So I dont know what this looks like in solargo. But in home assistant below is what I see when loadshedding happen. The inverter’s energy mode changes to “off-grid”. It does the same thing when I trip the main switch from eskom, or the breaker infront of the inverter. Yes my inverter has it’s own dedicated breaker on my non-essential db.
When my inverter is in off-grid mode, it functions at it best. Something on my PV strings is not optimal. When there is grid, the inverter doesn;t always max out the strings to charge the battery. But when it goes into offgrid mode, either by me changing the work mode or tripping that main switch or actual loadshedding, it maxes those PV strings to the max to both run the load and charge the battery at the maximum speed it can.
So it really is interesting, to see that someone elses goodwe inverter behaves the polar opposite to mine.
Thanks. This makes sense. A way to kick start the system is to put into off-grid mode. But then I might leave it there. I think the problem then would be that the non-essentials would still work when there’s grid, but wouldn’t get any assistance from the inverter (OK… they also wouldn’t impose any load on the battery).
I also need to spend the time to get HA up and running. I have Pi loaded with HA, but haven’t got any further.
Oh. I just checked. The “Normal (on-grid)” or the opposite does display in SolarGo. But I see you keeping history as well, SolarGo doesn’t do that, and I am distrustful of SEMS.
yeah Home Assistant is great to go back in see what happened at different times.
Here is what the logs look like when going into loadshedding and when going out. this was a 12am to 2am slot. The DoD message is just my automations adjusting DoD according to conditions around how much load there is at the time and expected kWh generation forecasted for the day.
We had loadshedding this morning from 6am to 8am. I think the batteries did well considering wife needed to dry her hair, not realising we in loadshedding.
Well shoot! The power never came back on after the 2am shed, and it’s overcast. So I’m still gathering data, but not what I expected and nothing helpful as regards my current woes.
A guy on the FB Goodwe forum says that he had this problem and that it’s firmware version specific. He says his firmware was rolled back, but he has a newer version than mine.
OK. So my installer’s help desk said that they would refer the matter to Goodwe. They called me back to get the serial number for the inverter, because that’s what Goodwe use to connect.
This morning I noticed I am now running on 2121E. So a firmware update was performed. My installer has a hotline to Goodwe that I don’t. And they don’t even install Goodwe any more (they now do Victron and Deye/Sunsynk). So good on them.
But the window of opportunity is small. Usually, on a sunny day, my battery will be charged by 11:30. Thereafter all the inverter has to do is service the loads from the house. I turned the City’s breaker off just after 11, but the battery hit 100% SOC at 11:20. The power from the panels dropped off then recovers (shown by SEMS, and seen by my on SolarGo). But I didn’t have enough time for a convincing test. I would have liked to have run it for a while in economical mode to see how it reacted.
Taking the load shedding schedule and weather forecasts into account, the next good test window looks like Tuesday morning.
OK… so I got some opportunity today in a morning of changeable weather, load shedding and an unschedule outage.
Scenario is
SOC at 96% at 12:00
Grid goes down at 12:00
Mode is “general”
And it kept on charging the battery. This took about 35 minutes, PV was very variable still, but it DID charge. In that mode. I need to try to get a gap to repeat this in eco mode.
This morning was a better test. Grid down at 7:00, lots of PV. In General mode, at about 8:15, the battery was charging. There was about a kW more PV than load, with the excess going to the battery.
I watched it for a while then switched to economic mode about 8:37. 2 minutes later the grid came back, so I still haven’t been able to check what happens with eco mode. But in general mode, with 2121E, it certainly is charging from PV when there is no grid.
There doesn’t seem to be a nice coincidence of load shedding and good weather for the next couple of days. Load sheds are early morning, when I’d be running of battery anyway, or in the afternoon when the battery is usually charged.
I’ve been wondering about eco mode. My default MO is to leave it on, with one charging slot per day - 15:00 to 16:00. But does that leave it in an ambiguous state for the rest of the day? Doesn’t know whether to charge or discharge or hold steady?
IDK
I also don’t understand the discharge side of eco mode at all.
Discharge is only useful if you have a grid/utility that have different price per time of day, on how much they charge you for using or pushing to the grid. So in SA its useless, but in other countries, the the spot price change, people use it to discharge their batteries into the grid when the price is very high (for both using and pushing to grid) and then they use charge from grid when the price is very low.
So that in-between state you then leave it for the rest of the day, is the so call emulated general mode, but not 100% general mode. So it sounds like, thats when your inverter doesn’t charge batteries from PV when the grid is gone.
Basically, you using eco mode for something different than it was intended for, which is what I described above in my first paragraph. At least, how GoodWe intended eco mode to be used.
I was considering to use eco mode too permanently, but if I am to run in to that same issue you have here, I guess I will rather stay with my current strategy, where I am in general mode most of the time, and rather manage the DoD with home assistant. Its definitely working better out for me.
In the interests of science, I watched my system on Tuesday when he had sunny weather, battery not charged, and load shedding. This was in general mode.
SEMS showed the battery charging. I changed the system into economical mode, but power was restored within 2 minutes and so I can’t say this was a useful test for ecomonical mode.
But I left the system in eco (In the interests of science) and this morning all the stars aligned. Load shedding at 6:00 (6:01 according to the inverter logs) and now look
By 7:45 the PV is double the load and the battery is charging, by 8:15 the PV climbs steeply (this is usual), the load increases a little (usual) but the battery is definitely charging. PV is not just tracking the load.
This with the safety code set to 230 V 50Hz default. I switched it back to South Africa, but the grid came up 2 minutes later, so, again, no useful test. Saturday morning looks like being the next useful test slot.
So this morning…
Grid: Not present
Inverter mode: economical
Safety code: 230V 50Hz default
Conditions: Clear and sunny
Latest. No combination of weather and grid outage has presented an opportunity for testing. But this morning I am running 2525J. A big step, but any step invalidates everything I’ve found so far.
Man I would get annoyed if the firmware was just upgraded without even an email to tell me why and what it was for… Do you have automatic FW update enabled??
Happens only if the installer the inverter is connected to request it, user request it, but sometimes, if the goodwe in said country, may also push updates, however this is the first I hear/read of it for Goodwe SA.
