GreenFields

The time of use settings (including power limit) only apply while the grid is on. Once the grid power goes off, the battery charge/discharge current limits will apply.

I don't think you need more panels. The 24x310W panels (7440W) is plenty considering that you've got spare unused capacity during the afternoons, and we're still in Winter, and you're only using 20kWh per day. Even if you replace your batteries, you'll probably not need to add or replace panels. Neither do I think you need an alternative timer or DC element or whatnot. Just be sure to resolve the shading issues. One of the no-cost options could be to increase the thermostat's temperature so that you have more heat stored to be usable later during the evening. Just be careful of scalding. You could consider also dropping from a 4kW to 3kW element. That will of course increase the heating time, but you could run off solar for maybe an extra half-hour in the afternoon to around 16:30 or later in summer, without switching yet to grid. I wouldn't go as low as a 2kW element for that. For the morning slots, try turning the timer off before the last person starts to use hot water. Perhaps 6am? That is to say, try to avoid re-heating spikes when there's nobody else that will need water next. Other thought with that, you don't need 200litres of hot water for dishwashing and handwashing. Maybe a small 10-20 litre water heater could be useful for those times. Or a dishwasher. In the end, the only way to save electricity during the early-morning time slot is to switch to a heat pump, but it's not clear if the savings will be worth the outlay.

Just a thought - really just taking a chance - try to set your Azimuth to 300 degrees and see how it looks. It might be, if this is a new feature being tested, that the degrees are being counted anti-clockwise rather than clockwise. Basically, I'm wondering whether your panels aren't being interpreted as pointing 60 degrees West rather than to the East.

I'd still be cautious, more curious actually. The current seems to diverge more at the lower irradiation of NOCT and maybe you are more likely to spend more time in the lower ranges.If you have an inverter that clips MPPT current at 13A the difference at STC is moot anyway. Can you post a link to the full spec sheets including the Current/Voltage graphs?

Can't answer, but would suggest to look also at Goodwe for 48V inverters, since they are supposed to have fan-less operation on some models. Perhaps some Goodwe owners can comment.

Personal 2c' worth. I wouldn't even do anything until next month when you've got the second inverter. Just install the inverter and the additional 5x555W panels to it all in one go. Those panels could generate at most around R250 worth of additional power in one month, hardly worth the changes and re-changes, and since you've got around 5.2kW of panels already installed, you're not even going to see that benefit. I'd just put the 5x555W panels straight onto the North-East side a month from now, and keep the existing panels where they are on the first inverter, and maybe later add another 3 of the 335W panel to the existing 2kW string.

This is how I manage it - solar-charging and top-up charge from grid by day, running off battery during peak times, and shifting as much load as I can to off-peak and standard time slots. Compared to the off-peak price of power it becomes less economically viable to install more panels or batteries, so I've settled on remaining grid-interactive, and then being able to cover loadshedding with basic amenities. The problem is just that I've got a small 5kWh battery and 5kW inverter, so it's easy to either drain the battery making dinner, or exceed the power rating of the inverter and then draw in from the grid, but the cost still doesn't justify making changes. Metering is in place as I've had grid-tie since 2015, using GPRS communications to the municipality, so that's barely an afterthought by now.

Show me yours, I'll show you mine. All before VAT. NMBM. Can't complain about the monthly basic charge of R117.7 Can't complain about the Off-Peak R1.92/kWh. Can't complain about Standard-time R2.94/kWh. Peak energy is a less fun R8.56/kWh. I've used 10kWh so far this month at that rate/time, which was unavoidable.

Doesn't look like your domestic is the sole culprit; there's high usage even on days when she is not there. You've given the solar stats right at around 21 June, the Winter Solstice, basically the worst time of the year for solar generation because of the longest night of the year. Before jumping to a decision, can you pull a similar month of stats for September and December? Equinox and Summer Solstice? Off the cuff I think you could do with up to 50% more solar panels, but not before checking how much you could discharge your batteries overnight, because you're not cycling them deeply every day, and not without seeing where you could maybe do better with other efficiency measures (eg. heat pump). Maybe also give a description of the equipment you are currently using, and what loads you are running from it.

With the additional 5kW you'd have, as you said, additional MPPT inputs. If you're going the route of an 8kW, just consider your current configuration of panels, and how you'd re-wire them in order to make the most of the two MPPT's. For argument's sake, if you could re-arrange your existing 5.2kW of panels into two equal-length strings of 2.6kW, you could place them in parallel into one MPPT of the 8kW, and then add more of whatever other panels. If it happens to be difficult to do that, maybe because of space or panel direction, or different types of panels, then that would speak more for adding another 5kW, rather than the 8kW. The cheapest would be to put the tenants/flatlet as a non-essential load, and upgrade the 2.0kW string to 3.2kW also, but that's going to cause a different form of inverter envy. And it won't get you off Eskom as much, it should just prevent the tripping.

Maybe more effective, but not more efficient. To really get to the bottom of this, can you state the make and model for the inverter as well as for the panels. And then also a sketch or a Google Map satellite image of the house, showing where the panels are now, and where you intend to put the panels, and where you are likely to get shading. Without that it's hard to give you much more than generalisations.

Always speaking under correction, but I'm not comfortable with the idea to use these panels. Under normal operating conditions, okay, the inverter can limit the current it draws. The question is, what would cause the inverter to draw Isc? Simply, a short-circuit. Quick google search suggests it might be due to a blown fuse, faulty isolator or contactor, or an internal rectifier fault, faulty panel, earth faults, electrical surges, etc. The Isc rating of the inverter is never specified at either STC or NOCT, basically the inverter doesn't see the solar radiation or ambient temperature directly, it just sees current on the inputs. Isc-Current at STC and NOCT is a specification inherent to the solar panel, the current you'd expect when short-circuiting the +ve and -ve cables of the panel. If the panel can provide up to 18A Isc at STC in the event of an inverter internal short-circuit, then that will be above the current that the inverter is rated and expected to survive under those conditions. Maybe it's only just above, but it still is above. Basically, my fear is just that a repairable fault may become irreparable. And the question I'd always ask along with that is, WHY? WHY would one want to use panels in a setup that will forever limit the power you can get from the panel to max 75% of rated capacity?

Try taking away the "sync" option/tick so that you run on an internal clock, not synchronised with an external time server.

Check the basic time settings of the inverter (the clock time, not the timer), and you might find it's set to GMT+2 (Cairo), as opposed to GMT+2 (Johannesburg/Harare). The issue if I recall is that Deye does not make the Johannesburg time a selectable option. In South Africa we do not observe daylight savings time, whereas in Egypt the clock changes from GMT+2 to GMT+3 between April and October. If this is the case, you could set your time slot one hour earlier, or play around with settings; I think there was an option to set the time manually and lock out changes. Might be wrong. It was some time back that I looked at this issue on my Deye, and I don't really know how the Sunsynk version handles it.

Agreed, there are so many variables involved that a universal economic model is very difficult. The cost of generating equipment, and the cost of storage, the site-specific factors impacting installation, installer's markups, the expected lifespan of various components, degradation over time, the user's needs and usage profile, time of use, the municipal purchase tariffs, feed-in tariffs, the terms and conditions of grid export, meter costs, and also cost-avoidance. At least there are online tools and rules of thumb that can help predict generation, and one should be able to draw up a budget or a spreadsheet based on your power bill of how you intend to generate and use that power, and know at what cost you'd no longer be saving enough money for it to be worthwhile. Or, if the "industry" would settle on the expected cost for a standard/benchmark kit that would fit most folks' needs, say a 5-10kW hybrid inverter with specific amount of panels and battery, plus installation parts, something that's available in every major town and city, say through a national retailer, with recommended fitment costs within a certain distance from the store, and a certain distance from the DB board, single-storey, etc. then that might be a way to work around the variation in quotes.

If the inverter disconnects when the grid parameters are out of tolerance, that's surely by design, and probably not a good idea to try to override it. If it still makes financial sense, get more batteries to store power and export at night.

When I first read this I thought I'd give the installer the benefit of the doubt, considering that the writer was a bit imprecise on units and terminology (sorry, nothing personal). I just understood that the existing 2x 6-panel strings were configured as 6S2P on MPPT1, while a new string of 6x585W panels was connected to MPPT2. If this is indeed what was done, then there's nothing else left to do, it's all fine. If it was configured as 12 x455W panels as a series strong on MPPT1, then absolutely, change it asap.

In principle a good idea, but I'm thinking to rather move the East-facing panels onto the upper-storey roof, into the empty space, and to connect that in series to panels 5 and 6. I'd rather keep panels 7 and 8 dedicated to the geyserwise system, as they appear to the naked eye to be a different type of panel. Rather keep the geyserwise system intact as-is. Then yes, it makes sense to connect 1-2-3-4 in series, and also if you're adding 2 more North-facing panels, include them in that series.

As above, you can set your Zero Export power" to a low value like 20W. Actually, if you've got a bi-directional energy meter and if you're approved for export to the grid, you can set this value all the way down to zero (0). Do you have communications to a BMS? What batteries are you using? Is there any particular reason why you are running in Voltage mode? If you are able to, go under the battery settings and change "Voltage mode" to "Lithium mode." Use the work mode "Zero Export to CT." I'm working on the assumption that the CT coil is installed correctly. Your settings look alright, but note that the inverter will only start selling surplus power to the grid once your battery is charged. It's hard to tell anything from the image provided, because you don't say anything about your battery SOC, except that you're always allowing the battery Voltage to drop to 46V, which is basically empty. Maybe try to draw the graph of your daily consumption and production, SOC and Solar export statistics, so that one can see if there's anything unusual, or any obvious issues.

Under the Battery Settings you can set the Max Charge current to around 150A. I don't think you have to run it at its limit. Personal opinion. The inverter can handle up to 190A I think, but right now it's probably sitting at about 40A, and it's wasting potential power generation. Also, maybe just double-check under the Li-BMS settings, what the battery's charge rate is being reported as. For your set of 5 it should be in the order of 200A or more, maybe 250A.

Yes, and with that you've got an option of what you want to do, which upgrade path to follow, depending on where you'd like to end up. As was said, you can safely increase each of your existing strings to 8 panels each, to get a total of 8720W. In this case the PV Input Voltage is the limiting factor. Might be all you need, and if so that's fine. Otherwise, you could go the route of 5S2P per MPPT, to fit 20 panels, ie. to fit 10900W of panels. In this case the PV Input Current is the limiting factor. Better for if you want to charge more batteries, cater for poorer weather, or end up going more off-grid. In each case the nameplate Wattage is over the inverter's maximum of 8000W AC power and/or 10400W DC power respectively, but the inverter will just draw what it needs.

Safest is to link to this Status screen off from the home screen, and read the values under "Solar Power" for M1 and M2. If you are using two strings, then under M1 and M2 you should find similar Voltage of around 200V on each MPPT, and current of somewhere under 13A. If you are using both strings on one MPPT, then only one of the strings will see the 200V voltage, and current of up to 26A.

It's 26A per MPPT. Right now it sounds like you are using 1 string of 5 panels per MPPT, and each of those strings will max out at 13A. What you could also consider, is to put your 2 strings in parallel (NOT SERIES!!!) onto one of the MPPT's, so that one can reach up to 26A. This works say if all the panels are facing one direction. That would leave you free to put anything else on the second MPPT, say if you had roof sections facing East & West, rather than all North. Bottom line, you have options.

You have to put each string on its own PV input. There is then one MPPT controller for each string, and it will work perfectly as it was engineered to. Don't try to connect the two unequal strings in parallel to each other into a single PV input.

Are all these loads already all on the Essential Load port? If so, the best thing to do to see if you can go off-grid in Winter is... just start! On paper you're fine in theory, but test the practice without 100% commitment. Just flip the mains switch off and try to survive for the next 1-2 months. Right now, with the sun being the least, and the temperatures being the lowest, is one of the perfect best times to test the theory. Maybe don't just test Winter, but your rainy season too. Then decide next year if you want to pull the plug.