Project: Low expenditure expandable solar “experiment” for home.

• Solar panels are rated with 25 year life. Batteries are expensive and fail. If we are not off-grid, panels are the better investment. So, we can "displace" some of the grid electricity with our solar, as opposed to "replace" it. We then utilize grid availability as our backup.

• Low power circuits (computers, lighting) can be auto-switched from solar to grid with inexpensive plug-in switches and a small backup battery supply (to avoid power loss when the circuit switches from solar to grid).

• Commercially available systems are expensive. I priced a 1600 watt system at about $5k – $1600 for panels, $1600 for 4 backup batteries, $1600 for a controller/inverter. That’s without permits and installation, pretty much as a “starter” kit – and the real-world result is more likely a less-than 800 [usable] watt system.

• On the other hand, DIY low end hardware is pretty cheap.

What I have learned so far: Don’t rely on solar “happy talk.” Advertised capabilities seem to me (based on real world testing) to be vastly over-optimistic. But subject to lots of limitations, it maybe, sort of works. There are a lot of "unknowns", meaning it's easy to do this unsafely and/or fry the equipment (Both of which sort of destroy the ROI.)

My current system:

Two 100 watt panels feed a 12v 50A controller, two 12v 18ah lead/acid deep cycle batteries and a 300w pure sine wave inverter. These feed 120vac to house, where I charge a 330wh lithium-ion portable “power supply that feeds a computer I always leave on.

Cost to date is under $650. In one month’s tinkering, I am now “up to” $.04 per day worth of electricity.

Discussion:

So far, I am under-utilizing the solar panels. My location has a lot of cloudy days and some impossible to avoid shade. Failure to “track the sun” loses a lot of solar potential, so I manually turn them a few times per day. The other challenge is to find a way to fully utilize the (relatively small) solar production without inconvenient loss of power. I have on order an automatic switching device and a backup power supply. This will allow better utilization by automatically switching between solar (when available) and grid as backup.

Grid is far less expensive and more reliable than battery backup. So my [current] goal is to maximize solar use, thereby reducing grid use.

To date, there is no actual benefit to the solar system.

My 100 watt panels max out [at best] around 70 or so. With full sun, I am getting 12 – 13 amps combined off the 12 volt panels. (Since I did not match the panels, it is possible I would have better total output if I did not link them together in parallel.) The portable “power supply” originally came with a folding 100w panel that didn’t work, so I returned it. Further, the charging controller of the power supply is apparently limited to 60 watts in, which I didn’t know when I bought the two current panels. But 60 watts has been the max charging rate on both solar and via the plug-in charger.

Issues:

• Apparently, PWM controller load voltage will be the same as the panel/battery setup. 24v-dc-load. I now have two 12v panels and two 12v batteries. I saw a suggestion that batteries are better in series (among other reasons, easier to spot battery failures). My controller can adjust for 12v or 24v. But, my inverter is 12v. So ...
  
  
• It seems that Renogy controller instructions say not to connect an inverter to the load. inverter shutdown issue. My cheap controller did not state this and I have not pushed the limits yet. There are issues: (1) can the load circuit, which is designed for a light load of lights, or whatever, handle the amps needed by the inverter. Notably, my 330w inverter can surge to 600w. (2) if inverter is connected direct to batteries, will the built in battery protection of the inverter be sufficient? So far, I have kept the wattage draw pretty low so I don't know. (3) My controller is rated at 50A, but it's unclear what the *load* amperage rating is. (4) I do not know if the controller will fry if the inverter draws more amperage than controller will handle.
  
  
• What is the correct shutoff voltage to protect the batteries? My inverter is not adjustable, but as connected I am relying on the controller anyway (unless inverter setting is more conservative?) I erroneously assumed default controller settings would be a good starting point. Having used that setup for about 45 days, I may have damaged the batteries. (They aren't "dead" as some articles suggest would happen. OTOH, I have some corrosion and deposits on a battery terminal, which I would not have expected from SLA batteries. The controller defaults were 14.2v for panel "off" and 10.1 for load "off", but after some use (and some reading) I found data that recommended 10.8 for "off". Along with the scary point that overdischarge (and overcharge) will kill the batteries. Batteries not dead yet, so I don't know. I adjusted to 14.1 and 12.3 (50% level), per linked article. deep cycle battery maintenance.
  
  
• The Flashfish Li+ box does not come with any charge/discharge instructions. So, I don't know whether to assume it has appropriately set defaults and will "protect itself" or whether it should only be depleted to 50% or so. It's a bit more pricey to run this thing to "fail" in order to find out! But, my experience with (for instance) battery powered hand tools is that if you don't use them, the batteries fail anyway. (I have 10 y/o NiMH batteries that still "work", but self-discharge at a rapid rate and are therefore usually dead. I bought some Li+ batteries which are working good, but not very old. Most of my older laptop computers have dead or weak batteries (which are cost-prohibitive to replace given age of computers), but I have a 6 year old Lenovo which I often leave plugged in for weeks at a time, and the battery still seems reasonably ok.)