Posted by: @old_scientist

↑

The first law of thermodynamics states that energy can't be created or destroyed, so where does the excess clipped energy go when a solar array is generating 4kW on a sunny day but the inverter clips it's output to 3.6kW? Is the 400W of clipped energy lost as heat from the inverter? My inverter is in the loft so I've no idea how hot it gets on a sunny day.

Einstein!!  I think we lose it in heat in the inverter, don’t need to worry about that in the winter, solar is useless then.  I’m thinking wind might be a better option if someone comes up with a decent device 

Posted by: @old_scientist

↑

The first law of thermodynamics states that energy can't be created or destroyed, so where does the excess clipped energy go when a solar array is generating 4kW on a sunny day but the inverter clips it's output to 3.6kW? Is the 400W of clipped energy lost as heat from the inverter? My inverter is in the loft so I've no idea how hot it gets on a sunny day.

It's a good question and one which I have asked myself.  We really need a solar panel physicist to answer, I am a lapsed physicist but never really studied this particular specialty.

Firstly I doubt the panel does generate 4kW when clipped, because it's not correctly loaded, but for sure the solar energy does still hit the panels, so the question remains valid albeit slightly modified.

I think  that when it is clipping the inverter presents less than a matched load to the panels, so I suspect some of the energy is absorbed by the panels resulting in them heating up more than would otherwise be the case.  I could easily believe that, in addition, there is an effect on the reflectivity, but definitely don't know this to be the case. 

As you say it has to go somewhere, the above seems plausible at least .

It's worth bearing in mind that solar panel efficiency has remained stubbornly around 20% for a decade more.  This being the case some additional excess energy lost to heat is unlikely to cause anything to melt.

I understand that it gets dissipated as heat, mostly in the panels but some in the inverter. Since most solar power generates heat in the panels (80%) because they are only 20% efficient the extra heat isn’t a lot. We don’t need spurious heat though!

In the solar panels.  

On a nice sunny day, a square metre of panels would get 1kW from the sun.  Of this we might expect 200W of electical power if a suitabel MPPT inverter is connected, and then 800W as thermal power warming up the panel itself.  If you disconnect the panel, then the whole of the available 1000W will heat up the panel. 

When the inverter MPPT backs off, it just leaves some of the power in the panel.  If could choose the best MPPT point and to dissipate the excess power in itself instead - but no inverter will do this, as it would cause overheating and unreliability in a physically small inverter, while the panels are designed to be able to dissipate the thermal power.

Not that it's a good idea to leave panels in the sun unconnected - I think eventually they do get damaged this way for some reason.

Why do panels get damaged if not connected?

When they are normally in the sun, they get 800W/m^2 evenly distributed, and they warm up everywhere.  Silicon doesn't like being super hot for too long, it will eventually degrade, but it's a long term problem.

I think if you leave a panel disconnected, you can think of the extra 200W/m^2 as actually electrically generated and then dissipated by the panel itself.  The graph above shows that as temperature increases, the open circuit voltage of a PV panel drops - a positive feedback effect.  So I would expect an unconnected panel to have hotspots where the current flows, and these will degrade it faster.

@robl thank you, makes perfect sense, and my wife is relieved to know the inverter isn't sat in the loft smouldering due to excess heat because of an oversized array.