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

previous 1 2

Move Thread

LAN_403

CT2

Tue Jun 09 2009, 02:55PM

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

Hey guys,

Thanks for the fast replies! Sorry for my slow reply...

I had to check with my boss to see exactly what we are going for here, as he was away when I first posted.

We basically want to see how much "scattered" or non perpendicular light contributes to power output of the panel. So say on cloudy or smoggy days we can get an estimate of how much power weâ€™ll put out versus a clear day where more direct light will fall upon the panel. The results donâ€™t have to be exact; we are just trying to get an idea of how much power weâ€™ll loose.

Steve that is exactly what we are looking for! Itâ€™s just too bad it cost six thousand dollarsâ€¦ But I may have found something which will give us good enough results for a LOT cheaper, at least that is what Iâ€™m hoping for.

My idea is to use these light sensors

they are light to voltage sensors, with a linear output of voltage to uW/cm^2 of power, hey perfect eh? And they just so happen to have about the perfect range of light sensitivity (although it falls off near the edges, 300 and 1100 nm). Iâ€™ll have one sensor in a black tube for the perpendicular â€œdirectâ€ light measurement, a 10k resistor on the output, and then Iâ€™ll measure the voltage. For the indirect Iâ€™ll use 5 of the sensors, one point straight up and the other 4 in a circle around the central one, but at an angle of say maybe 70 degrees to the central one. Iâ€™ll take the average of all these sensors to get the direct + indirect light. This is because the output drops substantially at angles nearing 90 and -90 degrees. Then just subtract the direct from the direct + indirect and we are done.

This will give me V/( W/cm^2) and with a linear relationship. This should get me in the ballpark for estimating the power we can get from the panel during different weather conditions. What do you guys think?

Once again, thanks for all the replies and ideas!

Chris.

Steve Conner

Tue Jun 09 2009, 03:36PM

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

You still don't seem to grasp the idea that direct sunlight is not the same thing as light perpendicular to the panels. Unless the array is mounted on a sun tracker, the results from the black tube method will be useless, because the panels have a cosine law response but the black tube doesn't.

CT2

Tue Jun 09 2009, 04:15PM

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

I understand that "direct" can mean different things, directly from the sun would be all the light the panel will ever see. But in our case we want to get the difference between light rays that have made their way down to the surface of the earth without bouncing of particles in the air, such as smog, rain, things like that (this is what we will consider "direct"). The black tube test isn't over a period of time, we will calculate the solar altitude angle and set up the tube at the same angle so that as the sun is shining down only light that hasn't been reflected, bounced, refracted or anything like that will make it to the sensor. We don't need to track the sun, we can just take the measurement, and if we need more data points we can manually follow the sun across the sky.

Does this not make sense? Or am I still missing something important?

As for the array, it hasn't been decided what angle it will be on, we do have a sun tracker, and a stationary array for research though. But this isn't for that. I'm thinking if we know the amount of perpendicular light versus light at another angle, for different times of day in different conditions then we can get an estimate of how much power the panel will supply in those conditions.

For example, its a hot smoggy day in toronto, because of this we get say 500W "direct" and 500W "indirect" per square meter. Now we can say that the power out of the panel will be 500W * efficiency + 500W * the cosine response of the panel to angles other then perpendicular * efficiency. That would give us a rough estimate of the power our panel will give us... atleast I think?

Steve Conner

Tue Jun 09 2009, 04:41PM

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

Oh, I was just assuming that you'd be logging data continuously for a year, or something. That's what you need to do to get the real world efficiency.

You could just do it at midday one day in summer and aim your collimator manually, I guess, but you'd need to somehow extrapolate the effects of other times of day and times of year, which are complicated. For instance, the spectrum of solar radiation depends on the sun's angle above the horizon (because of atmospheric absorption) and the efficiency of PVs depends on the spectrum and also the temperature, which in turn depends on the time of day and the season. We had software that would do that for us.

Might be worth writing to the guys at my old research group:

They ate problems like this for breakfast, and I heard that at least one of them moved to a university in Canada.

twist2b

Tue Jun 09 2009, 10:45PM

Registered Member #2086 Joined: Tue Apr 21 2009, 02:33AM
Location:
Posts: 117

Steve is right, you need REAL LIFE data. You should study a years worth of data IMO... or find someone who has recent data. The changing seasons effect the suns position from the earth.

BTW - fun project, I always have found solar panels to be a neat method for energy.

previous 1 2

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