Grid-connected is an entirely different ball game. You will not see any open source projects there, or at least not any that anyone will want to use.
Let's think about why not. Anything grid-connected, you REALLY want a licensed electrician to plan and install. And competent electricians will NOT go anywhere near a piece of equipment that is not UL certified. A company producing equipment is NOT going to go through the expense of getting UL certified and then just release their design, PCB, and schematics for free.
And I want to be clear that I am a strong proponent of open source hardware, there are just certain situations where the incentives in reality just don't line up. This is one of them.
Have you heard of balcony solar?
It's a solar panel, a microinverter, and a standard wall plug. It doesn't need an electrician to install any more than anything else, you just plug it in. Outlets work both ways.
LibreSolar doesn't seem to be working on any inverters, but a complete open source system like this would be great.
Also, there's no need for a transfer switch in any grid-tied system, whether plug-in or hard-wired. Grid-tied inverters shut off automatically if there's no grid frequency to sync to.
"Transfer switch" refers to a specific kind of switch that transfers load between two sources. There is only one source (the panels) and one load (the grid) on a grid-tied inverter, so what you're saying does not make sense.
There are more complicated solar setups that do involve transfer switches, but they are not applicable to the balcony solar use case and remain uncommon even for hardwired rooftop solar.
And indeed these are uncommon, mostly because they tend to be more serious devices. Victron and formerly Xantrex make nice ones, but the inverter alone probably costs more than a complete balcony solar installation.
For the solar balcony and more common rooftop solar setups there is only a simple disconnect, but both a transfer switch and a disconnect are the same thing: a (usually beefy) relay, but the transfer switch variety switches your house between the inverter and the grid whereas the disconnect just physically disconnects the inverters output. The downside of that setup is that if there is no grid but you do have solar that you still have no power.
Most of these wouldn't be able to power anything but the smallest installations anyway (300 W or so, typically), and don't have a battery to store any excess (as if there would be any...).
As soon as you add a battery it makes good sense to use the transfer switch: you can disconnect from the grid but the inverter can keep running to power your house and if you're lucky the solar will replenish it fast enough during the day that you can hold over for a while.
The big rooftop inverter that I have has a built in transfer switch but I'm not using it right now simply because I don't have a good way to route the wiring to and from the inverter. It is stuck in my garage with the main distribution panel on the other side of the house. In my old house in Canada that was all designed from scratch and there we had the house entirely off-grid with the transfer switch hooked to a genset if the power was out for longer than the battery could sustain us (48 KWh so that usually was good for a couple of days).
> the transfer switch variety switches your house between the inverter and the grid whereas the disconnect just physically disconnects the inverters output
> That simply is a transfer switch that is built in.
Then in this comment:
> but both a transfer switch and a disconnect are the same thing
These statements are false, and the fact that the second one was written even after being corrected once makes me think they still do not understand.
This user appears to blast paragraphs upon paragraphs of irrelevant noise at anyone who responds to them so that either the comment has internal conflicts (as you noticed) or any criticism seems nitpicky.
The units likely have "protected outlets" too that likely use an internal transfer relay to disconnect from the grid side, but at 15/20A it doesn't have to be terribly beefy.
I have a similar situation here but at much higher power levels, a single underground cable from my garage to the house carrying 16A tri-phase and a whole raft of consumers in the garage itself. There too there is the potential for overload with both consumers and producers on the same cable. The solution there was to have a secondary distribution panel, breakers on both sides of the cables, for the consumers and for the inverter guaranteeing that none of the wiring in the panel or to the house or the consumers ever exceeds its rating.
This was by far the most cost effective solution, saved adding another ground cable and relieves the main distribution panel of a lot of current going in and out of the garage.
Just curious - are you exporting power or zero export with a current monitor upstream of the main panel? Also same question regarding off-grid operation and a transfer switch ahead of the main panel.
I don't know what tri-phase breakers cost in NL but the second panel and feed-in breaker sound like the straightforward solution in the US too. Our wires cost considerably more, and we don't even have RCD in the breakers you'd use for that.
Hehe, ok. I think that's your insurance company calling on the other line.
> are you exporting power
Exporting 12 MWh / year or thereabouts.
> or zero export with a current monitor upstream of the main panel?
There is a current monitor (a Shelly tri-phase one), right now it is still economically viable to do so (though the utility companies are trying what they can to dissuade you by changing the deal through politics). If it is no longer then I will just install a battery and disconnect from the grid for the summer months.
And I'm not using a transfer switch because I don't have a battery to stabilize the system.
Well, technically those breakers are a manual transfer switch, only it is broken up into two halves and I can just disconnect the mains feed and run in island mode but I would still need to install a battery and a charger. House mains breaker off, solar on would be the house running entirely off the local stuff, I just don't trust that inverter without a battery behind it to be able to react quickly enough to load changes and by default it is set up to disconnect if the grid goes off, so you'd have to manually override that. The main issue with it being two halves is that you can not guarantee that the house net is in-sync with the grid at the moment you make the switch and that's a bad idea with a system this powerful, so I'd definitely get a proper automated one if I intended to do this for real, otherwise you might cause a load spike which could trip breakers and annoy the neighbors.
Right now I can't switch that on or off under load anyway because the large inverter would simply disconnect as well.
Tri phase breaker of the right amperage was about 150 bucks.
If I were to do this I would probably get a complete set from Victron, their stuff is amazingly well engineered, but if these open source people are going to make an inverter/charger combo then I might go for that and add a another manufacturers automated transfer switch.
An inverter is, complexity wise, not that much harder than a large switching power supply, there is some more instrumentation and some more rules but it isn't super difficult. It is much harder to make one that is commercially viable because those guys all cut corners to stay competitive. Ironically a proper case is probably the hardest part, there are also some larger inductors that might be tricky to source. And if you were to design one you should probably make the low voltage stuff (UI, CPU) on a completely separate board from the line voltage stuff and go for tri-phase right away because it is so much cleaner. Bonus points for modularity of the output stage.
(then again maybe someday I'll hit some wall with off the shelf MPPTs and find myself wanting to go down that rabbit hole lolol. but honestly AC coupling seems cleaner in terms of things like fault protection on longer runs - fault on a stiff mains circuit -> breaker will trip. Fault on a circuit where the current/power is intrinsically limited to what the solar panels can supply -> ???)
I asked about the transfer switch / monitoring because I've looked at the same problem here, first with a generator now with solar. Incoming power service is on the complete opposite side of the house from where I really want the power handling gear. The manual two breaker thing is practical (for a generator at least), but not code compliant here (no positive lockout).
I would think Victron would have an option for a remote transfer (/disconnect) switch, but I haven't really looked into it yet. It would still have to get the grid phase timing somehow to line them up before connecting, so something more than merely a dumb contactor.
Same in my house.
> It would still have to get the grid phase timing somehow to line them up before connecting, so something more than merely a dumb contactor.
Then you'll want a synchronous one. They match phase before making the switch, which is one reason why it's nice if you use one tied to your inverter, which already has the capability to steer its phase to match the grid.
But really an independent transfer switch wouldn't actually fully solve the problem - power at the main incoming service panel would still have to blip off-on for the downstream Victron to see the grid loss, disconnect from grid with the external transfer switch, disable anti-islanding, and then re-close the relay to backfeed up its own AC-IN. And even that would be a bit dodgy relative to proper certification for anti-islanding.
What one really needs for this topology is to move both the contactor and the current sense normally in the Victron, to the location of the incoming service panel. Which is why I was wondering aloud if they had a solution to do this, and coordinate with the inverter to maintain export rules, phase matching, etc.
I'm guessing the common answer is just run two sets of wires, as with generators.
In the United States, you are one small piece of sheet metal[1] away from any number of interesting power set ups on your side of the utility.
This means a combination of two circuit breakers is now your transfer switch. This is legal with all utilities and NEC compliant, etc.
If you are willing to sacrifice perfectly uninterrupted power, you can dramatically simplify your grid tie - and open up many other possibilities on your side of the physical interlock.
[1] https://www.daierswitches.com/products/sd-200vl-generator-in...
This is much easier to do if it is all integrated into the inverter itself, but that makes for an awkward bunch of wiring, because the inverters are typically not situated right next to the entry point for the grid connection. I'd have to rewire my distribution hookup completely for that kind of functionality, or to have a remote controlled disconnect while the inverter keeps feeding the distribution panel.
A transfer switch is much more applicable to emergency power or ship/shore power situations where you only use one power source at the time. For solar it is normally all on or all off or solar+battery(+wind) on all the time and backfeeding into the grid when it is available and grid power when solar+wind+battery are not available.
This can get complex in a hurry, fortunately there are a number of companies that make excellent components for these applications that you can just order and hook up and call it a day, without ever having to worry if your fancy setup has the right break-before-make order and whether or not it is code compliant. And they're not expensive compared to the rest of the gear you'll need.
Competent electricians are licensed professionals who (1) stand to make money on selling gear and (2) have customers that hire them simply because they don't want the hassle or the liability. Obviously a licensed professional is not going to install your home brew inverter, but at the same time if you can design a homebrew inverter you probably don't need a licensed professional anyway.
I've rewired lots of homes and have never had an issue with any of this and designed my first inverter when I was 17 to power my room when my betters decided I should go to sleep and cut the power.
This stuff is not magic. If someone designs a modern open source inverter I'm definitely going to build and install it. Fortunately insurance companies here are reasonable: if your homebrew device wasn't the cause of the mishap then you are still insured.
The one thing they are very strict about is gas, because there is no such thing as a 'fuse for gas'. But if you've properly designed and fused your gear then it should be no less safe than any other grid connected device, even if the magic UL or TUV mark isn't there.
The big one is EMI, that can be hard to get right and you need some gear for this, which is why it pays off to pool the money for an open source design to be certified. And once certified of course the design is 'type approved' and frozen, so you can't change any of the hardware without going through recertification. This is expensive, but if you don't do it every other week should still be well within the means of a properly set up open source project.
Why the fearmongering? It's not as if we're 12 here.
The main concern is exporting to a downed grid that line-workers are trying to restore.
If you're grid connected and see valid phase on the input for a certain amount of time of matching phase and measuring voltage you can provisionally connect at exactly that phase and voltage but without injecting power. After that you are allowed to slowly ramp up your output by leading the phase (while raising the voltage within certain limits) as long as you observe the effect that you have on the grid. If the grid phase drops away or there is any other anomaly (such as a voltage drop or rise of more than x V/s you are required to immediately disconnect, there are many other disconnect requirements but that's the main one with respect to line worker safety.
Three disconnects within a short period of time = no reconnect attempts for a much longer time. If the situation persists that's a failure and you are no longer allowed to connect to the grid until there has been an intervention and an inverter reset.
If your inverter is of the islanding variety then the rules are slightly different, then the transfer switch only gets energized when you match voltage and frequency but in the meantime the (usually battery backed up) inverter can supply local consumers.
By the time you come up with the idea of rolling your own inverter you have either become familiar with the requirements (which differ from region to region, and which in a properly designed inverter are mostly a matter of tweaking firmware parameters) or you will have to do so because you realize your responsibilities.
Anybody up for this kind of project will with a high degree of likelihood have the required knowledge because that knowledge is a lot simpler to acquire than the knowledge to build an inverter that isn't going to result in you being laughed out of the room when your EE buddies come look at your creation.
I would expect you to do a better job than 95% of the imported ones that I've taken apart and which all had massive shortcuts taken, good enough to pass first inspection and a year into warranty, not good enough for long term safe deployment. This ranges from unsuitable connectors, low quality inductors, even lower quality relays, undersized FET boards, insufficient cooling, bad cast aluminum housings, in general bad housings (not rodent and/or insect proof) and so on.
I'd prefer to just put the DIY inverter behind a transfer switch (with an adequate battery bank and maybe a small propane generator)... with the grid as emergency fallback.
https://fosdem.org/2026/schedule/event/7Q9HMF-open-solar-pow...
2025 for archive:
https://archive.fosdem.org/2025/schedule/event/fosdem-2025-6...
ZephyrOS: https://github.com/zephyrproject-rtos/zephyr
Would there be value in modeling this system with TLA+?
Why build another open product?
There are a few GitHub topics for solar electricity:
solar: https://github.com/topics/solar
photovoltaic: https://github.com/topics/photovoltaic
pv: https://github.com/topics/pv
battery-management: https://github.com/topics/battery-management
ups-management: https://github.com/topics/ups-management
inverter: https://github.com/topics/inverter
Photovoltaic system: https://en.wikipedia.org/wiki/Photovoltaic_system
I got solar installed by the local power company and while it's well done and was a great deal regarding the price, the inverter stats are locked behind a really terrible app. At least there isn't a subscription cost but I wouldn't be surprised if they add one someday.
Would gladly pay more for fully open and serviceable replacement.
Agreed - a lot of the inverters do some real BS moves around data management clearly a way to extract more value in a subscription mode. Its mind numbingly frustrating.
In Australia you can go from the notion of "hmm I might want to put some solar on my roof" to having it, approved, installed and running in a couple of weeks. Safely and legally. In the US you lose months at best. Half a year is nothing. And everybody involved wants to make a profit so that's where all the cost goes. The crazy thing is that even with all that cost and bloat it's still worth doing. Imagine how good it would be if the US could figure out how to do stuff efficiently.
Both cost and complexity have to come down there. A lot of the friction in the system is a combination of the fossil fuel industry lobbying very successfully against anything to do with renewables, local energy monopolists resisting change and the notion of competition from their own customers (or any form of competition), and inept politicians coming up with ways to keep those happy.
And as an EU person, it's not that much better here. Better than the US. But we can do better.
This project isn't being marketd to people who call up a company to white-glove the whole-home installation end-to-end. This is for DIYers who have enough knowledge to tinker with self-designed solar projects but not the EE degree required to engineer some of the more specialized equipment themselves.
I'll give a good example: I use solar to power a ham radio station for a weekend in the summer. However, nearly ALL of the equipment you can buy for the production and storage of solar power emits some degree of radio-frequency interference, which is bad when your whole goal is to power a very sensitive radio.
When it comes to charge controllers in particular, there are exactly two companies that claim to make RFI-quiet MPPT controllers. One has mixed reviews (some people say they work great, some say they are not any better than anything else), and one is very good but also very expensive for what you get. So, more open design and community feedback from people like me might get the cost of a reliable RFI-free charge controller down to where it should be.
Safety is important of course. But people are being very selectively paranoid about this stuff.
Same with EV batteries vs. ICE engines. Combustion engines that intentionally explode highly flammable fuels sometimes catch fire (surprise). People think nothing of parking them in garages connected to their house. Sometimes those cars are quite old. Maybe the wiring is a bit dodgy. Or the fuel hose a bit dried out and leaky. And the guy you pay to fix your old car is maybe not a trained professional. Or something else goes wrong. In short, it's extremely common for vehicles to catch fire (most common reason for fire trucks to get called to a scene). And lots of people die in vehicle fire related incidents. Almost all of which are good old ICE cars. Yet all people talk about is battery fires in EVs. Which are quite rare and pretty much never happen at all for certain newer battery types (e.g. LFP, sodium ion).
Libre does mean free as in freedom, but it also means available and released from prison.
Also, it comes from French and Spanish [https://www.gnu.org/philosophy/free-sw.en.html], it's the same word and same grammar, subject before adjective.
Not a chance. Language is a tool, not some kind of holy house. If the meaning comes across that's already more than you could hope for.
The whole idea that language is something to be condensed and frozen is relatively new, it used to be much more fluid which made it harder to have a decisive set of words (sorry, Scrabblers (tm)) but language could adapt faster to circumstances. What we consider the 'final form' of all of these languages isn't final at all, it is just what we are used to and more of a convention than anything else.
Some countries (including mine, unfortunately) have groups that will go out of their way to harass people to use the 'correct form' ignoring that what those forms are is subject to change and that not everybody grew up with the currently agreed upon preferential way to spell things, and meaning is so much more complicated than spelling.
I think the easiest way to deal with these kind of frustrations is to see them as compliments: at least someone was aware enough of your language to name something in it, that's more than you could say for many other languages, if only because the scripts are different.
- dc motor conversions for air conditioners and voltage controllers that can adapt to multiple panel types to drive it. There's a lot to be gained without going all-in an inverter. What if I stayed on grid but heavily offset my heating / cooling bill by having a wall mount unit that was free to run during daylight hours on a completely separate circuit?
- conversion kit to make the wall mount unit a heater in the winter months
- a DC home system for select appliances such as lights, computers, or even refrigerators. This requires more precise voltage regulators because DC is more finicky when you add / remove loads. But you save some losses in efficiency.
- a thermal battery, so my window mount cooler can freeze a solid block of ice all day during the sun, and use it to cool me when I get home. It would be sealed of course. But condensation would still have to be managed. Maybe a hot water tank adapter that uses excess electricity and dumps it in the tank, but not so much to explode. Again, a form of offset, not a replacement for gas. Another thermal battery could be a sewage tank that aborbs heat from the AC unit before it goes down the drain. This would reduce the load of the fan.
- a wind generator that works best in storms. It could dump straight to a heating element and fan indoors. Who wouldn't love free heat during violent storms? Maybe it could have a clutch (mechanical or electric) to tune the load to the wind gusts.
- a solar cooker, maybe with a molten fluid or superheated steam. The latter can go well above 500F so plenty hot enough for cooking. But of course very dangerous so would need a professional device, if it's even possible. But this would allow e.g. restaurants in the summer to offset or replace heating for their kitchens. It pains me to think of how much energy is wasted in a hot desert city to pay for a gas grill and then pay again to pump to he excess heat outside.
This somewhat limits the usefulness of the hardware anywhere you need to be insured, e.g. your house, boat or van.
(Also, Amazon is where most people get their solar equipment these days and you would be amazed how much of it is not UL certified either.)
The real reason this should give you pause is that you don't want your house to burn down regardless of an insurance payout. That is how your incentives remain aligned with the insurance company.
In my experience insurance companies just ask you a bunch of big-picture or specific questions (eg swimming pool), reserve the right to inspect even though they never do, and then jack the rates rather than trying to grok more unknowns.
I've never read an insurance policy where anything like this is explicitly mentioned. I suppose there is a legal path of being criminally charged by fire investigators for having performed unsafe wiring (non-Listed power handling equipment as part of the fixed wiring), and then the insurance company denying you because of that criminality. Or in states where DIY wiring is not illegal, perhaps declaring that the wiring itself is "illegal" (as it goes against the NEC (despite the NEC not being openly published as we generally expect from laws!)) and then hanging their hat on illegality regardless of criminality? But does any of this happen in practice?
Surely if insurance companies were concerned about fires caused by dodgy electronics, they'd address all of the people using non-NRTL-tested GENSYM brands from Amazon et al? They've got no ability to post-facto deny based on this (said devices aren't illegal), but they could surely make it an explicit condition of policies.
If you want to DIY everything you need self-build insurance.
I’m definitely not covered if I were to burn my house down with a dodgy inverter installation.
Many things you still have to get a permit and have it inspected afterwards just like a professional would. But if you skip that no one will ever notice unless it's major structural work (such as a new deck).
That said, if it's electrical and you skip the required inspection and then your house burns down and your work was at fault that might nullify your insurance but I'm not sure.
I wouldn’t run some AC motors, old AC clock, ham radio, or many other things on some generators.
The line is open to interoperation and never defined by the manufacturer. It’s blanket liability avoidance that confuses customers.
I've certainly had my fair share of square wave UPSs and devices that don't work on them.
There are certainly cases where harmonic distortion is a problem for a device. It’s just that everyone is left guessing, and there’s an overblown fear of devices being harmed.
Same goes for all the random Chinese inverters people are buying and installing in their Homes, Boats and Vans. Doesn't seem to stop them.
Of course, there is a rising group of actually really high-quality native brands starting to come out of China - Anker were an early one of these, Xiaomi is a massive one, UGreen is really popular on Amazon etc. These kind of brands actually will be properly certified and are scary good for how cheap they are. So it's changing a lot.
That being said, the Libre Solar components are also meant to be used as the basis for customization (hence, called building blocks). Some of the devices are used with minor modifications in certified commercial products.
Apart from boats/caravans, DC systems are used a lot for rural electrification in the global south. This is also where the communication features of Zephyr RTOS are very important.
The battery stuff is more risky (bringing lithium cells into the picture) but I don't think anyone should be worried by the MPPTs.
($3k would be for "unintentional radiator" device, i.e., not supposed to be a radio, $30k would be for "intentional radiator" device, i.e., supposed to be a radio)
*Testing and tweaking and then sign-off in grown-up labs.
A hobby BMS is usually a bad idea, as most kits from unknown origins prioritized cost over safety. Depending where you live, prior to roof installation there may be additional zoning and signed engineering drawing requirements.
It is not hard to find UL equipment, but expect to pay about another $600 for the BMS. Yet, it is better than a house burning down, and the insurance provider denying coverage.
Have a look at local certified installer companies, and make sure to get some real references in your town. Just like most HVAC companies... some installers are just over priced scams. Some folks claim https://www.pegasussolar.com/ was inexpensive, and might be worth a call. Best regards =3
UL or other certification is a very good idea. They can't automatically deny coverage for lack of certification, but it becomes a much harder fight for you to prove the non-certified equipment wasn't at fault.