New solar installation 2015-11-13
PumpsSolar
0 Comment
Having finished construction and testing of the water circulation pumps, the next step is to find a permanent power source. I’ve always been interested in the potential of solar for bringing power to any point on the land without needing an electrical grid connection. These pumps are ideally suited to solar, being almost 30x more efficient than the commercial centrifugal water pumps that I’ve tested and compared it to.
As such, these pumps have a much better chance of being driven effectively by solar power. Additionally, the circulation pumps we built use 24V DC motors, which increases the efficiency even further since it doesn’t need the inverter that an AC 220V pump would require. And on top of all that, the DC motors can easily be slowed down on a cloudy day if there is less power, which can’t easily be done with AC pumps.
So, we bought 2 280 watt 24V panels from Amorn Solar:

The panels are mounted to a custom built steel frame mounted on 4 pilings. Today we put up the electrical box for the solar charge controller, but haven’t wired in the controller yet:

Here is a view of the frame before the panels were mounted:

I ordered 4x 6V T-105 225Ah batteries for the system. That should give plenty of juice for this setup. In my reading a while back, I saw a masters thesis done in Thailand on solar electricity here, and one of the conclusions from actual experiments was that the average amount of energy produced by a solar panel per day in Bangkok is about six hours of the rated peak power. So for this setup,

I have 2x 280W panels, so the average energy it should produce per day is 2 * 280W * 6h = 3.36kWh.

For comparison, the batteries are 4 * 6V * 225Ah = 5.4 kWh

This should mean that the batteries should have no problem holding a days worth of solar electricity. And given that the pumps should be running all day, every day, the batteries shouldn’t have to buffer very much.

Another number of interest is that in this post, I measured the power draw of a single pump running at 12V to be 54W. So for 2 pumps running 24 hours, we get 2 * 54W * 24h = 2.6 kWh. So on an average day’s sun, the solar panels ought to be able to run the pumps at the speed I was testing it (at 12V) for 24 hours per day every day.

This calculation ignores a number of contributing factors such as:

  • The solar panels are 24V. While the pump motor is also rated at 24V, I don’t really want to run it this fast, so 12V is more than enough. In order to run it at the lower speed on 24V input, it will need a PWM motor drive circuit, which will have a little bit of power loss.
  • In order to run at night, the solar panels will need to charge the batteries. This charging process is normally only about 80% efficient, so there is more loss here.
  • Not every day will have an average day’s sunlight
  • Many of these numbers come from other people’s tests. Until I test it myself and get some real data, these numbers are just hypothetical

Despite all of this, the numbers should be ballpark right. And for the purpose of running the pump, if I can attach a battery voltage level sensor to the motor drive circuit, there shouldn’t be too much of a problem if the pump controller simply slows down the pump when the batteries are getting lower. This should mean that on days that the solar panels get less sunlight, the batteries will end up less charged, resulting in the pumps running at a slightly lower speed to compensate. This should work out just fine.

One of my favorite things is to calculate and theorize until I am ready to build and then build it to see how closely the reality matches the theory. I am very excited to get everything wired up and see how this all works in actuality now.