HennieL

... And, of course, if you are primarily interested in saving money, and if you pay "peak" and "off-peak" rates, then it makes sense to only switch the geyser on during the "off-peak" times. Once you can generate your own solar electricity then obviously your geyser should be switched on during good sunshine periods when you are generating free electricity, and off at night and/or during heavy cloudy days when it would require grid electricity. Most of us on this Forum are probably in this latter category of users. My geyser is set (through the smart controller) to switch on at 10:00 and off at 14:00 during winter with normal good weather, but I manually change this via my smartphone app as required to make use of any sunshine and/or bright indirect lighting. Tell us a bit about the company you work for - how do they manage home energy?

Hi Emma, and WELCOME to the Forum. You already have two options for your geyser - I am controlling mine via a CBI Astute smart controller such as this: https://www.voltex.co.za/product/cbi-astute-smart-controller-asc-wifi-enabled that works via Wi-Fi. As long as my home router is on, I can access the controller from work and switch the geyser on or off as I like. I can also check the power used for the day, and check daily historical power usage for some months. Best way to learn is to ask questions right here on the Forum.

Apologies to @Deanos for all the confusing posts - we do tend to dive in deeply, and forget that there are also "normal" people on the Forum... You do have a challenge finding the Goldilocks system that is "just right" for your unique situation: You already have a 24V PWM type inverter, and that does not work well with your modern high-power solar panels Upgrading to a modern hybrid-type inverter than would work well with your solar panels will require a 48-52V lithium battery (very expensive...) AND perhaps two more solar panels of the same type, and facing the same direction - you have confirmed that this is not a viable option Adding another battery or two would work if you are prepared to charge them from the grid, especially in winter, but this will be rather expensive Buying a simple generator to charge your existing batteries would also work, but would again be costly. I had such a system with my old 24V 2.2kW original inverter that I ran for nearly 10 years without solar panels, and that worked perfectly for me. My inverter was powered from a normal mains plug socket, and whenever I wanted to charge the batteries (I still had a bank of 12V lead-acid batteries) I would just plug a lead into the generator, and plug the inverter into the other end, and everything worked fine. Can I just bug you with one more question/suggestion: Looking at the two photos that you uploaded, there appear to be sufficient space for one solar panel in landscape orientation across the center section of your house above where your two panels are currently mounted. Judging by the size of the panels in the photo, you should easily be able to mount two panels either side of the center ridge. If this would be possible you would be able to upgrade to a 3kW, or even 5kW inverter with 4 solar panels if you buy just two additional units of what you currently have. This will be more costly than the additional battery, or the generator options, but will save you money in the long run by putting your study completely "off grid", and even supplementing the rest of your house - think about it...

Thanks @Deanos A picture paints a thousand words... Your slope angle is not too flat, so that's out of the picture 🙂 It is true that one cannot have one "string" of panels working as expected if there is even just a portion of one panel being shaded, so your installer gave you good advice on that... He did NOT give you good advice on matching the panels to your inverter. His crucial mistake was not matching the maximum voltage output of your modern, high-power panels (being 52.7V per panel) with the maximum VOC of the PWM of the solar charge module of the inverter (which is 80V). When one exceeds the rated VOC of the inverter module the inverter will become permanently damaged... so, you are limited to only one solar panel in series (one panel feeding through a second panel, the same as (say) two torch batteries feeding one through the second to give the required voltage. One can double the current (Amps) being provided by the solar panels by connecting two or more of them in parallel without increasing the VOC, but panels are more efficient when supplying power at higher voltage. What he did not allow for, was the fact that your panels are not optimally facing the sun (both in terms of slope, and especially in terms of directions relative to the sun's path (called azimuth). Because of this, your two parallel connected panels are not even producing sufficient voltage to reach the ideal "operating zone" of 30V - 32V as stated on the inverter label (your first image posted in the first post). If he had recommended lower rated older technology panels (as @Denns stated he did in his last post above) then ironically you would have been better off, as you could then have connected two panels in series and used most of what they would have produced...

I respectfully disagree. Inverters can accommodate some over-supply, as they just limit the input. Even if this inverter cannot do so, the maximum output power of the panels is at standard testing conditions, which might be achieved in some places in South Africa at the peak summer time between 12:00 and 13:00. It is highly unlikely that the panels would produce this power at the rather flat slope (vertical angle) that they are installed at - and the azimuth angle is also working against this. I would be very surprised if the panels can even reach 50% of this rated power... What I do note when comparing the specifications and output information provided by @Deanos in the first post is the following: Solar charge module rated current = 50A @ 24V Charge module VOC = 80V Solar charge module operating voltage range = 30V - 32V PV input voltage (measured) = 26.6V PV input current = 11.0A It is clear that the PV input voltage is below the PWM operating window. At the time the measurements were taken, the batteries were being charged at approximately 26.6 V × 10 A = 266 W (about 0.05C for a 200 Ah battery bank). A 24 V, 200 Ah battery bank stores approximately 4.8 kWh of energy, and charging from 50% to 100% SOC would require about 2.4 kWh of power. At a constant charging power of 266 W, this would theoretically take about 2400 Wh ÷ 266 W ≈ 9 hours. In practice, however, charging will take even longer because the charging current tapers during the absorption stage, and the available solar power varies throughout the day. @Deanos can you please confirm if the PV input voltage reported by the inverter is the true input voltage (as stated in the last screenshot of your first post), or is this the voltage of the batteries being charged? According to that last image, the battery voltage is reported as being 27.03V, and if that is the case then the batteries were not being charged at all at the time of measurement... From all the above, it would appear that your system would have worked better with older solar panels that have lower VOC ratings, where you could connect two panels in series, and not in parallel (you cannot do that with these panels as this will exceed the critical VOC limit), and with perhaps a second "string" of 2 series connected panels then paralleled with the first "string". If you cannot do this, then the only option left would be to face the panels correctly (facing close to north, and at the correct slope (between ~30 and ~40 degrees from horisontal, depending on where you live)). PS. This is assuming that the panels are in fact connected in parallel - if connected in series the I agree with what @Denns said above (I was busy typing when he posted...) BUT then there is a real risk of exceeding the VOC limit in winter...

Thanks, this helps us to understand your situation. It's difficult to make out much detail on the photo, but it looks like the panels are on the eastern roof that is actually sloping down to the west, with an east facing roof to the left of the panels... Can you please confirm if this is the case? Something else to check is the slope of the panels (degrees from vertical) - those panels appear to be quite flat (horisontal), and with the winter sun being much lower in the sky that would explain why your summer yield was OK but the winter yield is much less. Would it be possible to raise the angle of the panels? If you care to tell us where your house is located (town) we can easily calculate the optimum slope for summer, winter, or all-year yield for you.

Can you please elaborate on your roof size and layout - it is quite strange to hear of any roof that can only accommodate two solar panels... Don't you also have a roof facing west (or a carport of something facing north...) where you can install another two panels? Also, you mention that you have two 100Ah lithium batteries installed. Lithium batteries don't like to be drained deeply, and if your system was installed just after covid lockdown then they are already about 6 years old. This would not be a problem if they were only lightly discharged, but if they were regularly discharged deeply then their capacity could already have reduced enough to become troublesome.

I concur - there's just too many variables to be able to compare one system with another - even in the same town... I generally only use about 50% - 60% of my system's capacity, and that alone will skew my PV harvest, as my (lack of) consumption would limit what the panels can produce... Still, nice to compare figures as long as it's not taken too seriously 😉

In my opinion: I'm sure it will... As @Kalahari Meerkat said, the panels will only produce what is demanded, and unless your consumption exceeds the inverter's rated power, the panels won't exceed that (plus losses...), so it's unlikely to be a problem. It would, however, be wise to set your battery charging current at a value that this plus your house load will not exceed the inverter rating (i.e. don't charge at (say) 200A for two hours - rather charge at 50A for 8 hours).

Interesting - here in Bloemfontein my average solar harvest for 1-24 June is only 21.7% below my average for April (best performing month for this year...) My panels are facing true north, and at a slope of 27 degrees.

I agree - how do you connect a single phase inverter to a 3-phase system?

Hi Francois, It will help us to advise you if you can supply more information: Where are you (or should I say the house) located - the Cape has much less PV production at this time of the year than (say) Prieska... Are the panels facing exactly to true north, or are they offset a bit to the east or west? Without knowing this one can not make an accurate estimation. Can you confirm if the 9x425W panels are on one string (connected to one MPPT), and the 8x545W panels on a second string (second MPPT)? Are all panels in a string connected in series? Do you have any shade on any of the panels during the day? Even just a fraction of one panel shaded can drastically reduce the production for the whole string. Are you actually trying to pull more load than what is being produced? PV panels will only produce the amount of power being demanded from them, and if e.g. you are only demanding 2kW from the panels at (say) 13:00 on a clear sunny day, then the panels will only produce 2kW of power... Lastly, it will help to know what make/model of inverter you have, and what brand(s) of solar panels.

How are the batteries connected to the inverter - through a busbar, or with cable going from one battery to the next? (thinking that if one cable is longer than the others it could result in under-charging...) Also, were these batteries all fully charged/balanced individually before connecting them to the inverter? (if not, then it is possible that the less charged battery will progressively become even less charged, as the inverter could stop the charging when the first battery signals that it is fully charged...) I'm not sure if the Deye's software is similar to that of the Sunsynk (I do believe it is...), but if it is then you could download *some* battery data (more than displayed on the inverter screen or software. I discuss the process here: https://powerforum.co.za/topic/34104-finding-and-understanding-the-sunsynk-operation-data/ I can unfortunately not advise directly, as I do not know this battery - but I hope that someone can help you.

According to my Sunsynk installation manual the CT coil's arrow must point to the inverter - see screenshot below: I'm not a sparky, but given that your inverter + PV panels should feed your non-essentials while the sun is shining, is it correct to "feed back" to your non-essentials from the "output" side of the earth leakage, as shown in your diagram?

That's shocking to hear, and they want us to trust them... Seeing that Solar Deity does not want to answer my simple questions regarding other "agents" for Pylontech, I did some digging via Google AI: "Pylontech operates through a shared agency network in South Africa, meaning they do not use a single exclusive importer. Instead, they distribute their products through a select group of officially appointed distributors and service agents. The primary official channels managing Pylontech imports, distribution, and official support in the country include: 1. Primary Appointed Distributors (Importers) The bulk of official Pylontech stock enters South Africa through a few large-scale solar and electrical wholesalers who handle direct import and distribution: SegenSolar: One of the most prominent official distributors in SA, operating multiple distribution offices and service centers across the West Rand, East Rand, Cape Town, Durban, and Bloemfontein. CNBM (China National Building Material Group): A massive global entity that imports and distributes Pylontech equipment alongside other tier-1 green energy solutions in South Africa. ACDC Dynamics: Operates as a major official reseller and distributor, leveraging its massive nationwide electrical supply network to distribute Pylontech residential and industrial battery solutions. Current Automation: Another major technical distributor that officially stocks and supports the Pylontech range through main hubs in Johannesburg, Cape Town, and Durban. 2. Appointed Service Agents Because Pylontech does not have a direct brick-and-mortar corporate footprint for public walks-ins in South Africa, they delegate technical repairs and warranty assessments to specialized domestic partners: Solar Deity: Officially designated as the appointed Pylontech Service Center for South Africa. Based in Gauteng, they handle physical diagnostics, technical support, and component repairs for units in and out of warranty. 3.Crucial Warranty Note If you are dealing with a faulty battery, Pylontech's corporate rule dictates that warranty claims must follow the exact supply chain of purchase. For example, Segen Solar will generally only process and approve a warranty claim if the serial number proves the unit was originally supplied out of their specific warehouse." I also note that Pylontech has a very low Trust index (only 2.2) and more importantly, a -100NPS (likelihood of being recommended)