Solar - Direct vs. Indirect Light / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Solar - Direct vs. Indirect Light

1 2 next

Move Thread

LAN_403

CT2

Tue Jun 02 2009, 04:19PM

Registered Member #180 Joined: Thu Feb 16 2006, 02:12AM
Location: Ontario, Canada
Posts: 187

Hey guys,

I'm currently working for a company as a co-op student (no thanks to my school...); we are working on installing solar systems, and the research that goes along with that. It's a really cool job and I get to design a lot of things and do some cool research and testing. I was hoping I could get some help from all the knowledgeable people on here.

Hereâ€™s the problem Iâ€™m having, we would like to be able to measure the amount of direct light (perpendicular to the panel) versus indirect light (reflected off atmospheric particles or nearby objects, at an angle other then 90 degrees to the panel). Sounds simple enough, my idea was to use two solar sensors, one would be open to all sources of light, and the second would be within a small black tube, that way only light coming straight would make it to the bottom and hit the sensor.

Now the more I thought about it the more variables seemed to start popping up. The panels that we would install on a roof are covered in a special glass, this helps capture light, so the sensors we use would need to have the same glass covering if we want the results to be comparable. Also the type of material the actual cells are made of would need to be the same (Monocrystalline silicon most likely)

. On top of this the absorption spectrums of the sensors and panels would need to be the same as well (which is a hard figure to find, not many datasheets specify the range). And then the more I looked into it nothing seems to be linear, the voltage output per amount of light on the panel isnâ€™t a linear relationship, and it doesnâ€™t seem to be that the current output is either.

So what seemed like a simple test is actually turning out to be way more complex then I had originally thought. I mean I could just use two tiny solar cells

an we could get some results, however I donâ€™t think they would be accurate or even usable due to all the variables and the non linearity of everything. Maybe a cadmium sulfide resistor would work, but then it doesnâ€™t really tell you how the actual panel will respond to indirect vs. direct, it will only show you the different of how much light is direct and how much is indirect. My last thought would be to have the company that makes the panel to send us just one single cell (well two, but not connected) covered in the glass as it would be in the module, and then use those to do the measurements.

Any ideas or thoughts would be greatly appreciated, because the more I think about it the more complex and immeasurable it seems.

Thanks, Chris.

Electroholic

Tue Jun 02 2009, 05:06PM

Registered Member #191 Joined: Fri Feb 17 2006, 02:01AM
Location: Esbjerg Denmark
Posts: 720

yea and you must keep them at the same temperature, because you loose 0.5% efficiency for every degree C temperature rise.
To get power in, the best way is probably look at the actual output maximum power point. If you don't have an active tracker, 80% of OCV should get you close enough.
Also, so long as the spectrum of direct light and indirect is the same, and you find the relative difference, it should be good enough.

So how accurate do you want it to be?

Proud Mary

Tue Jun 02 2009, 05:28PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

How about covering the 'indirect' panel with a tent of crumpled cling film, which will attenuate the light very little, but wshould simulate indirect lighting fairly well.

You could do a similar experiment with packer's bubble film, and then create some kind of median between that and the cling film tent.

Or you could maybe deposit/evaporate some harmless colourless crystalline substance onto the 'indirect' panel simulation - sugar perhaps - to create a diffusing surface. Or try that kind of frosted glass people have in toilet windows, and then allow some kind of attenuation factor for losses in the extra sheet of glass.

Just a few ideas that might help.

HV Enthusiast

Tue Jun 02 2009, 05:35PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

You need a shadow radiometer.

MinorityCarrier

Tue Jun 02 2009, 07:55PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

For what application do you need to make this comparison (if it is not confidential)?

MOTVolt

Tue Jun 02 2009, 10:58PM

Registered Member #1996 Joined: Wed Feb 25 2009, 03:56AM
Location: Blackburn South, Melbourne, Australia
Posts: 10

If you just wanted a simple percentage comparison, you could just use two identical LDR's. One open to the all light, the other in a dark tube as you suggested, then simply subtract the tubed LDR voltage from the open LDR voltage. That will give you the voltage of the ambient light by itself.

coillah

Wed Jun 03 2009, 02:06AM

Registered Member #1517 Joined: Wed Jun 04 2008, 06:55AM
Location: Chico CA
Posts: 304

Electroholic wrote ...

yea and you must keep them at the same temperature, because you loose 0.5% efficiency for every degree C temperature rise.
To get power in, the best way is probably look at the actual output maximum power point. If you don't have an active tracker, 80% of OCV should get you close enough.
Also, so long as the spectrum of direct light and indirect is the same, and you find the relative difference, it should be good enough.

So how accurate do you want it to be?

If you can find the exact relationship between efficiency and temp, you do not need to maintain the temp, just record it.

Oh yeah, and what kind of accuracy are you shooting for?

...

Wed Jun 03 2009, 02:43AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

A lot of the non-linearities of your proposed plans should end up canceling themselvs out. For example, the difference in absorption spectra between the solar cell and your device will not matter much, since sunlight is a more or less constant spectra--so there will just be some constant calibration factor you can easily measure. Simarily, the power output from a solar cell (which is what you should be measuring, since that is what you are trying to predict) is fairly linear with light intensity, and again it would be trivial to take a few data points with a real panel vs your tester, and if it turns out the difference in linearity between the real cells and your tester is too great you could always do a curve fit and make a calibration curve to use.

The light angle test is somewhat trickier, and without a piece of the special glass to test with your best bet would probably to use a cone instead of a tube to try to mimic the larger acceptance angle of the panel.

Steve Conner

Wed Jun 03 2009, 08:56AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I did a lot of this stuff as part of my PhD. I worked on the installation of three PV systems, two of which were instrumented for research purposes.

One instrument we used was the pyranometer:

The cosine law response of this instrument takes care of all your direct vs. indirect light worries. The efficiency of a solar panel is measured by comparing the watts per square meter that it produces (number of watts of electricity that come out, divided by its area in square meters) to the irradiation (also in watts per square meter) measured by a pyranometer sitting next to it.

Your reasoning doesn't really make sense anyway: "direct light" should surely be light that comes directly from the sun, therefore your collimating tube would not measure direct light unless it tracked the sun. You can get devices like that, they're called pyrheliometers, and intended for mounting on PV arrays that track the sun. But for non-tracking arrays we just used a pyranometer aligned with the plane of the modules. Or left it pointing straight up (since that's the ISO standard way of measuring irradiation) and applied a correction factor.

Another way of doing it is to use two pyranometers, one with a shade ring in the shape of the sun's track (available as an accessory) that blocks direct sunlight. The unshaded one measures direct+indirect (aka, "global irradiation") the shaded one measures indirect only, so you can calculate the direct component from that. The shade ring needs alignment, and readjusting as the sun's height changes with the seasons.

An improved version of this is available in one unit from Delta-T:

We also had Pt100 precision temperature sensors stuck to the backs of some modules, and voltage/current sensors, the whole lot hooked up to a portable datalogger.

Sulaiman

Wed Jun 03 2009, 11:04AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

A simple solution may be to measure the output normally
then shade the module with an opaque object such as a piece of aluminium sheet
hold the sheet far from the pv array so that direct sun is blocked but not skylight,
this should give just skylight energy.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob