maintenance
-
Hey gang, viernullvier here. I've been the designated guardian of compost.party for a few days and used the opportunity to play around with the solar power setup. My goal was to come up with a setup that is able to power the phone throughout the notoriously bleak Berlin winter while doing away with the need to unplug the panel every morning. TL;DR: We're not there yet, but I learned a lot along the way and have a couple of ideas for future improvements.
Of course, the hardware was again gathered in true compost fashion: I've been able to get hold of some outdoor-grade 40 W solar panels that have been decommissioned at my workplace as they are superseded by a new hardware generation. I also found a cheap second-hand solar charge controller for lead-acid batteries that has been sitting unused in an RV for some time. The hardware was completed by a couple of 12 V 5 Ah lead-acid batteries from someone's abandoned DIY residential solar battery project.
Before I took over compost duties, I've been dry-running the setup in my room for a few days while keeping an eye on the reported voltage. The first results were inconclusive: With one panel connected to three batteries in parallel, the controller was at least able to maintain the state of charge with the battery voltage fluctuating by about 0.2 V. I was a bit worried that this wouldn't be sufficient for actually powering the phone, so I wired up a second panel in parallel and used only two batteries (I assumed that fewer batteries will charge faster with the same charge current).
The results: Even though compost's reported battery level had been steadily sitting at 99% the whole time, the test was a failure. The phone was mostly running off the batteries, the solar power only delayed the battery discharge during daytime. At some point I even had to switch over to a second set of full batteries to prevent over-discharge. In the end, it turned out that the simple charge controller was not really fit for the purpose - and sure enough, infamous electronics youtuber Big Clive already did a thorough teardown of this very device.
In which way did it fail exactly?
- The controller has no backfeed protection. At nighttime, the battery voltage was actually running through the the panels which acted as a large resistor, draining the battery. Two panels in parallel even decreased the resistance, making the problem worse.
- The controller's USB outputs are hooked up directly to its internal voltage regulator, so they are always active and can not be switched off if the battery's state of charge is too low. It's possible to over-discharge the battery by using the USB outputs.
- The controller is just a sophisticated switch. It doesn't do any voltage conversion in order to optimize the solar panel's output, so it is not able to charge the batteries at all in low-light conditions.
Where to go from here: While 1. could be fixed by adding blocking diodes in series to each panel and 2. could be circumvented with a 5 V buck converter on the properly switched 12 V output, I would prefer to keep the setup as simple as possible. I'd rather go with a different charge controller that already has the proper protections built in and supports maximum power point tracking in order to address 3. as well.