# So why are photoelectric panels flat? (Part II)

In this post So why are photoelectric panels flat?, I've asked if we could use flat photoelectric panels with hemispheric cells. Most of the answers seemed to say it would not work.

Anyway, I'm asking again, with a simple design (this is only a draft of course):

The idea is to use hemispheres connected to inverted hemispheres, something like eggs boxes to maximize inter-reflections.

Of course, people will say that there are shadows, etc... But we have to consider a whole day, not only beginning and ending of a day, with maximized absorption. Also, for the same panel size, the surface receiving light is wider. What I'm asking is if inter-reflections could help to absorb more light finally.

• I thought I pretty well answered this situation in your previous post. – Carl Witthoft Jan 5 '16 at 13:18
• When a leaf lays on a cell this is a stress situation for the cell. I think you will get areas of different resistance and his led to hot spots. There are experiments with nanowires, which are also some kind of making area bigger. The newest trend is to produce multilayer cells to earn different wavelengths in different layers. Furthermore they produce half cells, different string and bar patterns and last not least junction cells with strings on the backside. – HolgerFiedler Jan 5 '16 at 13:39
• You still didn't do your homework to understand the angular dependence of the total received radiation. It's pretty much in chapter one of everything you can read about solar design. – CuriousOne Jan 5 '16 at 16:36

• This has nothing to do with the geometric problem that the OP doesn't understand. The surface patterning is meant to increase the quantum efficiency of the cells, which is limited by the mismatch between the optical properties of the vacuum ($\epsilon_r=1$) and that of the semiconductor ($\epsilon_r$ complex with real part >>1). – CuriousOne Jan 5 '16 at 16:39