | about 12.3Kx3.41/(42x8x60) = 2 F above the
duct duct | 24-hour shaded-pool wet bulb temp.
. panel panel |
-----------------|------evap pool-----
| ~6' | ~8' | ~1'
......................................................

--------------------------------------
| | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| P P | 10' wide |
|-----------------| EPDM rubber | 16'
| V V | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| | |
--------------------------------------

With sun elevation theta, the panels would see approximately
800x0.9x0.9/3.28(6+16x0.9) = 1185(sin(theta)+2.4cos(theta)) = 3082 W/m^2
(3 suns max), when theta = 26.2 degrees, producing about 3051 watts
of electricity with about 83.4K Btu/h of net heat on the floor, so
the panels might be about 83.4K/(42x60x8) = 4.1 F warmer than
the 68.4 F (20.2 C) 24-hour wet bulb temp in Phila in July.

Nick

10 PI=4*ATN(1)
20 AXISELD=12'axis elevation above horizon (degrees)
30 AXISEL=PI*AXISELD/180'axis elevation above horizon (radians)
40 FL=6'focal length (feet)
50 HEIGHT=1.33'reflector height (feet)
60 N=9'number of strip reflectors
70 PRINT"strip #","bottom (ft)","top (ft)","spacer (in)","tilt (deg)"
80 PRINT
90 A=ATN(HEIGHT/FL)'initial est angle to focus (rad)
100 FOR STRIP=1 TO N'reflector strip # (from bottom)
110 TILT=(A-AXISEL)/2'reflector tilt from vertical (rad)
120 D=HEIGHT*SIN(TILT)'spacer (feet)
130 TH=TB+HEIGHT*COS(TILT)'upper strip edge height (feet)
140 AH=ATN(TH/(FL-D))'new estimate
150 IF ABS((A-AH)/A)>.001 THEN A=AH:GOTO 110'iterate to 0.1%
160 SPACER=12*D'spacer width behind top of strip (inches)
170 PRINT STRIP,TB,TH,SPACER,180*TILT/PI
180 TB=TH+D*TH/(FL-D)'next bottom edge height (feet)
190 NEXT STRIP
200 CONC=N*HEIGHT/FL'concentration ratio
220 PRINT
230 PRINT"height=";TH,"conc=";CONC,"ohb=";OHB;"%"

strip # bottom (ft) top (ft) spacer (in) tilt (deg)

1 0 1.329988 0.069425 0.2492357
2 1.331271 2.653542 1.718152 6.180075
3 2.718412 4.021652 3.185456 11.51298
4 4.207816 5.485535 4.430827 16.11825
5 5.845247 7.095126 5.455744 19.98893
6 7.676832 8.899349 6.285437 23.19267
7 9.75055 10.94771 6.952877 25.82627
8 12.11791 13.29236 7.489846 27.98812
9 14.83565 15.99017 7.923422 29.76574

height= 15.99017 conc= 1.995 ohb= 11.00475 %

5. ### Scott WillingGuest

Well, it's a long way from superconductor territory, but PV cell
voltage does droop with increasing temperature, so there is some
advantage to cooling the panels if ambient temperatures are high. A
good PV controller can take advantage of this.

Practical photovoltaics (Richard Komp) briefly describes a couple of
home-brew hybrid panel ideas for shuffling heat off the back of the
cells into air or fluid.

I think this might be more common in commercial products except that
linking the two functions together would narrow the market.

The next few years may be interesting however.

-=s

6. ### George GhioGuest

The clue to Nicks advice is the frequent use of the word "Might"

Full sun at dawn? Not likely unless the array is mounted on top of a
building with clear view of the horizon.

What shape ducts?

If they are tubular what is the effect of lens effect of the water on
energy production?

What are the losses through a couple of inches of water and two layers
of plastic?

How much energy is lost to pumping the water and moving the panels or
mirrors?

If moving the panels what about the transfer of weight of water?
What is the duty cycle of these pumps?
"Might" have several lbs of frozen water sitting on the panels.
Love this bit. A building under the basement.

I can picture something like this, Nick Pine in building under the
basement with an automatic banana feed. Free hot air for heating.

For a domestic installation the cost would exceed the gain. If it was at
all an economical proposition it would be done everywhere.

Nicks plan would have at least 3 motors running for the greater part of
day light hours. Four motors if duel axis tracking was used.

The number of moving parts of the drawn design would guarantee an
unacceptable amount of maintenance.

George

7. ### Guest

Flattish... 30" 6 mil UV poly film greenhouse air distribution ducts
(about 40 cents per linear foot) laid flat on top of the PV panels
so they are about 44" wide and 2" tall when the sun is shining, and
empty at night.
Almost none for the water itself. The air-plastic interface would have
a Fresnel loss of about 4%, with a lot less for polyethylene-water and
polyethylene-glass interfaces, with some sort of matching goo. You might
look up the refractive indices and work this out.
....70 watts max, as mentioned below. The panels and mirrors don't move.
Perhaps 8 hours per day in July. Less in winter, when more heat is needed.
Intelligent control.
You may be confused, my good man. Pick one:
1. Basement heat store with PVs on a ground rack, or
2. No basement heat store, with a building under the rack.

Nick

....something like this on the ground, made from 15 18' pieces of Unistrut,
viewed in a fixed font:

|
. |
F| At dawn, in full sun, the panels might
. r| see about 0.9x0.9x0.9x800W/m^2x16'/6' =
e| 1555 W/m^2... 6 62"x32.5" 165 W Sharp
Dynaglas . s| panels might produce 1539 watts, if
n| they are cooled by a few inches of water
. e| flowing through plastic film ducts on top.
18' l| 16'
r| -1539 = 12.3kW of net heat. Two Attwood V1250
. e| 24V 35W bilge pumps could move 42 gpm of
f| water through the ducts, keeping the panels
. | about 12.3Kx3.41/(42x8x60) = 2 F above the
duct duct | 24-hour shaded-pool wet bulb temp.
. panel panel |
-----------------|------evap pool-----
| ~6' | ~8' | ~1'
......................................................

--------------------------------------
| | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| P P | 10' wide |
|-----------------| EPDM rubber | 16'
| V V | |
| | |
| P P | |
|-----------------| |
| V V | |
| | |
| | |
--------------------------------------

With sun elevation theta, the panels would see approximately
800x0.9x0.9/3.28(6+16x0.9) = 1185(sin(theta)+2.4cos(theta)) = 3082 W/m^2
(3 suns max), when theta = 26.2 degrees, producing about 3051 watts
of electricity with about 83.4K Btu/h of net heat on the floor, so
the panels might be about 83.4K/(42x60x8) = 4.1 F warmer than
the 68.4 F (20.2 C) 24-hour wet bulb temp in Phila in July.

strip # bottom (ft) top (ft) spacer (in) tilt (deg)

1 0 1.329988 0.069425 0.2492357
2 1.331271 2.653542 1.718152 6.180075
3 2.718412 4.021652 3.185456 11.51298
4 4.207816 5.485535 4.430827 16.11825
5 5.845247 7.095126 5.455744 19.98893
6 7.676832 8.899349 6.285437 23.19267
7 9.75055 10.94771 6.952877 25.82627
8 12.11791 13.29236 7.489846 27.98812
9 14.83565 15.99017 7.923422 29.76574

height= 15.99017 conc= 1.995 ohb= 11.00475 %

8. ### George GhioGuest

Oh I am sorry I thought that the pumps ran for more than an hour.
Explain this;

At dawn, in full sun, the panels might see about
0.9x0.9x0.9x800W/m^2x16'/6'

If neither the panels or mirrors move it would perhaps be better that
the panels face south. Faceing east to see sun rise?
No that is the expected run time. What is the rated duty cycle of the
named pump? 20%? 50%?
Sorry but you said;

With a basement heat store, this might also provide some
wintertime space heating or hot water for showers.

It could have a building underneath.

All this aside what is the cost to set this up?

Oh sorry, I forgot you don't know. You have never done this. Never will
do it. If you want to sell this the materials will need to be new. It
would also have to be cheaper than adding panels to the same energy
value.

When you get to the point where you can say;

'I have built this and it supplies X amount of energy beyond panels
without cooling at a cost this much less than adding panels to the same
energy gain'

then you may be taken seriously.

George

9. ### George GhioGuest

Oh yeah, 70 watts is .56 kWh a day.

70W X 8hours.

10. ### Guest

Glass is 1.526. Water is 1.33...
Watts vs watt-hours makes it clear the pumps can return
something like 50X more electrical power than they use.
The polycarbonate cover transmission is 0.9, the mirror reflectance is
about 0.92, and I underestimated the duct transmission as 0.9. The 16/6
is the concentration ratio at dawn. How much sun bounces back out through
the cover or off the reflective endwalls?
Horizontal panels seem easy to mount and cool, with a vertical reflector
facing south.
....100%, with a \$33 price and a 3-year guarantee. They'll be in fairly clean
cool fresh water most of the time, altho they're tested in hot salty mud,
among other things.
Here's a partial materials list:

\$3000(?) 6 Sharp 165 W PV panels
2000 24 V grid-tie inverter
800 280' of Unistrut
12 32' poly film ducts
66 2 24 V 21 gpm pumps
300 pump controls, hoses, etc.
400 288 ft^2 Dynaglas
50 180 ft^2 EPDM
88 256 ft^2 1" foil-faced foamboard
20 192 ft^2 metalized Mylar film
-----
\$6736 for 3 kW peak, with useful hot water, vs

\$5800 for 800 W peak, with no hot water, for
a system with only the first three parts.

Nick

11. ### Guest

The air-plastic interface would have a Fresnel loss of about 4%, with
And low-density polyethylene is 1.51, so we have about 0.96 transmission
into the poly, 1-((1.51-1.33)/(1.51+1.33)^2 = 0.996 from poly to water and
water to poly, and almost nothing from poly to glass (compared to 0.96 with
panels in air?), so it looks like the net penalty is less than 1%.

Nick

12. ### George GhioGuest

So in 8 hours you use 560 Wh for a return of 28000 Wh
I see the panels are fixed and the mirrors are fixed. So the mirrors
will direct sun across the panels for what, An hour in the middle of the
day, Maybe a whole ninty minutes and for the rest of the day the panels
will recieve a fraction of the available energy due to poor angle.

Anyway it is all a moot point until you build it. You can't, and even if
you did it would cost at least twice what you quote and not perform to
your claims. And the pump will last for about 5 months or less. 1000
hours being about it for such a motor. You see Nick, you have no
experience. Theory is great but more often than not fails to prove true.
This is the difference between theory and fact.

So just once in your life. Do it. Prove you're right. While most people
here are doing things you are all talk. Hell, even Wayne has managed a
working system even if he can't explain it.

George

13. ### Bill ShulerGuest

Yes,
cooling a pv does cause it to harness more energy. I proved this with
a pv b4 putting it up. Using a dmm, I checked the voltage and amperage
on the warm, dry pv in the sun versus wetting it down. The power increased
when the pv was water cooled. I do not recall how much, for I did not make
note.
I am thinking of building a system with pv s mounted on a heat collector
with
a liquid medium w/i to help heat water being circulated to a heat exchanger
within a water heater tank- thereby harnessing
more power- electric and heat.

Bill

Also, there is no such thing as a stupid question.

14. ### wmbjkGuest

You might try talking about what *you're* doing, rather than criticizing
others for using a discussion group for <gasp> discussion. I see that
I guess you figured out that the more often the site gets mentioned, the
sooner it will be on the top of any web search for your name. So your
plan is to keep prodding me to mention the URL? I knew you must be doing
something productive.

Wayne

15. ### George GhioGuest

Well Wayne I do want people to look at your site. It is an interesting
design you have chosen. People can learn a lot from your site. Of course
they could learn a lot more if your site had realistic numbers that
would add up. But still and all worth the study.

16. ### George GhioGuest

Been a while so here are some pics of my work for those that have not
seen them before. Charts and circuits as well as some of my
installations around central Victoria.

George

17. ### wmbjkGuest

(updated for the second time this morning) is an excellent primer for
anyone considering sending you money for your book.

Wayne

18. ### wmbjkGuest

Invisible pictures? Excellent strategy.

Wayne

19. ### Fred B. McGalliardGuest

....
How about using an airflow to cool this. With aluminum heat fins into a
light aluminum channel to carry the air, you could arrange it to naturally
convect, then just add a bit of forced air with a fan. Well, it's an idea.
Easier to build than a water based system.

Good one!