Solar panel array question

[Powermax]

I am on the UVa Solar Car team, and I was thinking about the best approach to how to best wire our solar panels.


Currently we plan to use 29 in series, but I am concerned that due to the design of the car (rounded) that the panels will not receive equal irradiance, and this may severely impact losses. Especially if one of the panels becomes damaged during the "scrutineering" process.

For instance: If you wire cells in series, and one of them is not producing the same amount as other panels, then the increased series and drop in voltage would equate to power loss. I figure that it's possible to low-barrier schottky diodes in anti-parelell to each cell, which would start conducting as soon as the voltage drops to -0.1V or so. So at least the rest of the cells in the series are still supplying current even with one out of operation, with minimal loss.

But what if cells are wired in parallel? Then when one cell is not receiving light, will the leakage current cause problems? Will the "dead" cell act as a short-circuit to the rest of the good cells? Adding a diode in series with every panel seems like a bad idea as that would cause a drop of 0.1V for every cell. The cells we are using output 2.083V at optimum power output.

The solar technology being used are off-spec dual-junction allium arsenide (GaAs) solar panels. (not the typical monocrystalline or polycrystalline silicon used for consumer gear) and unfortunately there exists no widely available datasheet for these panels as they are very much obsolete, out of production, and they are off-spec. (not good enough for NASA, in other words.)

[Le_Bassiste]

you are raising many valid points there.
series cabling has indeed some decent advantages over paralleling strings: for the same power output, series connection has higher voltage and lower current, so you have either less I^2R losses in the cabling for a given wire diameter, or you may save some weight because of possible wire size reduction. ymmv, pun intended. also,  have a search for "optimizers" for solar PV applications.for example, https://www.solaredge.com/us/products/power-optimizer#/.
these are most likely not suitable for your application, but you get the picture     these give you the advantage of running all PV modules at their individual MPPs. most helpful when trying to get the most out of an array that is subject to partial shading.
for the parallel connection, you get some form of "redundancy" in case of a wire harness(connector) - , or module-failure. also, you can keep system voltage at safe low levels. however, as you said, the decoupling diodes will impact system efficiency.
so, i'd opt for series connection here.

[mycroft]

Two interesting pages are:

  - https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19650020213.pdf
  - http://www.avdweb.nl/tech-tips/pv-panels/bypass-diodes.html