Quick one: unused solar panel on the roof, better short or open circuit?

[gnuarm]

Solar panels slowly degrade, so leave it open and put a cover to preserve it.

May I ask why if it is so obvious? Just curious...

Isn't it obvious? A short circuit would dissipate all the generated power in the panel itself, accelerating aging even more.

Solar cells don't work that way.

[gnuarm]

Open is probably thermally the worst case.
Short is a little bit better, a part of the electrical energy gets dissipated into the wires.
Best is to put a load on them (light bulb or so), so they run really cooler at peak power.

In the open circuit case, there is no external current, so no power.  There can be leakage current within the solar cell, but that will be rather slight, otherwise it would significantly impact the performance of the solar cell. 

In a closed circuit case, there is not external voltage, so no external power.  There can be internal resistance, which will dissipate internal heat. 

None of this will be significantly different from what is observed in normal operation.

[gnuarm]

Open is probably thermally the worst case.
Short is a little bit better, a part of the electrical energy gets dissipated into the wires.
Best is to put a load on them (light bulb or so), so they run really cooler at peak power.

And here we have the right answer. It's surprising to see how people think their first instinct must be "obviously" correct even when they have absolutely no idea about the actual physics involved.

It's also surprising how people will offer an opinion, without any supporting information.

[gnuarm]

You'll also be dissipating energy inside the solar panel heating it even further while it is already baking in the sun.

Why should a shorted panel heat up more than an open one?
In both cases there is no energy supply besides the sun, which heats them up both.
Where should the energy come from to heat it up even further?

Only when you withdraw (electrical) energy from the panel, you cool it down a little bit.
But in order to withdraw energy, I*V must be > 0, i.e. neither open nor shorted.

With a shorted output, you will have maximum current, which produces heat in the solar cell from the internal resistance.  With an open output, you have no current, but maximum voltage, which will push current through the internal leakage resistance, producing heat in the solar cell.

[gnuarm]

Wanting to short a solar panel is idiotic. A solar panel consists of cells in series and you'll be pushing full short circuit current through each cell which kicks the crap out of the weakest cells and degrading them quicker. You'll also be dissipating energy inside the solar panel heating it even further while it is already baking in the sun. More degradation.

The current still flows in an open-circuit (unloaded) solar cell, it just flows through the same intrinsic solar cell diode that establishes the V_OC.

Easy enough to find out.  Put a solar cell in the sun, with the output open.  Let the temperature settle and measure.  Connect the outputs and let the temperature and measure.  The heat dissipated with impact the temperature.  If you can't measure the difference, then it won't matter.

[gnuarm]

Again, this question is really simple to answer by applying some common sense and logic. Are solar panel manufacturers recommending to short solar panels when not in use? I have not been able to find any that say the panels should be shorted when not connected to an inverter. Secondly, are solar panels / arrays of solar panels shorted while an installation is not in use? I have not been able to find any recommendation to do so. Even though in a larger install (say hundreds of panels) it is likely solar panels are fitted long before they are hooked up to the inverters. So where does that leave manufacturers of solar panels to optimise their panels for in case there is no load connected? Short or open? The obvious answer is 'open' because in the end that involves the least amount of work.

And there is another electrical reason to keep panels open as well. When shorted: In case of partial shading you'll push the highest current possible through the bypass diodes which prevent cells from becoming reverse biased.

Let me understand.  You find no information telling the user to short the panels, so that means the panels should be left open?  I guess you can get very interesting results if your searches are not very good.

[Someone]

I framed that carefully as the no load situation. It's not uncommon for people to oversize the panels relative to the inverter or have curtailment.

It happens sure but it's not common. Again, probably 95%+ of the time the energy from the panels is being used, then there is a small section of time it is not (like Dave with his undersized microinverters). This is different from 100% of the time the panel sitting unused, which is what OP is talking about.

Its not that different at all. If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

[Zucca]

With a shorted output, you will have maximum current, which produces heat in the solar cell from the internal resistance.  With an open output, you have no current, but maximum voltage, which will push current through the internal leakage resistance, producing heat in the solar cell.

Oh yes, I agree 100% with the above. It would be interesting to know which resistance produces more heat... the internal leakage resistance or the internal resistance.

Also I would assume that internal leakage resistance >  internal (series) resistance, so if the current is the same the internal leakage resistance is burning more heat than the other one...

It looks like you knows how solar cell works... can you please confirming me than it is a device that generate a current and not a power when exposed to light?
In other words, the power generated in a solar cell is an effect of the current "created" by the light.

Many thanks!

[Zucca]

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

surprised as well about the lack of studies in this situation, or.... maybe for the experts it is obvious that in short circuit it will be better and they no bother to study it?
Only God knows...

[f4eru]

Only God knows...

Hmm. It is scientifically proven that God does not know thermodynamics of PV panels.

[gf]

...can you please confirming me than it is a device that generate a current and not a power when exposed to light?
In other words, the power generated in a solar cell is an effect of the current "created" by the light.

In the end, does it matter whether the chicken or the egg came first? For a given light incident, you can observe a particular I vs. V characteristic from the outside, and if you connect a load, then the cell delivers a corresponding amount of power to the load.

Furthermore, conservation of energy cannot be circumvented. The solar power absorbed by the cell must be equal to the electrical power deliverd to the load + the power disposed by cell to the ambient (via IR radiation and via thermal transmission to the air). Eventually, the cell reaches a temperature at which an equilibrium between incoming and outgoing power is established (the hotter the cell, the more power it can dispose to the ambient).

And does it make a difference at the end, whether (say) 15% of the absorbed solar irradiance power heat up the cell directly, or whether the same 15% are converted electricity first and then this electricity is dissipated inside the cell and heats it up? At least from the power balance POV, it does not make a difference.

[gnuarm]

With a shorted output, you will have maximum current, which produces heat in the solar cell from the internal resistance.  With an open output, you have no current, but maximum voltage, which will push current through the internal leakage resistance, producing heat in the solar cell.

Oh yes, I agree 100% with the above. It would be interesting to know which resistance produces more heat... the internal leakage resistance or the internal resistance.

Also I would assume that internal leakage resistance >  internal (series) resistance, so if the current is the same the internal leakage resistance is burning more heat than the other one...

It looks like you knows how solar cell works... can you please confirming me than it is a device that generate a current and not a power when exposed to light?
In other words, the power generated in a solar cell is an effect of the current "created" by the light.

Many thanks!

It's not that I "understand" solar cells more than anyone else here.  I look at the data (mostly the I-V curve) and accept what it tells me as the truth. 

I don't remember data ever lying.

I don't think the solar cell produces a fixed current or a fixed power from a fixed illumination.  Look at the I-V curve, and tell me what is going on.

[Siwastaja]

"macroheating" of the cell, i.e. heating while ignoring spatially small effects like hotspotting, is the easy part, as the constant nature of emissivity regardless of load is easily observed: both open-circuit and short-circuit conditions give roughly the same heating and hence directly heating-related aging. Only connecting a load (in absence of MPPT tracking, even just something equaling roughly the I_mpp at V_mpp, is an improvement) can reduce the temperature of the cell. This is all obvious from basic conservation of energy, so pretty much common sense to everyone except nctnico.

Now the macroscopic temperature probably isn't the only aging mechanism of the cell, and what else happens requires serious understanding of the detailed cell physics apparently no one here has any idea about, me included.

It's not a surprise this is not discussed in cell datasheets. It makes little sense to invest money in this expensive stuff and yet more money to install it, only to not use it. And short periods of non-usage would be irrelevant anyway, clearly the cell does not age while open-circuit for weeks or months, so clearly aging does not increase by orders of magnitude, compared to maximum power point load situation. If aging rate increases by 30%, 50% or even 100%, this does not matter to the manufacturer as the panels are not intended to sit unused for years. 1000% increase would be something that needs to be discussed in datasheets/instructions, and would be a problem when inverters in some situations have to cut production e.g. due to voltage or frequency rise, or utility demand control signals.

[gf]

I don't think the solar cell produces a fixed current or a fixed power from a fixed illumination.  Look at the I-V curve, and tell me what is going on.

The commonly used equivalent circuit consists of an ideal (solar irradiance dependent) current source and a diode (and some resistors). It is supposed to model/approximate the observed I vs. V curve. But this is still just a behavioral (i.e. "as if") model. In practice, of course you cannot dismount the current source and the diode from the cell, as separate components.

Now the macroscopic temperature probably isn't the only aging mechanism of the cell, and what else happens requires serious understanding of the detailed cell physics...

That's true, of course. I guess there exists no trivial model for aging.

[Kleinstein]

Usually the aging of the crystalline silicone solar cells is not very fast.  Looking at hot spots and not the overall temperature is a good point: it is enough to have hot spots fail. There are different types of hot spots:
1) points with high current flow (e.g. near contacts), that would get hot especially under short circuit.
2) points with locally higher leakage current (lower "diode" drop" that would get hot under open circuit conditions as there is more voltage and thus the chance that areas with defects can get hot.

Good modern panels should be checked for both types of weak points, e.g. by IR monitoring the cell under both conditions. So I would not worry too much about the hopefully relatively short time with the PV panels stting unused.

[mikerj]

This paper covers the accelerated degradation of an open circuit panel in a desert, so not necessarily applicable to less harsh environments.

https://www.sciencedirect.com/science/article/pii/S2211379716301280

[CatalinaWOW]

This thread demonstrates one of the features of human behavior.  There are a variety of approaches to the answer, with the conclusion similar in a substantial majority of the answers.  But the OP doesn't like the result and searches the outliers and creates new explanations of why his preferred answer is best.

There is much to learn here, both from the varied ways to skin the cat, and from the demonstrated social behavior.

[thm_w]

Its not that different at all. If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

It is different: ~5% of the time unloaded vs 100% of the time.
It is a significant cost: FET that can handle full shorted current of all panels + additional heatsinking and thermal dissipation required. And again if you want to do it optimally, you need a huge load dump bank.

Its like asking why car manufacturers don't all include an AC charger to keep the internal battery maintained when the car sits unused.

This thread demonstrates one of the features of human behavior.  There are a variety of approaches to the answer, with the conclusion similar in a substantial majority of the answers.  But the OP doesn't like the result and searches the outliers and creates new explanations of why his preferred answer is best.

Well OP hasn't even told us how long they expect the panels to sit. If its some days or months or years. They were somewhat clear about not wanting to spend effort or money on it, so all of those solutions are out.
Its still an interesting discussion to have.

[gnuarm]

Its not that different at all. If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

It is different: ~5% of the time unloaded vs 100% of the time.
It is a significant cost: FET that can handle full shorted current of all panels + additional heatsinking and thermal dissipation required. And again if you want to do it optimally, you need a huge load dump bank.

Its like asking why car manufacturers don't all include an AC charger to keep the internal battery maintained when the car sits unused.

Poor example.  All EVs include the AC charger in the car.  Perhaps you are thinking of the EVSE that provides the connection from the AC outlet to the car's port.  Not all cars include that when you buy the car.

[bdunham7]

If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

A shorting FET would be at least some cost, and even more relevant an additional potential failure point.  If I were the product manager that would be a hard no unless the benefits were well established, well known and in demand by customers.  I think if you raised this issue with 99% of installers and 99.9% of solar customers, you'd get a blank stare, so the last two are out.   

[wraper]

...can you please confirming me than it is a device that generate a current and not a power when exposed to light?
In other words, the power generated in a solar cell is an effect of the current "created" by the light.

In the end, does it matter whether the chicken or the egg came first? For a given light incident, you can observe a particular I vs. V characteristic from the outside, and if you connect a load, then the cell delivers a corresponding amount of power to the load.

Furthermore, conservation of energy cannot be circumvented. The solar power absorbed by the cell must be equal to the electrical power deliverd to the load + the power disposed by cell to the ambient (via IR radiation and via thermal transmission to the air). Eventually, the cell reaches a temperature at which an equilibrium between incoming and outgoing power is established (the hotter the cell, the more power it can dispose to the ambient).

And does it make a difference at the end, whether (say) 15% of the absorbed solar irradiance power heat up the cell directly, or whether the same 15% are converted electricity first and then this electricity is dissipated inside the cell and heats it up? At least from the power balance POV, it does not make a difference.

I'm not convinced that when no current is drawn 100% of the light that would otherwise provide electric power actually gets absorbed by the panel and converted into heat rather than reflected. Also I don't think something like "15% are converted electricity first and then this electricity is dissipated inside the cell and heats it up?" will happen with no actual current flow. It either should be converted into electric energy and resistive losses if current flows or directly into heat if no current is drawn with possible change in amount of light reflected. It needs some actual data rather than guesses.
IMHO the worst that can happen with open circuit is no change on how much non loaded panel heats up compared with short circuited. I'm surprised about the idea that solar panel somehow can internally convert electric energy into heat with no path for current flow. There will be electric potential but no path for current flow other than small leakage current. Short circuit does not take away the energy from the panel, all dissipation still happens within the panel. The only way to take the energy away from the panel is providing normal load that will dump the heat somewhere else.

[Someone]

If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

A shorting FET would be at least some cost, and even more relevant an additional potential failure point.  If I were the product manager that would be a hard no unless the benefits were well established, well known and in demand by customers.  I think if you raised this issue with 99% of installers and 99.9% of solar customers, you'd get a blank stare, so the last two are out.

Which is exactly what I said, there is no well known benefit for this, customers aren't demanding it, neither are manufacturers pushing it. IF shorting during no load was beneficial we'd expect to see it implemented. Therefore it is pretty safe to conclude that there is likely no benefit, even without some solid evidence to prove that.

Cost of implementation is trivial compared to possible gains.

[bdunham7]

I'm surprised about the idea that solar panel somehow can internally convert electric energy into heat with no path for current flow.

A photon of sufficient energy bops an electron up and across the junction.  If the photon has energy above and beyond what is required (which it will for any wavelength of photon shorter than the absolute maximum), the electron loses energy (mostl as heat although there could be re-emission, I suppose) until it settles down to whatever the energy level (voltage) is on the other side of the junction.  The remainder is your available electrical energy.  Once enough of those electrons get bopped across the junction to raise the voltage on that side to the junction voltage, they flow back down through the junction.  Since the junction is a diode, the voltage is about half a volt or so.  So there's your continuous current flow.  The diode nature of the junction is what limits the open circuit voltage of the solar cell.  If you short the cell, then the voltage on the other side of the junction is very low and the electrons that get bopped across that junction lose even more energy--almost all of it--as they settle down to that lower voltage level.  Then the very low voltage is just enough to propel them around the shorted external loop, but not back down the junction.  Either way (open or short) the photons that interact with (bop) electrons end up seeing almost all of that energy dissipated in the cell.

I doubt there would be much reason for the emissivity of the panel to change much at the wavelengths sufficient to bop the electrons enough to matter.  The panel isn't going to run out of them since if even 1% of available electrons were bopped across the junction at one time you'd probably have thousands or millions of volts, so the supply of available electrons for the photons to interact with isn't likely to go down enough to matter.  The current flowing back through the junction at ~0.5V might (I think it does, anyway--I've read about this effect somewhere) cause the panel to emit IR radiation but at a much longer wavelength (0.5eV ~ 2400nM).  This would yield a slight cooling effect for the open circuit case. 

[gf]

The current flowing back through the junction at ~0.5V might (I think it does, anyway--I've read about this effect somewhere) cause the panel to emit IR radiation but at a much longer wavelength (0.5eV ~ 2400nM).

Dou you mean IR radiation like a LED (not just the blackbody radiation due to surface temperature)?

[nctnico]

I framed that carefully as the no load situation. It's not uncommon for people to oversize the panels relative to the inverter or have curtailment.

It happens sure but it's not common. Again, probably 95%+ of the time the energy from the panels is being used, then there is a small section of time it is not (like Dave with his undersized microinverters). This is different from 100% of the time the panel sitting unused, which is what OP is talking about.

Its not that different at all. If there was a life extension to panels under no load conditions by shorting them, and (as you agree) those conditions exist some of the time in the real world, it would make sense that products would be marketed to match that. Adding a shorting mosfet to an inverter is not a significant cost.

We're lacking the evidence that shorting panels will increase lifespan. Not lacking is the occurrence of this, or the low cost solutions.

Yep. We don't know what solar panel manufacturers do to optimise their products for the actual use case down to the semiconductor physics & chemistry level. Graphs and simplified models won't tell you that. Just like you can't get SOA information from a simplified transistor model. The question from the OP is about how a solar panel ages / deteriorates. The simplified model doesn't answer that so it is not relevant to the question.

And you are right about solar panels needing to be able to support / designed for being partially loaded or even unloaded for prolonged periods of time. Inverters and/or the grid may not support solar panels working at full capacity. In the NL there are areas where the grid can't handle the amount of solar energy on a sunny day so the inverters switch off due to the mains voltage becoming too high. That leaves the panels unloaded in what you would call normal operating conditions. And it is not unreasonable to assume panels are oversized for the inverter in a significant number of places. When I look out of my window I see several >4kWp installations and I doubt all of them have an inverter capable of handling the full capacity of the panels. Likely most of these installations are on a 16A (ballpark 3600W) circuit as this is what solar panel installers offer by default.