Another option is the Enphase products, HTTP binding and JSON transform. Just setting my one up now, there is heaps of very granular data if needed.
2 Likes
I do have Fronius inverter and all works perfect with openHAB
George, what you have done there is just what I am trying to do!
I’m new to openHAB but have managed to configure the Enphase production output. Regarding the inverters I have the http url Thing fetching the json, but I’m struggling to get the correct JSONPATH expression to cope with the multiple inverters.
Are you able to share your solution? Thanks
Hi Simon,
No worries, I will organise my notes over the next day or so and post it.
Cheers
1 Like
Would you share your Node-Red node?
This is interesting as I’m currently trying to decide on a solar system. Does anyone manage to write to their inverter? i.e. to ask the battery to be charged at cheap rate or discharged to grid if export is good?
Catherine
1 Like
I did but battery control abilities depend very much on the inverter model, and are not documented for the most part so it’s still quite a lot of trial & error.
There’s no real charging at cheap rates, even if you have a hourly changing tariff, it’s all way more expensive than to use your self-farmed energy.
I myself have a Sungrow inverter which I like (it’s a 3 phase full house UPS, yeah).
Advertising: I’m commercially offering an energy management system based on openHAB rules
(German only sorry, but I promise it runs on non-German computers, too 
… just use a translator)
This can operate your washing machine, heat pump and basically everything that openHAB can control on cheap PV power.
1 Like
I plan to use RTC Solution . Can you share you Experience please ?
Sorry for replying so late - not logging on on a daily basis. The above mentioned script is running well - pushing it every couple of seconds via node-red exec and MQTT to Openhab.
This is definitely something I’d be interested in. In my ideal setup, the battery would serve a dual function: aside from storing excess solar power, it would serve to displace all grid consumption to the cheapest tariffs, which would be especially useful in winter when the panels would probably not deliver enough power to cover my usage.
I gather some inverters can be made to take power from the battery or the grid as required, but I haven’t seen much about control options for charging the battery.
It seems that the Tesla Powerwall, at least, can be forced to charge from the grid by manipulating the reserve percentage number. But automating this does not seem to be easy:
https://forum.logicmachine.net/showthread.php?tid=3308
It seems an obvious thing to want to do. No idea about other makes of battery.
I’m happily running a Sungrow inverter that allows for battery charge control.
In the EMS I’m selling (see my last post) I charge or discharge it as needed to maximize own consumption. As the EMS is implemented in openHAB rules, anyone can easily add his own rules to charge/discharge based on variable tariffing or anything.
1 Like
At the moment I’m just stumbling around, trying to work out what is even possible. I’m in an apartment building, so installing solar panels may be tricky. Depends. Good to know that the idea is feasible, though.
Another possible use case for the battery is getting rid of gas for hot water. Due to archaic rules in Spain, heating water instantaneously with electricity from the grid is practically impossible (domestic wattage is low and increasing it very expensive), whereas it’s easy with gas - a nonsensical situation in 2022. I wondered if a battery that can deliver 10kW might solve the problem (apparently if you can get lots of power from a battery, that’s allowed). Is it possible to combine the power from the grid and the battery to power appliances, or is it an either/or situation? Could you, say, pull 6kW from a battery and 5kW from the grid to get 11kW?
You won’t be explicitly ‘pulling’ but that’s what will happen anyway in probably any standard setup with a hybrid inverter.
FWIW, I’d get a use water heat pump in that situation. If that has a buffer it will also flatten the curve i.e. peak power needs.
1 Like
The problem with a heat pump is that you need a big water tank. I’m not keen on water tanks because you either limit the amount of hot water you have available, or oversize the tank on the off-chance you might need more. Either way, it’s less efficient than instant heating, because you’re keep water warm that you don’t need for 23 hours out of 24. Another problem is that a lot of water tanks won’t fit where the boiler is currently located.
I have a heat pump but didn’t include hot water for that very reason.
OK, a hybrid inverter. Noted.
Nope that’s comparing apples to oranges. Instant heating is all-electric while a heat pump is giving you an efficiency factor of 3+, compared to that the ‘loss’ in storage is totally negligible.
And that ‘loss’ is really one of those old myths that maybe was true once some 40 yrs ago but was carried over again and again ever since although it’s not true any longer with a decently isolated tank and up to date control technology.
With some OH intelligence you can even control how much water it’ll prepare, minimizing losses.
Heck, you can even use the water buffer as a cheap thermal battery to store excess solar power and that even all throughout the year. I’ve built that capability into my EMS for that very reason.
2 Likes
It seems that the Tesla Powerwall, at least, can be forced to charge from the grid by manipulating the reserve percentage number. But automating this does not seem to be easy
Actually, with OpenHAB rules, automataing this is pretty easy. Setting the reserve higher than the current charge level will grid charge if there is insufficient solar. I now take a Solar Forcast (from solcast.com.au) and modify my reserve % based on what the next day’s solar forecast is.
1 Like
That’s correct, I hadn’t factored the efficiency of the heat pump in. But heat loss from water tanks doesn’t seem to be quite so trivial. Every source I’ve looked at, including a study from a few years back, suggests that heat loss is well over 1 kWh/day for a typical (insulated) water tank. A 3-minute shower at 40 degrees probably requires less than that to heat the water.
In any event, I’m not sure I have the space for a decent-sized tank. I’ll take another look at it.
Which only applies to the remaining water level so numbers in reality also depend on the control’s intelligence how full it’s keeping the tank. Something I bet those studies have been ignoring. There’s not much ‘intelligence’ in classic HVAC controllers and numbers are often misleading as HVAC guys typically fail to think out of their (pretty limited) box.
For example, if they factor in solar energy at all, it’s usually based on direct-solar heating of water. Which is totally different from heat pump generated hot water as you can intelligently control the amount of the latter while for the former you can not. It’s there, it’s what’s filling the tank even in summer when you cannot make good use of it - it’s wasted energy.
A more intelligent system based on PV avoids generating hot water in the first place and puts excess power to the grid.
Nope, that on average takes more than 1 kWh (a little, depends on more factors of course).
A 10 minute shower would probably be the maximum to keep and it’s <= 150l so a tank that size should be more than enough. Ok let’s stop, too off-topic now.
FWIW, electric battery storage efficiency is also only ~90% i.e. 1kWh/day for a typically sized battery.
Of course it’s true that if you have leftover solar power, you can use it to heat the water, rather than returning it to the grid (which will only net your energy usage down to zero here, it won’t pay for any excess power you put into it - at least not yet). I calculated that in summer I would have a substantial surplus of solar power here, even with just 4-5 panels, but then that’s also when heating water uses the least energy. In winter, I would typically get less power from the panels I could feasibly install than I would use.
The basic problem with tanks remains - you always have to massively oversize your tank (to cover the maximum possible usage), or risk running out of hot water. On a typical day, I maybe only use 24 litres, but I’d probably need a 100-litre tank to cover the outlier scenarios.
It’s also true that batteries have losses, too, you’re right.