Traditional generators exert more rotational resistance when drawing current. I once saw a video (sorry, no link) of a guy who had set up a long plastic screw thing enclosed in a pipe, fed by water from a river, connected to a generator, connected to a battery. When the switch was off (when no current was being drawn) the screw turned quite quickly as the water flowed through. As soon as the switch was turned on, and current was being drawn, the screw turned much more slowly.
This must have something to do with current inducing a magnetic field in the coil which resists moving past the permanent magnets (I don’t know, I’m not an electrician). But it got me wondering: does drawing current from a solar panel cause any visible changes to the solar panel, compared to when no current is drawn?
Sent from my iPhone using Tapatalk
I can’t think of any mechanism that would cause the appearance to change. It’s just a piece of silicon semiconductor. It absorbs a certain percentage of light, creating electron/hole pairs. I’ve never heard that the absorption rate would depend on whether the electrons/holes are flowing out of the wire or not. (And I think I would have heard of it if it happened, because that would imply that applying bias voltage on a silicon detector would change its efficiency - a very useful thing in my line of work. Of course bias voltage would affect the charge collection efficiency, and a high enough voltage might amplify the charge, but I’ve never heard that the charge creation rate would change.)
However, I have been told that if you apply power TO the solar cell, it would glow. I haven’t found confirmation on it, but I do know LEDs work as light sensors, so at least some semiconductor light emitters/receivers work both ways.
Back to the OP, solid state isn’t my field, but I can say some general things. The same amount of energy is impingent on the panels, whether they’re attached to a load or not. When they’re attached to a load, some of that energy is leaving as electrical energy. That means that at least one of two things is true: When they’re not drawing a load, either they heat up more, or they reflect more. If they reflect more, that would certainly have a visible effect. And if they heat up more, well, that might still have a visible effect.
Whats happening is that the electrons are still being knocked off silicon atoms and over to the p-type silicon where they collect, making the p type negative and the n type positive.
After the p is charged up, the voltage is reached that the electrons are thrown back due to the strength of the electric field, creating photons and heat.
There maybe some photons released , but it seems to me its going to translate the energy into heat much more greatly.
The PV is made of crystal silicon If the silicon was visibily affected, it would have be to re-crystalised and lose its PV ability. ergo, its not significantly affected in structure.
If it got hot enough to glow, it would melt… again, causing it to fail to be an effective PV cell.
Note that there is glass and resin in the PV cell assembly, to keep the semiconductor totally clean and dry. So that may be affected by heat in some different way.
That bit about voltage causing the electron to go back from where it came ? That was a gross simplification.
What limits the voltage of the PV cell ?
The number of minority carriers at the junction edge. The number of minority carriers injected from the other side is simply the number of minority carriers in equilibrium multiplied by an exponential factor which depends on the voltage and the temperature. Therefore, minimising the equilibrium minority carrier concentration reduces recombination. Minimizing the equilibrium carrier concentration is achieved by increasing the doping;
Having short diffusion length (the neutral gap between n and p) would mean recombination occurs at greater rate, reducing voltage. But increasing doping reduces diffusion length, so there is a sweet spot combination of diffusion length and doping that maximised power (voltage times current) produced by the cell.
The limit is because the diffusion length is basically an inate property of the silicon… its only so thick, and so can only hold off a defined voltage… trying to make it bigger, just makes it thinner,after the sweet spot.
Great answer. I think you’re saying it would (or could) emit a tiny amount of light when it’s illuminated without being connected to a load? It would be in the wavelength range that the PV cell is sensitive to (i.e. middle of visible light range), right?
Isilder, what’s the false dichotomy? I said that it’d either heat up more, or reflect more, or some combination of the two. You said it’d probably mostly heat up more. That’s perfectly consistent with what I said.
And it wouldn’t need to be red-hot for there to be some visible difference due to heat. One might, for instance, get a miraging effect above the panels, like what one sees on a paved road on a hot day. Or maybe not, but that’s why I put “might” in that sentence.
EDIT: On looking back, I see that I didn’t actually mention the possibility of some combination of the two. I meant to mention it, but I guess that somewhere between my brain and my fingers, that slipped away from me. So if that’s what you mean by a false dichotomy, I apologize; it was my mistake.
I noticed this from a side effect of the solar panels installed on my roof. Besides providing electricity, they also effectively provide shade for the shingles on the roof. So they stay cooler, and the attic stays cooler, so my air conditioning runs less. An entirely unexpected bonus from my solar panels.
This is an interesting discussion. Although I was an ops engineer on the tests of the very first ISS solar panels, and had another acre of commercial panels next door, I never heard anything about this.
I will point out that the cells only convert 10% or so of the sunlight hitting them into electricity so it will probably be a small effect, what ever it is. Also, the hotter they get the less electricity produced. So much so that in order to properly characterize a cell, we had to illuminate it a few microseconds at a time to avoid heating loss.
You’d have had the same effect if you’d simply erected a tent of ordinary tarps or nursery shade cloth a few inches above your roof.
Back when I lived in the desert I often wondered why a false roof of shade cloth on, say, 12" stanchions was not bog-standard house-building practice. It’d have made a huge difference in cost to air condition even if the cloth itself needed replacing every 10 years or so.
Yes, they look different in infrared. Commercial plants will sometimes fly drones over panel arrays to look for malfunctioning cells. (But they’d probably also have per-panel monitoring anyway)
I used to live in AZ, and before that, NM. I like the idea of a roof heat shield, but I can tell you why such a thing was never implemented in AZ:
The housing market there is very cyclical (boom/bust)
During busts, no one is building houses, so no one is building forward-thinking solar features into houses.
During booms, builders compete primarily on price and amenities. Neither the builder nor the speculative buyer assign any value to insulation.
During booms, the people buying the houses (speculators, mostly, but also people from out of state) are not the ones who will pay the utility bills.
The ultimate buyers, who will pay the utility bills, tend to visit in the winter and it never occurs to them that they might pay $600/month for air conditioning if they keep their houses at 80 degrees F. And they will pay that much.
It’s a tragedy-of-the-commons situation: no one cares about insulation, much less solar heat shields. When the houses are built and first sold, no one has an incentive to complain. When the ultimate owner takes possession in May, it’s too late to argue that utility costs are higher than you expected.
In general, wherever you are in the world, residential & many commercial buildings are designed and purchased with zero thought to operating costs. Businesses are slowly coming around to understanding that they can pay a little more up front for a high efficiency building and reap a truly monstrously good ROI on it. A very few consumers are getting the idea too, but so far it’s mostly the greener-than-thou fringe.
The fact so many of both residences and commercial buildings are owned by one entity and occupied by another doesn’t help. The landlord has almost zero incentive to pay more during construction to save the tenant money later on utility costs.
Often “tragedy of commons” is actually “failure of government”. Apply a nationwide tax on inefficiently constructed new buildings and the marketplace would react accordingly. The tax is just defraying the cost of pollution mitigation, defending Middle Eastern oil flows, etc. that are made larger by the shortsighted decisions to build inefficient buildings then waste vast amounts of energy running them all.
As long as our government is run of, by, and for either children or scoundrels I don’t hold out much hope for grown-up decision-making.
Ok, think of it this way. Imagine what would happen if you build a device that pumps water when moving water falls on it from a waterfall. Like a paddle wheel.
So waterfall water falls down, hits the paddle wheel, and the wheel turns a mechanism that moves a piston. There’s a “high” pressure side of that piston, and a low pressure side, and the water is getting moved in a loop to do some work.
Now what happens if you close valves so that no water can be pumped? (same as turning off the switch to a solar panel because you don’t need it’s energy). The paddle wheel will freeze, because it can’t turn. Water will slam into it and no work is done. Therefore, the excess energy must be expressed as heat at the paddle wheel.
Same thing with a solar panel. The quantum P-N stuff doesn’t make any difference to what will happen : all a solar panel is doing is pumping electrons, down one wire and they come back on another wire. If you disconnect or the inverter or battery charge circuitry closes it’s FETs, electrons can no longer flow. Thus the panel has to get hotter, as energy is not flowing away from it in the form of electric current.
Since common solar panels are only 15-20% efficient, it’s not much of a difference. You would barely notice it.
You might actually be able to notice if you’re using the high end, 50% efficient cells they use in satellites. They might get noticeably hotter when disconnected from a circuit.
As for “looking different” : actually, if you have a passive IR camera, you would be able to see a clear difference between panels that are “on” with current flowing and panels that are disconnected or “off”.
From my memories of solid state:
Take a regular diode & remove it from its’ plastic, opaque case. Shine light on it and it will produce a current. Run a current through, it will emit light. Making it emit or absorb enough to be useful depends on band gaps, chemistry, and other stuff I forgot.
And Chronos, you neglected a couple (smaller) possibilities. In the most general case, if you shine light on something it will be:
[ul]
[li]absorbed (turning into electricity &/or heat)[/li][li]reflected (ooh, shiny. at least the calculator solar cells)[/li][li]or transmitted (for a solar cell, this means a photon goes through the silicon without reacting, so you put a mirror on the back of the cell to cheaply double{-ish} the efficiency)[/li][/ul]
I spent too much time starting at those equations.
