the Apeironian

Solar project technical update

Some technical info about the Diabugu school solar power project:

The existing panels consist of two arrays, each installed at different times and both connected to the water pump controller. The first array has six 135 Watt (12V nominal) panels, all in series. The second array has three 230 Watt (24V nominal) panels, all in parallel. The two arrays are wired in series to the water pump controller. The total nominal open-circuit voltage of the whole system (calculated according to the labels on the panels) is about 170 volts. I measured the open circuit voltage of the whole shebang at around 150 volts (temperature dependent – voltage drops with increasing temperature). Voltage under load varies a LOT, but is usually around 100 volts.

The grand idea behind this project is to arrange everything so that 100% of the available solar power can be used for EITHER the water pump OR battery charging OR some combination of the two, without having to flip any switches or fiddle with anything. It would have been much easier to electrically separate one of the two solar arrays and use one for charging batteries and one for the water pump, but that would have left some solar power going to waste on a regular basis. To achieve minimal waste, I plan to connect the solar charge controller and the water pump controller in parallel to the existing (slightly modified) solar array.

If you have worked with solar charge controllers, you will know that they usually come in 12, 24, and 48 volt nominal configurations, matched with equivalent voltage battery banks. If you have a nominally 24 volt solar array, you have a nominally 24 volt battery bank, and vice-versa. I have a nominally 96 volt solar array to work with. Even if I constructed a 96 volt battery bank, it would be very dangerous (a short circuit with a metal wrench could liquefy the wrench, the battery, and the unfortunate butter-fingered victim), and they don’t make commonly available DC to AC inverters that take 96 volts input. Fortunately, there is a magical new class of solar charge controllers called MPPT, or Maximum Power Point Tracking, which do two primary things differently than traditional PWM (pulse width modulation) charge controllers. First, it compares the voltages of the PV (photovoltaic, or solar) array and the battery bank. Tracking along the current-voltage curve of the PV system, it then uses a beefy DC to DC converter (mine can do 45 amps) to convert the PV input voltage to the voltage at which the battery bank can accept the highest current for its particular nominal voltage. The end result is that you can have a much higher PV input voltage, which then gets converted down to the correct voltage for the battery bank, while maximizing current flow through the system. In practical terms in this context it means 1) I get to do my project, and 2) I get to use thinner wire to run the 80 meters from the panel to the office, which saves more money than the extra I paid for the MPPT unit over a regular one.

These MPPT controllers are not available in this country, so I imported one at my own expense. It’s possible I may be able to reimburse myself.

The max open circuit input voltage for this bad boy is 150 volts. It starts to attenuate current at 140 V though, so I am going to have to reduce the array voltage a little. I plan to re-wire two of the 135W panels into a parallel configuration, which will in turn be wired in series with the rest of the 135W panels. This should shave about 19V open circuit off the total, which should be adequate. This change should not affect water pump operation.

I plan to use a 24V nominal battery array. I want to get 8 200 Amp-Hour deep cycle sealed batteries, but may have to reduce the number depending on cost. I plan to get a 1000W 24V pure sine wave inverter (best quality I can find).

One remaining big mystery (which nobody seems to be able to answer for me) is how the charge controller and water pump controller are going to behave being wired in parallel. My instinct tells me that if they are both running, they will split the available current depending on their individual resistance characteristics. The resistance for the water pump should be pretty consistent (it’s either pumping or its not), but the charge controller will be all over the place, depending on the charge level of the batteries and the load on the inverter. If you know about this stuff and can predict more intelligently what this beast is going to do, I would be very grateful for your input!

I also have to figure out what conduit I’m going to run the wire in underground, and a few other details. Oh yeah, if anyone knows whether I need to ground this thing on the charge controller end, even if it’s already grounded at the pump controller (wired in parallel), please let me know!

I think that’s about it.

MC