Bug with solar panels exposure

[Sharpy]

The screenshot pretty much says it all. Sun exposure of panels is taken from a point somewhere by the base. A panel exposed in 90% is still "blocked" if that specific point is blocked.

Windows, build id = 00861

2015.06.23 at 17:19:17 CEST

Branch: master

The only mod installed is KER, totally unrelated to the issue.

[sumghai]

Solar panels in real life are composed of an array of individual photovoltaic cells wired in series (like Christmas tree lights). If only the tip of the panel is shaded, the rest of the cells continue working to supply ElectricCharge of the vessel. However, if only a tiny portion of the base is shaded, it cuts off access to the rest of the (illuminated) cells, and so you get no ElectricCharge.

In short - that's pretty much how real-life solar panels work, and probably isn't a bug.

[Sharpy]

Solar panels in real life are composed of an array of individual photovoltaic cells wired in series (like Christmas tree lights). If only the tip of the panel is shaded, the rest of the cells continue working to supply ElectricCharge of the vessel. However, if only a tiny portion of the base is shaded, it cuts off access to the rest of the (illuminated) cells, and so you get no ElectricCharge.

In short - that's pretty much how real-life solar panels work, and probably isn't a bug.

I don't know how solar panels are wired internally, but I can with absolute confidence say that wouldn't be the consequence of wiring them in series. Yes, *damaging* any panel if they are wired in series would break the circuit. But by merely not receiving sunlight a panel doesn't start acting as isolator. Plus series wiring means specifically that every single element of the circuit has the same influence on the rest of the circuit.

Besides, wouldn't it be silly if a single bird dropping could disable three meters of a solar panel, just because it splattered on the one special cell?

[Claw]

I would guess it probably has more to do with the fact that KSP only calculates shadowing at the part's root, rather than over multiple points. I didn't confirm that is how it's done, but based on what I know of other parts I am fuessing this is the case. So yes, this will likely lead to exactly what you describe.

The alternative is casting multiple rays or testing multiple points. Probably feasible for low count craft, but adds up. Idk if this will get addressed, but it's the first time I recall this specific complaint. Probably something to keep an eye out for, but I can't recall having this problem myself for arrays that pop out of cargo bays.

Cheers,

-Claw

[Sharpy]

One more alternative is picking data back from the graphics engine and checking how much of the panel is actually lit up by the Sun - skip separate physics calculations, use the lighting rendering algorithm which is active at all times.

[Skalou]

Solar panels in real life are composed of an array of individual photovoltaic cells wired in series (like Christmas tree lights). If only the tip of the panel is shaded, the rest of the cells continue working to supply ElectricCharge of the vessel. However, if only a tiny portion of the base is shaded, it cuts off access to the rest of the (illuminated) cells, and so you get no ElectricCharge.

In short - that's pretty much how real-life solar panels work, and probably isn't a bug.

Right , that's why there is a margin between the root and cells IRL,

so ksp one's are a bad conception?

but i find it bad from a gameplay point of view, if you want to reproduce the irl problem, give us some good tool to solve it.