Solar Cells Efficiency?
Jerry Flanders
in NC. They were unmounted, just the glass panel with the bare metal
coating on the back. I was told they were rated at about 900 mA at
about 18 volts each.
I have finally gotten around to actually connecting electrodes and
testing them out. The first one only put out about 250 - 275 mA in
bright SC sunlight yesterday (charge current into a 12 v lead-acid
battery, with an isolating diode in the circuit (same without the
diode - I checked).
I thought that perhaps the first one was bad, so last night I wired up
connectors to the remaining two and tested them out today. Same thing
- about 250-275 mA each. Right now, all three are going, with a
separate isolation diode for each, charging a total of about 800 mA
in full sunlight. I briefly shorted the three while connected
together, and measured only about 1 amp total (about 27V open
circuit).
Keep in mind that I am talking about 6 square feet of collector area,
and there is as yet no protective glass or cover of any kind. Quickie
rough calculations tell me my efficiency seems to be way down.
Any comments? What charge current do you guys get per square foot with
PV cells? They have been unused for a long time - any reason to expect
improvement in the future? Maybe I just got rejects - would rejects be
as consistent as this? Do rejects deliver low current, or low voltage,
or what?
Jerry W4UKU flan...@groupz.net
Michael Stein
>in NC. They were unmounted, just the glass panel with the bare metal
>coating on the back. I was told they were rated at about 900 mA at
>about 18 volts each.
"glass panel" implies to me ("cells" are build by thin layers on
the back side of the glass) that you have amorphous silicon cells
which are about 1/2 as efficient as crystalline cells.
I was playing with one 12" x 12" today, about 17 V open circuit
and about 200 mA short circuit (was also doing about 13+V into a
165 ohm resistor).
>I have finally gotten around to actually connecting electrodes and
>testing them out. The first one only put out about 250 - 275 mA in
>bright SC sunlight yesterday (charge current into a 12 v lead-acid
>battery, with an isolating diode in the circuit (same without the
>diode - I checked).
>together, and measured only about 1 amp total (about 27V open
>circuit).
Really 27V open circut? Wow, how many cells are in series for that?
Perhaps they aren't designed for 12 Volt output, rather 24 V?
So if you used a switching (stepdown) regulator, 24 to 27 volts
in and 13 volts out (at about double the current) you'd have
twice the power going in to your batteries (rather than just
wasting it the extra voltage...)
>Keep in mind that I am talking about 6 square feet of collector area,
sunlight is about 1000W/sq yard -> 1000W/9 sq ft 111W/sq ft. A
book says these are about 6% efficient so perhaps you should get
12W / panel. At 24 volts that would be 500 mA each. Your panels
sound close to this (all very very approx...).
Gary Coffman
>Last September, I bought three unused 12" x 24" PV panels at a hamfest
>in NC. They were unmounted, just the glass panel with the bare metal
>coating on the back. I was told they were rated at about 900 mA at
>about 18 volts each.
>
>testing them out. The first one only put out about 250 - 275 mA in
>bright SC sunlight yesterday (charge current into a 12 v lead-acid
>battery, with an isolating diode in the circuit (same without the
>diode - I checked).
>
>
>connectors to the remaining two and tested them out today. Same thing
>- about 250-275 mA each. Right now, all three are going, with a
>separate isolation diode for each, charging a total of about 800 mA
>in full sunlight. I briefly shorted the three while connected
>circuit).
>
>rough calculations tell me my efficiency seems to be way down.
>
>Any comments? What charge current do you guys get per square foot with
>PV cells? They have been unused for a long time - any reason to expect
>improvement in the future? Maybe I just got rejects - would rejects be
>as consistent as this? Do rejects deliver low current, or low voltage,
>or what?
Jerry, the current rating for solar panels is the *short circuit*
current. And the voltage rating is the *open circuit* rating, both
in full sun on a clear day with the panel oriented dead on perpendicular
to the sun's rays, and the panel held to a temperature of no more than
25C. You won't ever get all of that at the same time, so the charge current
into a battery will always be less than the panel's current rating.
The curve relating panel voltage and current is non-linear, and
there is one point where you get maximum panel *power* output,
but it's rarely the point where you hook the panel to a battery
through a isolation diode. For high efficiency charging, you
need a DC-DC converter that keeps the load on the panel at the
high efficiency point and steps up the voltage to charge the
battery at the required rate.
Your panels are obviously of an older design, and don't have
the efficiency of some of the newer panels, but the way you
are operating them may be hurting you more. The efficiency of
solar panels declines with temperature, so you need to arrange
to keep them as cool as possible for best output too. And of
course they need to be placed dead on perpendicular to the
sun's rays, a tracking mount is required for best efficiency.
And a proper charge controller is needed to load the panels
properly and deliver the most effective charging voltage to
the batteries. If you don't watch out, however, those
"improvements" can wind up costing you more power than you
gain.
The attitude of many solar power advocates is that since the
"fuel" is free, they don't care about efficiency. But as you've
discovered, it's the cost of the physical plant that limits
how economically you can obtain solar power.
Solar panels don't improve with age, in fact they degrade,
so they won't get any better, but you may be able to improve
how you are using them in order to get the most out of them.
Your panels, properly oriented in bright sun, should be
intercepting about 300 watts of solar energy. If they are
older panels, the efficiency will be less than 10%, so you
shouldn't expect more than about 30 watts out of them if
you load them properly, less if you don't. The *best*
panels get up to 30% efficiency, so that would give you
around 90 watts, but you don't even want to know what
panels like that cost.
Gary
--
Gary Coffman KE4ZV | You make it, | Due to provider problems
Destructive Testing Systems | we break it. | with previous uucp addresses
534 Shannon Way | Guaranteed! | Email to ke...@radio.org
Lawrenceville, GA 30244 | |
mkeitz
In article <1996Apr15....@ke4zv.atl.ga.us>, ga...@ke4zv.atl.ga.us (Gary Coffman) says:
>
>In article <4krn8k$h...@news1.sunbelt.net> flan...@znet.groupz.net (Jerry Flanders) writes:
>>Last September, I bought three unused 12" x 24" PV panels at a hamfest
>>in NC. They were unmounted, just the glass panel with the bare metal
>>coating on the back. I was told they were rated at about 900 mA at
>>about 18 volts each.
>>
>>I have finally gotten around to actually connecting electrodes and
>>testing them out. The first one only put out about 250 - 275 mA in
>>bright SC sunlight yesterday (charge current into a 12 v lead-acid
>>battery, with an isolating diode in the circuit (same without the
>>diode - I checked).
>>
Most likely you have amorphous panels. They can be identified by their
uniform brown color. These perform very poorly in less than absoulte
full sun, which doesn't happen often. I've always been disappointed by
them, except maybe in a solar calculator or other application that
requires only a miniscule fraction of the rated output. Trade them for
crystalline (appear blue) panels instead. The cost is about twice as
much per claimed watt, but you can expect to get close to the rating, if
exposed to a reasonable level of sun, and even some useful power in
cloudy conditions. Also the crystalline panels develop more than twice
the watts per area. I wouldn't install amorphous panels even if they
were given to me (well, maybe if I were in a low-latitude desert and could
count on full 1 KW / m^2 sun all day), so if that is what you have you
may have some trouble liquidating them.
>Jerry, the current rating for solar panels is the *short circuit*
>current. And the voltage rating is the *open circuit* rating, both
>in full sun on a clear day with the panel oriented dead on perpendicular
>to the sun's rays, and the panel held to a temperature of no more than
>25C. You won't ever get all of that at the same time, so the charge current
>into a battery will always be less than the panel's current rating.
Solar panels act pretty much as constant-current sources from shorted
up to near the rated voltage. The current decreases by maybe 5% over
most of the range. Slightly before the rated voltage is
reached, the current output starts to decrease sharply. The rated
voltage and current are the optimum operating point to draw as much
*power* from the panel under full-sun conditions. This point is slightly
down on the "knee" of the current vs. voltage curve. If you just heard
the rating by word-of mouth it may not conform to this convention of
course, but usually the rating is xxx mA @ xx V with the understanding
that this is the peak power point under nominal "full sun".
A full-blown "charge controller" automatically finds the optimum peak
power point, which is neither the maximum voltage or the maximum current.
For a small installation, it is more economic to just obtain a panel
with a rating of 15 to 18 volts (for 6-cell lead acid battery), go on
the current rating (thus slightly oversizing the wattage rating) and dump
the panel output current straight into the battery. If you are
considering more than 50 W or so of capacity, then charge controllers,
tracking mounts, etc. should be considered in their cost trade-off against
buying more or larger panels.
With a direct (through diode) connection to the batteries, no regulator
at all is required unless the battery can't tolerate overcharging, in
which case a shunt regulator is probably best since it doesn't reduce
performance at less than full charge. Yuasa sealed lead acid batteries
are designed to overcharge for moderate times (a few hundred hours) at up
to 10% of capacity but other brands or wet cells may not fare as well.
-Mike KD4QDM
PFZouave
output of your solar supply, whatever its fabrication technology, is
*tracking*. If I remember the figure correctly, a panel that is set up to
track the sun over the course of the day is supposed to produce 40% more
electricity than an equivalent south-facing fixed panel.
/=====================================================/
/ PFZo...@aol.com
/
/ The absurdity of a claim is not a valid measure of its probability
/
/=====================================================/
Rich Alloway
I know I haven't read the whole thread (and that is my fault) but I have
read a couple postings and noone has mentioned the actual efficiency of
a solar cell.
Just a few months ago, a relatively local company (somewhere in Viginia I
believe) was visited by a electrical engineer friend of mine. He has been
helping them increase the efficiency of their solar cells. After he
"meddled" with their operations a little bit they were amazed at getting
15% efficiency!!! (Yes...fifteen percent!)
Hope this helps a little,
Rich
Gary Coffman
We have to be careful when talking about solar cell efficiency to make
sure we're all talking about the same thing. Violet cells have achieved
36% efficiency in the lab, but that's the efficiency for converting the
wavelengths to which they are sensitive to electricity. That's not the
efficiency from total solar input to electricity.
On a clear day, the total solar input is 1 kW/m^2 at the surface with
the panel perpendicular to the sun. Now the solar flux approximates
a black body, and consists of a spectrum of photon energies (and hence
frequencies). A solar cell is a quantum device, and can't use photons
with less than its bandgap energy. That means that the considerable
part of the solar input which consists of IR is useless for our purpose,
and only serves to heat the solar panel, which reduces its efficiency.
And photons above the bandgap energy throw away their extra energy
above that needed to liberate the electron into lattice recoil, IE
heat. So only a part of the solar energy contained by shortwave photons
is fully utilized.
Further, the panel area is not 100% active. There is the metalization
which blocks solar input over part of the cell area, and there is the
mechanical mounting of the individual crystaline cells, which always
leaves some waste space. So we always wind up with much less than
1 kW/m^2 of electrical output from a solar panel.
In other words, we can't approach 100% efficiency very closely at all.
When most cell makers quote efficiency, they disregard the input energy
in the solar spectrum which can't bridge the bandgap, and they disregard
all area except active cell area. This makes the efficiency number look
relatively good, but when you factor in those things, the efficiency
number drops rather dramatically. Add in other factors like metalization
resistance and recombination losses, and panel efficiency drops to only
a few percent at best.
A panel that could deliver 150 watts/m^2 of electrical power (without
concentrators) would be a very remarkable achievement. But a panel that
can convert 15% of the portion of input energy at or above the bandgap
energy striking just its active area is rather mundane.
Richard Hager
>
> Keep in mind that I am talking about 6 square feet of collector area,
> and there is as yet no protective glass or cover of any kind. Quickie
> rough calculations tell me my efficiency seems to be way down.
>
> Any comments? What charge current do you guys get per square foot with
> PV cells? They have been unused for a long time - any reason to expect
>
> improvement in the future? Maybe I just got rejects - would rejects be
> as consistent as this? Do rejects deliver low current, or low voltage,
> or what?
>
Hello Jerry,
You didn't include enough real information to exactly answer your
questions, but I can get you started.
"Reject" solar cells come in many flavors. Bear in mind that calling a
low-output cell a 'reject' is like calling a 90ns RAM a reject. Most
ram's come out 70ns, but a few come out at 50ns and a few at 100ns.
There's always a statistical distribution in the production. Same with
solar cells or any other similar device. A 2:1 range in parameters is
normal.
In re: efficiency and charge current
Solar cells come in three main types, amorphous, poly-crystalline, and
single-crystal. They could also be made of two major material groups,
silicon or gallium arsenide. On top of that, they can be
single-junction or multi-junction. Normal efficiencies for all these
types will vary from 1 to 25% depending on the type.
Amorph Sil - 1-4%
Poly Sil - 3-9%
1-crys Sil - 8-15%
1-cry GaAs - 10-17%
above are typical ranges for production old enough to be surplus.
Finally, incoming energy, or 'insolation', is commonly taken to be
800-1000 watts/sq meter. Sq m = 8.45 sq ft.. At 900w/sqm = 106.5
w/sqft. If array is 12% efficient, that's about 13 watts/sqft, or 912
ma at 14.0 volts (12v batt charge voltage) per sq ft of -active- area.
Most cells produce maximum power when loaded down to a specific voltage
(varies with type and mfg.). If you've got normal single crystal
silicon, I think it's around .55v, but that's a hazy memory.
You can easily find the best point for your stuff by measuring current
while loading to various voltages, and plotting the points as a power
(I*E) curve.
Finally, power falls off if the angle to the sun isn't accurate.
Regards,
Richard Hager