Table of Contents
I've been long fascinated by the concept of solar power. The idea of putting some weird blue plate in your garden or on your rooftop and getting free electricity sounds almost like a joke and yet, not only is it serious, it's a reasonably cheap project to accomplish as long as your requirements aren't super high.
That being said, I had a slight apprehension to messing with solar. Well, not even necessarily strictly with solar, rather with anything electric. My late grandfather was an electric engineer and he used to tinker on a small couple-dozen watts panel as part of his many retirement projects. I don't remember the exact details because I was very young, but I recall seeing a bunch of huge boxes and incomprehensible cabling, which put the subconscious idea into my mind that this is far too complicated for someone without a proper education in the field.
However, fate would see to it that just a couple of days before I started writing this post, nearly fifteen years after I was first enamoured with photovoltaics, I was reading the excellent LOW-TECH MAGAZINE's About the Solar Powered Website article, which details how they managed to wire up a self-sufficient Raspberry Pi(-like) based system as their server, which could, technically, run out of power if the weather was unkind.
What struck me as somewhat shocking is that the setup itself didn't seem all that complicated. Yes, there were several parts, including stuff whose purpose I wasn't aware of yet, but as a whole it seemed remarkably approachable.
This ultimately led me to the Youtube channel Footprint Hero, which features a guy called Alex Beale, whose entire shtick is to make easy to understand videos about solar power and how a complete beginner could start utilizing it themselves.
I was so impressed by how turn-key the entire process seemed based on his videos, that I became convinced that I want to try this stuff.
Disclaimer: I am a programmer by trade, not an electrician. While I have a generally decent understanding of software, hardware is a whole another beast and I am by all means pretty much a complete beginner. I tried to ensure my post contains correct information, but I cannot claim that I haven't made any mistakes in this post.
Please keep this in mind while reading and do your own research! It's well worth it, solar is a hella cool topic. Also corrections are very welcome in comments, if you find this post on social media.
1. The plan
Still, before jumping into doing things, one needs a reasonable target to aim for. Obviously, I wasn't going to just start filling my roof with panels and running my entire home off solar. Such a project would cost many thousands of euros, which I simply don't have, would require hiring a bunch of engineers and workers, and finally it likely wouldn't even pay for itself in a reasonable amount of time due to the solar-unfriendly laws in my country.1
So, I decided to aim quite a bit lower. I have previously bought myself a plug-in watt-meter, mostly just out of curiosity to see how much power the different devices in my room use. The most interesting finding was when I noticed that my PC rarely seems to use more than 150-200W and, when I'm not using it for any strenuous tasks, it usually dips even lower into the 90-110W territory.
There is a law in place that mandates that below a certain threshold households pay a discounted price for electricity. We happen to be usually below this line, but sometimes things don't work out quite how we hoped and the meter creeps up slightly above the threshold. When this happens, the bill is suddenly a lot higher than it should be.
I figured, hey, if we only just go slightly higher every once in a while, perhaps by taking my PC out of the equation we could ensure we only ever stay in the discounted range.
The jury is still out to see whether my theory works in practice, but this was enough of an excuse for me to start looking into buying the necessary things to build a small system.
2. Parts of a solar setup
If you already understand how PV works, feel free to skip this part, I just want to get everyone up to speed first. To run a tiny solar system, one needs the following parts at the very least:
- A battery: Stores excess energy in the form of chemical energy. Nowadays batteries are usually made from Lithium-ion salts and range from a couple hundred up to thousands of watt-hours worth of capacity.
An inverter: As solar panels generate direct current, while most appliances use alternating current, you need a so-called inverter which can turn one into the other.2
Figure 1: Schematic of an inverter. Notice how the path of electricity keeps turning around (i.e. inverting) with every switch. By C J Cowie - Own work, CC BY-SA 3.0.
Inverters use a set of transistor-pairs, which can be switched on and off extremely fast. By flipping between these at twice the speed of the intended frequency of the grid (50Hz in the US, 60Hz in th EU), the current is forced to rapidly switch directions, thus inducing an alternating wave, i.e. AC electricity.
Sadly this process is not 100% efficient. Generally inverter efficiency ranges between 85-95%, which means the true, useful capacity of your battery can easily be 15% less than stated in the specs.
A solar panel: Stars of the show. Generates electricity based on more or less the same principle as LEDs just in the opposite way. Did you know that you can actually use LEDs as an ad-hoc light sensor? I found it a pretty cool piece of trivia.
Photons radiated from the sun knock electrons free, which do a lap around the circuit the panel is wired into. This is usable electricity.
MPPT: Also known as a "Maximum Power Point Tracker". It is a small device that connects to both your battery and your solar panel and can, as the name implies, track the panel's maximum power point.
The maximum power point is the intersection of the ideal amount of voltage and current with regards to the solar panel's "Fill factor" (which basically quantifies the panel's quality) that provides the highest amount of usable power.
In an ideal world, where the panel is always in constant sunlight and isn't affected by temperature, this device would not be needed, as with those parameters frozen, the panel would generate a practically constant voltage and current (without regards to degradation, which would slowly worsen the output over the course of a couple decades).
However, realistically this is never the case. As seen on the graph below, the hotter a panel gets, the less voltage it's able to generate (see also the section on NMOT):
Figure 2: The current-voltage curve of my particular panel. At 50˚C it provides nearly 6V less voltage, than at 0˚C. Taken from the reference sheet.
What's worse the output also varies massively by how much of the panel is shaded. If you only used the raw output of a solar panel you'd find that even if only one or two of its cells were shaded, you would not only lose, say, 5-10% of your expected power, but up to a quarter of even half of it.
This is because the cells on a panel are wired in a shape kind of like a "string" or "pipe" and even a single shaded cell can act as a blockage in this "pipe", because not only do they not generate current, they also act as resistors and thus allow very little current to flow through them.
To prevent completely choking the circuit, modern panels have so-called bypass diodes built in, which allow going around entire rows of cells in case one of them becomes shaded.
While this allows for the rest of the panel to operate unhindered (and the shaded cell won't become damaged from the huge amounts of current trying to flow through), it also means that a single shaded cell can knock out at minimum an entire row, but with cheaper panels (which don't use nearly as many diodes) easily a third of the panel and thus drop the voltage considerably.
Figure 3: Power/Voltage curve of a solar system, with both shaded and unshaded curves. By Staberder - Own work, CC BY-SA 4.0.
MPPTs exploit Ohm's law. By utilizing a DC-DC converter, it can freely vary the apparent resistance (and thus the amount of voltage) of the system and, per the aforementioned law, a change in voltage will induce a change in current as well. By careful modification of these values, the MPPT ensures that the panel is always operating at the exact voltage that generates the greatest power and thus delivers the highest wattage based on the current conditions.
And as an added bonus, MPPTs can even account for things like the composition of your battery and make sure it only ever charges with safe parameters of input, as certain materials require vastly different "charge profiles" (i.e. mix of voltage and current) to operate safely.
Cabling and adapters: Solar panels generally use MC4 connectors. These are deep plugs, which slot tightly into each other, providing a good amount of insulation against water and dust as these panels and cables are expected to be exposed to the weather all year long.
However, power stations (see below) generally do not come with MC4 plugs. Rather they are most often equipped with XT60 connectors. These are easiest to recognise by their unique shape and yellow / orange coloration. From what I understand, these were originally made to connect car parts, but I couldn't really find any good sources on them.
Generally (unless you're doing things fully DIY, in which case you don't need this explanation in the first place) you will likely need an MC4-XT60 adapter and these are usually not part of the package. Some solar panels do come with them, but you're largely expected to pick them up yourself. From what I've found, they are surprisingly expensive.
Power station: If you don't want to DIY everything, you can also get a power station, which will serve as your battery, inverter, and MPPT at the same time.
On one hand going this route is much easier as you get a complete device that you just have to plug in and use. It is also far safer for a beginner, because you don't have to measure things out and make sure everything is properly grounded, insulated, and well integrated. Finally, you also get a warranty and even a couple of shiny toys in the form of a companion app that can allow you to monitor and manipulate the station from afar. This is the option I ultimately went with.
On the other hand, it's usually more expensive to buy a complete solution compared to building one from parts. I've seen videos where people mentioned that they can easily get double or nearly triple the capacity from building their batteries from scratch for the same price as a pre-built station would cost. You also learn a lot more this way and can swap out any parts if you want to upgrade, which is not something a complete station can usually accommodate.
3. My particular setup
With the basics out of the way, here's an itemized list of everything I've bought for this "experiment" / pet project along with my comments and reasoning for them. The list will start of small and get extended as new parts are introduced.
3.1. Power station
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| EcoFlow RIVER 3 Plus | 107380 | 275 |
| Sum | 107380 | 275 |
My first purchase was an EcoFlow RIVER 3 Plus (which I'll refer to as "RIVER 3" for the rest of the post as it's a mouthful). This is a surprisingly cheap power station, that can consume 220W of solar power, which pretty much the highest I found in this price category. There are stations that can be charged with 300W or even as high as 1000W, but seeing how my PC rarely goes above 200W, I figured 220W should be plenty for the time being.
Note from the future: As it later turns out, it wasn't quite enough.
I'll touch on the "why" later, but first I'd still like to make it clear that I think the RIVER 3 is an excellent product if you don't have major capacity needs (you only really need an uninterruptible power supply or want to run minor appliances in places where you have no access to electricity) or if you are able to get good solar coverage for most of your day.
It's small, cheap, and has a quite impressive phone application with a lot of useful data and controls. Furthermore it's easily lightweight enough to carry around even in a backpack if you happen to need much more power than a powerbank could provide.
3.2. Solar panel
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| EcoFlow RIVER 3 Plus | 107380 | 275 |
| 220W Solar Panel | 46790 | 120 |
| Sum | 154170 | 395 |
I was originally going for a 220W panel by VOLT Polska, since it felt logical that I should match my expected wattage to my power station's. I also liked, that the panel itself is relatively "small" (147x67 centimetres). The price itself was good too, at only around 120€, so I placed an order and started waiting.
However, not a whole day has passed and I quickly realized I may have been a bit too hasty. The more I continued to read about the topic, the more it seemed like it may not be such a good idea to perfectly match the two ratings, because you only really get 4-6 hours of max power. Which means the moment those ideal conditions pass, there is nothing guaranteeing my panel's output could keep up with the PC's requirements.
Figure 4: My current solar-setup. The panel stands on stacked bricks, with a piece of cardboard taking the brunt of its weight, so that the cables are unburdened.
Instead, I learned what one should do is to "overpanel". What this means is to find a panel (or multiple) whose voltage (and preferably current too) is within the limits and its wattage above the power station's maximum, so that even if the panel is operating only at a fraction of of its full strength, you're still getting as much power as your station can take.4
I am lucky enough to live in a home with a relatively large garden, so I had enough space for a much larger panel than I was originally going to go with. This is good news for my budget, because solar panels tend to have an inverse relation between size and price. This only makes sense: The more panels you can place in the same area, the more power you can generate.
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| EcoFlow RIVER 3 Plus | 107380 | 275 |
| 410W Solar Panel | 24999 | 64 |
| Sum | 132379 | 339 |
Because of this I was able to find a panel from RISEN Energy that is rated for a whopping 410W and which costs only half the price of the previous one at 64€. Excellent deal in almost all aspects, with the only ugly detail being that this panel is a monster in terms of size: 176x110 centimetres. A simple napkin-calculation tells me that it's almost 1.2m2 bigger than the old candidate.
I'm not a big car guy in any meanings of the words, so I only drive an old 20+ years old Ford Fiesta. It's a tiny car, but I figured I don't need no shipping for my panel, "I'll just put the panel in the trunk and drive it home!" Yeah… A tip for future generations: A 2003 Ford Fiesta cannot accommodate a solar panel of this size.
Don't ask me how I learned.
3.3. Cabling
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| EcoFlow RIVER 3 Plus | 107380 | 275 |
| 410W Solar Panel | 24999 | 64 |
| MC4-XT60i adaptor | 9489 | 24 |
| Sum | 141868 | 363 |
I also found out that the RIVER 3 does not come with a solar cable, which was a little frustrating to learn, but I quickly found a store where such an adaptor is sold. It came in three different lengths, 2.5m, 3.5m, and 5m.
I ultimately picked the smallest, 2.5m one. Anything over that would've been overkill, since I was going to place the solar panel right next to my window anyway, which is right next to where I intended to store the battery.
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| EcoFlow RIVER 3 Plus | 107380 | 275 |
| 410W Solar Panel | 24999 | 64 |
| MC4-XT60i adaptor | 9489 | 24 |
| MC4 "Super-Flat" cord | 8080 | 21 |
| Sum | 149948 | 384 |
I also ended up buying a so-called "Super-Flat" cable, which allows you to run it under either a door or (as in my case) through a window-frame without needing to cut a hole.
Figure 5: The cable slips under the window's insulating foam, allowing for the window itself to stay closed.
It was a bit annoying to dish out another 20€ or so, but I figured if this whole thing doesn't pan out, I'd be left with a hole on my window-frame, which (even if small) would likely negatively affect the comfort of my room.
4. The initial results
It was late in the afternoon when I finally received all the necessary parts. I was very eager to try things out, but when we plugged everything in and I excitedly checked the RIVER 3's display, I was a little horrified to find that the panel (which, to remind you, was supposedly able to output 410W in ideal conditions) was only giving me a measly five watts. What's worse, I also saw some weird discoloration on the solar cells and the protective cardboard cover that the panel came in had a footprint on it, so I immediately started thinking of the worst. Not to mention how I'd get the panel back to the store.
Still, we figured it'd be worth waiting at least till the next day to see if the low power is just caused by the lethargic weather or if we really received a malfunctioning panel.
The night passed and during the next morning, to my great relief and joy, I saw that the station was charging at 30W and the amount was rapidly climbing in tandem with the sun. I messed with the panel's position and angle a little and soon enough (around 10-11AM) I started getting a constant 220W, the maximum the RIVER 3 is capable of receiving.
Figure 6: The RIVER 3 in operation at peak performance.
I waited until the power station was fully charged and then disconnected my PC and monitor from the grid and switched them over to the battery.
At first the PC didn't boot. Even after I pressed the button a couple of times, it seemed completely dead. This was a bit puzzling, but then I realized I just needed to turn on the station's "X-Boost" mode (which allows power surges of up to 1200W) and suddenly it sprang into life without issue.
My assumption is that during boot the PC draws in way more electricity and (without the special mode enabled) the resulting load may have been so large, that the power station thought it's an electric fault and blocked it with surge protection.
| Weather condition | Time of day | Minimum wattage | Maximum wattage |
|---|---|---|---|
| Direct, strong sunlight | 9AM - 14PM | 190 | 220 |
| Early afternoon | 14PM - 15PM | 30 | 60 |
| Cloudy / indirect sunlight | 16PM - 17PM | 15 | 40 |
| Minimal sunlight / after sundown | 17PM - 9AM | 0 | 5 |
4.1. Shortcomings of the RIVER 3
Here's the list of negatives I've experienced using the RIVER 3:
While the mobile app of the RIVER 3 is far superior to the replacement I went with, I really don't like that it is an always online IoT device.
I'm not super paranoid or anything, nor do I necessarily think that someone could hack the home network through a power station, but it's still not really nice that it's always out there listening.
In an ideal world you'd have a physical switch whether you want IoT capabilities or not (the replacement has this) and pairing should work without any online account necessary, with purely local authentication (think the LED screen showing a code that you need to enter on your phone or similar).
Charging with 220W proved to be not nearly enough. While I do get many generally sunny hours, the panel is only in direct sunlight for a fraction of these.
Therefore, because only these hours provide the maximum power the panel is capable of, I am forced to utilize this period to get as much energy as I can. And the fact that my station "artificially" caps my panel at pretty much half it should be capable of kinda sucks and severely limits my daily net wattage.
But even if it was capable of collecting more from the panel the biggest bottleneck of this setup was the size of the battery inside the RIVER 3.
Much of the 220W was wasted at peak hours as the station was already fully charged and, once the low-light hours arrived, the battery's 268Wh worth of capacity could only power my PC for just about two hours.
Now, this didn't mean I was out of luck in the afternoon, because the RIVER 3 comes with a feature that allows the user to set a certain percentage below which the station switches over to mains. I set this to 30% capacity, to ensure whatever happens I still have some power left.
The rub was, however, that with this setup I wasn't fully "sun-sufficient".
5. A hasty, but needed upgrade
This didn't sit too well with me. The idea that I'd pretty much just waste a large share of the power I generate because the battery was too small felt like a half-victory at best and an annoying complete failure to accomplish my original goal at worst.
I quickly looked up how much it'd cost to get an expansion battery for the RIVER 3, but what I found to my horror is that it'd actually cost more than the base device itself. It goes without saying, that it would have also provided an almost 3x increase to my max stored wattage (bringing it from 268Wh to 840Wh), but at that point it genuinely felt like it'd make sense to simply look for a device that can both store more power and take more input from my panel at the same time.
Figure 7: The Anker SOLIX C1000 sitting on a tiny chair. It provides power to two PCs at the same time.
And, if I wanted to be that, there was little better choice (with my budget and circumstances anyway) than the Anker SOLIX C1000, which is a beast on a whole another level. Not only is its capacity near five times the RIVER 3's (meaning the Anker outdoes it even with the extension battery's additional storage), but it can also be charged with up to 600W of solar.
Obviously it's also quite a bit more expensive (costing about 2.5x the price of the RIVER 3), but at this point, I was pretty much all-in:
| Item | Price (HUF) | Price (EUR)3 |
|---|---|---|
| Anker SOLIX C1000 | 240000 | 615 |
| 410W Solar Panel | 24999 | 64 |
| MC4-XT60i adaptor | 9489 | 24 |
| MC4 "Super-Flat" cord | 8080 | 21 |
| Sum | 282568 | 724 |
Sadly this does mean I almost doubled my initial budget, but 725€ is still not exactly a major expense compared to a proper solar setup.
However, the changes are immediately obvious and almost all positive. Due to the size of the new battery, I'm able to run not only my own but also my partner's PC and by the station's own measurements, both should be good for nearly the whole day without accounting for the solar input.
| Weather condition | Time of day | Minimum wattage | Maximum wattage |
|---|---|---|---|
| Dawn | 6AM - 7AM | 50 | 100 |
| Early morning | 7AM - 8AM | 150 | 180 |
| Peak sunlight hours | 9AM - 14PM | 200 | 350 |
| Afternoon / Past peak hours | 14PM - 18PM | 40 | 100 |
| Minimal sunlight / after sundown | 18PM - 6AM | 0 | 10 |
A quick note about my results: Look, this is the n=1 measurement of a single guy living in Budapest, taking random samples of his solar setup in a single season and in a very short time period.
Please do not use these numbers to guesstimate your expected production / setup needs. Solar depends on a ton of variables, including sunny hours, placement of your panels, temperature, latitude, materials used, length of your cabling, quality of your inverter, etc.
My experiences might give you an alright idea what to expect, if you live in similar conditions, but that's about it. If you get a hankering for that sweet free electricity, please spend your first week just reading up on stuff (including laws) before jumping in. In the best case, you just gave yourself a huge head start, in the worst case you spared yourself a lot of headache and several hundred euros.
As for the solar, at the time of writing I was able to get a consistent 300W and above through the sunny hours, which is a night-and-day change compared to the 180W averages I've had with the RIVER 3. While it is still not quite the 410W you see on the tin, I'm not disappointed. As it turns out manufacturers generally provide two different metrics:
Output at Standard Test Conditions (or STC): Basically ideal, lab-made conditions where the panel can deliver as much power as it's able. Think measuring what power is generated when a very bright light is aimed dead-on a reasonably cold panel.
STC serves as a good measurement at the maximum possible output you can expect (this can be handy to pick the right wires and safety mechanisms) and it is the one number you'll definitely see on all solar panels, no matter their manufacturer or place of origin. However, it does not necessarily give a realistic idea about the panel's general performance.
Output at Nominal Module Operating Temperature (or NMOT): What STC fails to account for is that the output of solar panels has an inverse relationship with their temperature.
This means the hotter your panel gets (and it will get hotter, after all, the sun is shining on it all day long), the less power you'll be able to harvest.
Solar.com found that for every degree Celsius above 25˚C, your panel's efficiency will drop by around 0.3-0.5%. Which means on a hot summer, where your panel might heat up to about 35˚C, you can expect a 3-5% drop in efficiency. Considering most panels generally operate between 20-25% efficiency, this drop can be significant.
NMOT is the temperature which your panel is expected to reach under the conditions of 20˚C ambient temperature, slight wind, and 800W/m2 irradiance. In my panel's case, this is 44˚C and the manufacturer reports a 0.34% drop in power per degree. Therefore, if I understand things correctly, I can expect a
(44-25) * 0.34% ≈ 6.5%drop in efficiency at peak temperature.By RISEN's own reckoning this means the panel will provide in general about 310W if I both get a ton of sunlight and high temperatures. Indeed, while I was able to get higher wattage already (even hitting higher than 410W for very brief periods), my general sustained peak wattage does seem to be around 325-350W.
Beyond the station itself, the package also came with a parallel-connection MC4 cable, so if I ever want to upgrade my setup by buying a second panel, that's one less item on the list. It's by no means a must-have nor is it nearly the most expensive part, but it's a nice gesture from the company, especially compared to the RIVER 3's bare-bones accessories.
In terms of non-solar charging, the station is able to be charged by a whopping 1000W using a wall socket (1300W if you enable "UltraFast charging", though the manufacturer warns against using it too much as it can harm the battery in the long run), which is a lot more than the RIVER 3's 600W and would charge a completely drained battery in practically an hour (and a bit, due to inefficiencies), which is excellent for its relatively large capacity.
I still intend to stick with solar-only (this was the main reason I switched in the first place), but if I'm ever in a pinch, this kind of charging speed could give me a serious second wind.
5.1. Shortcomings of the Anker SOLIX
This is, however, where the panel's not insignificant list of negatives start too:
You cannot set min-max values for the battery. These are a set of percentages below which the station automatically shuts its outputs down and above which it'll refuse to charge.
People usually set these to 20% and 80%, which give you ample usage, while also serving a dual purpose: The limit allows some leeway in case you are caught with your pants down in a blackout. Even if you were using your battery, you still have 20-30% to keep your machines on for a little while. The other reason is to protect the Lithium-ion cells, as both completely discharging them or filling them to full causes excessive wear.
Anker claims, that the second part is a non-issue, as the battery they use doesn't suffer from such limitations. However, if you happen to use your battery constantly like I do, you have to make sure yourself there's always enough juice in it in case a blackout hits. It's not a complete deal-breaker, but it's a bit of a shame.
You also cannot set the same "only charge from AC if the battery dips below X%" logic as with the RIVER 3.
I've seen posts that this was introduced in the Gen 2 version of the SOLIX C1000, but that's not the one I have and while they're similarly priced, I'm not too keen on sending yet another station back, paying even more, and then waiting for a new one to ship.
Not to mention the Gen 2 is also not expandable like the Gen 1 is, which would currently not be an issue, but if I ever want to upgrade, I would like to have the option.
The phone app's metrics are not nearly as informative. The RIVER 3 counted how many watt-hours you've used and how much you've charged back by type of input, which was perfect for my purposes of seeing how much electricity I can use without relying on the grid. Its charts were also a lot longer and higher-resolution, while the Anker's feels more like an afterthought / gadget to gawk at instead of something you could seriously keep an eye on.
In general, the app is a bit unreliable. The battery's status (percent charged and current input/output) updates slowly and gives no indication that the data you see is outdated. I've had instances where the app said the battery was around 80% charged, while the box itself said 95%. Similarly, the reported charge speeds can vary a lot too. It can take 5-10 seconds for the box to transmit new data and update to the real values, so I tend to just check the box itself.
I also had a couple weird disconnection issues, where I had to reset the box completely and delete then reconnect from the phone app to properly establish a connection between the two again.
- The fan of the box is somewhat louder, especially when you hit peak solar hours and get performance above 300W. I wouldn't necessarily call it annoying (I don't even hear it in headphones), but if you need a quiet accumulator, because you either cannot bear a constant fan noise or your work requires silence, the RIVER 3 wins by a long shot.
The box is also large and heavy. Which is not really a surprise, nor is a fair comparison, considering the difference in capacity, but it's still worth emphasizing.
You can throw the RIVER 3 on your desk and it'll not take too much space, nor is it terribly heavy (at 4.5kgs). You absolutely cannot do the same with the SOLIX, as it's nearly 13kgs heavy and also the size of two shoe boxes on top of each other.
I'd still ultimately classify it well within "easy to carry around" compared to even larger stations, but it definitely requires far more planning where to place it compared to the RIVER 3.
Finally, the box tries to be very smart about the power it wastes, so whenever you're not using one of the ports, be it input or output, it automatically closes them down.5
This sounds great on paper, but during late dusk and early daybreak the panel has juuust enough sunlight to fluctuate between 0 and 1W. The station, being ever so diligent, switches its input port on and off whenever it perceives that electricity starts or stops flowing. And since these ports are isolated using relays which make an audible metallic pop with every actuation, you get about 20 minutes of rhythmic clicking which is both very grating in general and loud enough to wake me whenever I forget to pull the cord.6
Apparently this is known and "expected" behaviour. I intend to fix it by purchasing a circuit breaker (which will also serve as a convenient fuse to protect the battery) and just turn it off at night and thus prevent these blips, but I really wish this was as convenient as with the RIVER 3 which just completely avoids the issue by not reacting to such low inputs.
6. Conclusion
And yet, despite all these issues, I still think I made the right choice and, if my rough calculations are correct, I'll be able to save around 60-80kWh a month by driving the two PCs by solar.
Now, 60-80kWh might not be a life-changing amount, but considering that building and owning this project gave me a lot of joy and atop that it's paying for itself (even if extremely slowly) I think it was more than worth it.
Another interesting side-effect is that I'm feeling a lot more conscious about my PC usage. Now, I'm not expecting some miracle here, where I'll reconnect with nature and barely if ever turn on my PC, I am fairly dependent on my computer. But turning electricity from something you don't ever think about into something that I have a nominally "finite" supply of gives me at least some pause before turning it on and incentivizes me to take longer breaks to allow for the battery to charge more / faster.
Finally, I learned a lot about electricity while working on this post and project. While I did have physics classes both in middle school and university (also an actual electrics class in the latter), the materials there sadly mostly entered in one ear and left through the other. I don't think it was the fault of my teachers, more me being a bit too checked out to care. Now that I'm learning at my own leisure and interest, I find things are sticking a lot better and I'm building intuition for things I kinda-sorta understood on a surface-level before, but never to the extent to really comprehend what's going on. Physics is actually really cool!
Take care and thanks for reading!
Footnotes:
While the power company is obligated by law to buy your excess solar power, they only pay about 5 HUF / kWh. At the time of writing that is just barely above a single euro cent.
Assuming it'd cost about 10.000€ for a complete solar setup for a home (there are grants you can apply for, but for the sake of simplicity, let's ignore those) and about 40 kWh / day excess, it'd take you 68 years to make that money back without accounting for your power bill.
And if you do account for that (our household generally pays about 310-350€ a year for electricity) it'd still take 30 years to make your money back. Which far less terrible, but is still a far cry from the US's 5-10 years.
And if you're curious how to turn AC back into DC, look no further than FULL BRIDGE RECTIFIERS.
Assuming 390:1 HUFEUR exchange rate.
Every article about overpaneling will stress this and so will I:
It is generally okay to go over the current limit, because (besides malfunctions) your power station will only ever pull as much as it's able to and the rest is just wasted. Do consider fuses though, they're far easier and cheaper to replace in case of an issue.
It is NOT okay to go over the voltage limit. If current is "pull", voltage is "push" and it will push as hard as it can. Meaning at best you'll blow your station's internals and waste a bunch of money, at worst you'll burn down your home. Electronics are not a toy.
More specifically input ports are opened and close automatically always, while output ports are fully manual by default, but you can set up either one-time "shut it off after X hours" or general "shut it off after X minutes/hours of inactivity" timers for them.
Despite my annoyance, I'm more than glad that this feature exists, even if I think it could be implemented in a smarter way.
Solar panels can technically cause "backflow", meaning electricity flows backwards and is wasted through the panel in the form of heat and (invisible) light. By physically severing the connection, such issues are avoided.
I just wish they implemented some sort of exponential retry. I.e. if you freshly plug in something, it immediately latches and tries to start charging. If it finds power, then that's that. If it doesn't, it retries a second later, then two seconds later, then four, etc. until it reaches some sane upper limit or the user manually pulls out and reconnects the cable.
This would ultimately still allow patchy inputs, still protect against backflow and would solve the constant annoying popping.