Solar Cell - Equivalent Circuit Question

[fourfathom]

So here is a common equivalent circuit for a solar cell:

My question: in a real solar cell is power being dissipated when there is no external load?  We will measure V_OC of about 0.6V (or a bit more), and according to the model the cell-generated current I_L is passing through the internal diode and through R_SH. 

But does this current really flow in an actual solar cell?  I'm curious about internal power dissipation and panel self-heating when the panel is open-circuit.  I'm not worried, since I know this isn't a problem in practice, I'm merely curious as to how well the model describes the actual cell's inner workings.

[TimFox]

In that model, with no external load connected, when you apply light flux the photocurrent flows through the internal resistance (which is not linear) until the forward-bias on the internal diode is sufficient for substantial current through that diode.  Thereafter, the photocurrent (proportional to the light flux) flows through the forward-biased diode (at roughly 0.6 V drop), which dissipates power.
With a substantial load (low external resistance), the voltage does not reach the forward bias for substantial diode current, and almost all of the photocurrent flows into the load resistance at < 0.6 V.
(I find that analysis a good mnemonic reminder to determine the polarity of photocurrent in a given wiring diagram.)

[fourfathom]

In that model, with no external load connected, when you apply light flux the photocurrent flows through the internal resistance (which is not linear) until the forward-bias on the internal diode is sufficient for substantial current through that diode.  Thereafter, the photocurrent (proportional to the light flux) flows through the forward-biased diode (at roughly 0.6 V drop), which dissipates power.
With a substantial load (low external resistance), the voltage does not reach the forward bias for substantial diode current, and almost all of the photocurrent flows into the load resistance at < 0.6 V.
(I find that analysis a good mnemonic reminder to determine the polarity of photocurrent in a given wiring diagram.)

Thanks Tim, but I know how the model works.  I've simulated it, and also done many load-line tests on actual panels (I use them on my boat and have set up solar/battery rigs for remote power).  The model is quite good at mimicking the behavior of an actual solar cell.

What I am trying to find out is if this model photocurrent that flows in the open-circuit case is a real thing, causing physical heating in a real cell, or is it just an artifact of the model?  (I do not understand the physics of photocurrent generation beyond a hand-waving model level)

[Siwastaja]

Yes, as far as I have understood, panels do heat up more when disconnected from any load.

[floobydust]

Studies are basically saying PV cells always absorbs a fixed percentage (~85%) of the incident solar energy. Whether or not some is converted and going out (electrically) say 13% depends on the electrical load.
With no load, the cells run that much hotter and apparently can degrade a bit due to the higher temps and thermal cycling.

"A study has declared that only about 13% of the total incident solar energy is converted into electricity and little of it is reflected by the cell surface, thus >85% of the incident energy must be dissipated as heat by the cell [5]. In this condition, the solar cell operates at a relatively high temperature; the situation is even more serious if the module is left in open-circuit."
"... amorphous silicon (a-Si: H) solar cells operating under open circuit conditions can degrade more when compared to similar cells operating under maximum power conditions after 13 days of field exposure." {in the desert}

I would think Spice model is correct then, in that there is a shunt load on the photo-current.

[fourfathom]

Thanks, everyone.  Yes, it does seen that the model effectively represents the real-world behavior in the no-load case.  Now, what about the short-circuit case?  I suppose the cell power dissipation is just I_L² * R_S.  I'm ignoring the solar heating effects, just the electrical factors.

[TimFox]

Thanks, everyone.  Yes, it does seen that the model effectively represents the real-world behavior in the no-load case.  Now, what about the short-circuit case?  I suppose the cell power dissipation is just I_L² * R_S.  I'm ignoring the solar heating effects, just the electrical factors.

Yes.  Under short-circuit conditions, the voltage drop across the internal series resistance should be too small for any substantial current to flow through the forward-biased internal diode.

[Zucca]

Thanks, everyone.  Yes, it does seen that the model effectively represents the real-world behavior in the no-load case.  Now, what about the short-circuit case?  I suppose the cell power dissipation is just I_L² * R_S.  I'm ignoring the solar heating effects, just the electrical factors.

As far I understand it, a PV panel is a current source and it is super happy with a short circuit at the input.
Symmetrically a voltage source with nothing attached is also a happy camper.

So I bet the PV panels heat up more with nothing attached, rather than a short circuit on the output, but this is just theory we need to measure/know the parasitic R in the systems...

BTW measuring the current in a shorted PV is the way to measure how much sun energy it is grabbing, with almost no temperature influence.

[fourfathom]

Thanks, everyone.  Yes, it does seen that the model effectively represents the real-world behavior in the no-load case.  Now, what about the short-circuit case?  I suppose the cell power dissipation is just I_L² * R_S.  I'm ignoring the solar heating effects, just the electrical factors.

As far I understand it, a PV panel is a current source and it is super happy with a short circuit at the input.
Symmetrically a voltage source with nothing attached is also a happy camper.

So I bet the PV panels heat up more with nothing attached, rather than a short circuit on the output, but this is just theory we need to measure/know the parasitic R in the systems...

BTW measuring the current in a shorted PV is the way to measure how much sun energy it is grabbing, with almost no temperature influence.

Yes, this is my understanding as well, but I may eventually get around to some experimental validation.  I have two identical 100W panels that aren't doing anything, and some IR temperature measuring gear, so I one of these days I will do some side-by-side open/short/max-power testing and check the panel temperature.  I couple of years ago I did tweak the solar cell model parameters to fairly-closely match another panel's performance, and I can do it again.  I don't have a particular reason to do any of this, I just find it interesting.

[Zucca]

BTW just because I do not want that somebody here miss it...

https://www.pveducation.org/

The above is IMHO hands down the best place to learn how PV works.

[fourfathom]

BTW just because I do not want that somebody here missing it...

https://www.pveducation.org/

The above is IMHO hands down the best place to learn how PV works.

Thanks!  This looks like the perfect amount of detail for me.

[Sredni]

BTW just because I do not want that somebody here missing it...

https://www.pveducation.org/

The above is IMHO hands down the best place to learn how PV works.

Thanks!  This looks like the perfect amount of detail for me.

If you don't mind the simil-Schwartzeneggerian accent, this short course from UT Delft might be right down your alley

https://www.youtube.com/playlist?list=PLI-7Sbz0TyYWWbQUYd8B4YPc2jJXsQcfA

(The guys name is Arno :-) )

[ejeffrey]

The equivalent circuit should not be taken literally as individual discrete components, it's all happening within the junction.

However, solar cells do heat up more when open circuit.  100% of the absorbed light gets turned into heat.  When you use them to power a load, some percentage (the efficiency) goes to the load, and the remainder is heat.

This is true as long as the absorption of the cell (it's color) doesn't change with load, but AFAIK it does not, or at least only to a very small degree -- not on par with the 20% of light that is turned into electricity,