One problem with solar systems is the need for high-efficiency DC-DC converters. For example, if lead-acid 12V storage batteries are used, then the user is faced with the problem of down converting that 12V source to many devices that use 3V, 5V, 9V, etc.

One way I thought of to side step this problem would be to use battery cells aggregations, instead of single large batteries.

For example, lead-acid cells provide 2V and Ni-Cad cells provide 1.2V. Therefore, you could combine 2 lead acid cells with 1 Ni-Cad cell to generate 3.2V electricity. By using this strategy, you could make a single large battery that would output the array of DC voltages needed. With this scheme the user would only need very simple regulators on the outputs.

The disadvantage would be the need for a more complicated charger.

Would this idea be a superior to using a 12V charging system?

  • 2
    \$\begingroup\$ ewww... don't go there! A few other disadvantages : different discharge and cell lifecycle characteristics, multiple points of failure, abuse one cell and all higher voltages will also fail. One cell will always be the weakest link. Also ... a helluva lot more expensive than one large battery and some converter modules (and LDO regs for low current rails). People did this in the 1920s and moved away from it as soon as transistors came along. \$\endgroup\$ – Brian Drummond Jul 22 '16 at 14:52

Every battery chemistry has it's own charge / voltage curve, it's own internal resistance, and it's own aging characteristics. This means that even if you solved the charging issue and sized the cells to give the same nominal capacity (which wouldn't just be a case of making them all the same size, it would depend on the nominal current draws of each voltage rail) you would still have fluctuating relative differences in your voltage rails depending on current draw on each rail and battery charge level.

All of those are fixable with a simple LDO on each output. But a battery voltage can easily change by at least 30% between full and empty. That means that assuming you want a constant voltage over the full charge range your average LDO efficiency is going to be under 85% for a perfect, 0 drop out device. That's well within the range of a good DC-DC switcher.

So your power supply is simpler but about the same efficiency. Your charger is now far more complex and non-standard, easily adding more complexity than you saved in the power supply. And you've replaced your nice standard battery for an application specific custom battery that costs vastly more.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.