Hybrid Car – More Fun with Less Gas

Photovoltaic cell efficiency question. - Page 2

register ::  Login Password  :: Lost Password?
Posted by Russ in San Diego on March 10, 2009, 10:03 pm
 

This comes up ALL the time.  If you search this group's archive, it
comes up at least once a month. Yes, the idea has merit, in that it
increases the panel efficiency by cooling it.  Unfortunately, you
really would prefer to run a flow rate sufficient to keep the panels
at a temperature well below what you want for domestic hot water.
This is achievable by use of a heat pump, but that's almost certainly
going to consume more power than is produced by the additional
cooling...

...so you could use it to preheat water going into your conventional
water heater, which ought to save some money.

BUT you're still left with the problem of having a liquid circulating
close to sensitive electronic equipment.  You certainly don't want to
use mains water as your coolant -- it's too mineral-laden and
corrosive.  Instead, you would want to use something similar to what
runs through an automotive radiator.  And in fact, that's how many
solar water heaters (including the one I use) work.  They are pumped
by convection.

The bottom line is that it's very questionable whether the additional
efficiency will pay for the additional complexity required.

Posted by nicksanspam on March 13, 2009, 11:06 am
 



Just trickle a bit of water over the face.


Yup. At the same time, you can also double sun power with a mirror.


Just a fact of life.


OR with a tiny water flow :-) Try numbers...


Does it ever rain on your planet? :-)

Nick


Posted by zzbunker@netscape.net on March 22, 2009, 6:19 pm
 
   Yes. but that's dramatically reduces the most attractive part of Pv
Cell Energy,
   in that you can make them automatically track the sun.
   Which is where the greatest engineering utility comes from.





Posted by frigitar on March 22, 2009, 10:31 pm
 Hello everyone,

I've just joined this group,

I'm studying RE technologies right now and I hope I can start
contributing with this.

When it comes to losing over 85% of solar energy (or as you say, 82%)
the problem is well beyond reflection or heat. I'll try to keep this
non technical.

Now the Sunlight as we see it has a mixed composition of many colors.
Some small part is uv, then say there is 30% green, 20% red, 10% blue,
some IR , so on and so forth.
The solar PV material, lets say for a crystalline silicon module, has
a certain spectral response. What this means is that it has this
"curve" which is characteristic of Silicon. In this curve,we see that
the amount of energy that the module can absorb or convert into
electricity is different for different wavelenghts. So if you took the
same amount of red light and the same amount of green light, the
electricity you get from the PV module would be different for these
two colours! Surprising is'nt it?

Hence this is where the biggest chunk of losses occur. Now the Sun
throws out lots of Green light, unfortunately, our PV panels are not
so efficient at absorbing green. They are great at absorbing UV, for
example, but then we dont get a lot of UV cause its trapped by the
ozone layer.

Apart from this there are indeed losses as mentioned due to reflection
off the surface of the cell but these are quite low as compared to the
internal losses. Also, a lot of the Sun's energy just passes through
the module because the energy is not enough to excite the PV cell to
produce electricity. This is an explanation involving quantum physics
and can be read about on the internet.

Also, I've heard that even if we did manage to absorb all of the Sun's
energy, the maximum efficiency we could get for a single junction cell
is around 35% .. its called the shockley limit or something but I'm
not sure about this one.

-Frigitar

Posted by BobG on March 23, 2009, 4:55 pm
 So if you made a PV wafer that was doped with the same stuff that
makes LEDs green, would it capture more green wavelengths, or reflect
them more?

This Thread
Bookmark this thread:
 
 
 
 
 
 
  •  
  • Subject
  • Author
  • Date
please rate this thread