FixAMess

The "C" thing should not be your primary criteria for buying a battery. 
It only becomes an issue when one only has a few batteries, and with current loadshedding having 1c vs c/2 is irrelevant.  One needs batteries to last at least 6 hrs.
With an 8kW inverter, you'd need at least 2x5kW batteries. So in effect you'd drain your battery in just over an hour at 1C. What about the other 4,5,6 hrs of loadshedding?
Also, if you have split your db do you actually need 8kW of power on the non-essential loads?
 

See here....
https://www.solarquotes.com.au/blog/home-battery-testing-results/
 

See here....
https://www.solarquotes.com.au/blog/home-battery-testing-results/
 

I present my solution, in the circuit that you can see in the image, to be able to connect 2 Strings to a single MPPT input of any power inverter, in my case East string and West string with automatic switching at a time close to solar noon to the PV2 input of my inverter.
I have a Goodwe ES5048 inverter, for those who don't know, this almost 5Kw inverter has only 2 photovoltaic inputs (2 mppt) for 2 strings.
The idea arose a little over 2 (two) years ago when I wanted to expand the area of solar panels to increase my solar production but I found myself with the problem of not having more space in my house to install more panels to the North.
I was not interested in buying another inverter to have 2 more mppt, not only for a matter of physical space to install them, but because the Goodwe is not parallelizable with other equal ones and the fact of having to modify the electrical circuits of my house downstream of the inverters was a resounding "no" for me
I have very good roofs at home to the West and East, so I began to investigate if there was any commercial solution for this purpose and I did not find anything.
I designed this Circuit that I present to you now, but that I have been working at home since March 2021, exactly tested for 2 (two) years, working perfectly with the DC RELAY Values that appear in the photographs of the Real Board installed in my home. It has been tested in all temperatures and conditions.
Before March 2021 I tried other 40 amp relays but they got too hot and I was afraid they would burn out. That's why I bought these that are 150 Amps and with the suggested aluminum heatsinks, even in the middle of summer with temperatures of 40 degrees, the heat dissipated is not high.
It's a very simple circuit that anyone with an idea of elementary electronics/electricity can do.
I could comment much more about it, but I prefer that you ask the questions that may arise.
The relays and their heat sinks can be purchased on aliexpress. From 600 Volts DC they have up to 400 Amps.
The timer can be purchased at any electrical/electronics store for a DIN board. It only matters that you have a Normal Open contact and a Normal Closed contact.
The other thing that is needed is a low voltage DC source to open or close the relays, it can be from an old cell phone since it supports more than 5 Vdc.
Really for me it was a solution that I am very happy with. None of the relays have burned out in two years of use. with an investment of a few dollars, minimal compared to the benefit it brings.
The scheme is self-descriptive, so as not to lengthen this Post.
 
 

Here is a diagram given to me by SYC. Shows how the extra pair of cables are attached
.

Here is a diagram given to me by SYC. Shows how the extra pair of cables are attached
.

I'll double up on the std pylontech cables..

Forgot to mention that if you are planning to discharge and charge at 198A, then you should consider doubling up on the Pylontech cables

It does't. Let's just apply Ohms law.
Let's look at a hypothetical example. Parallel a 100Ah and a 200Ah batteries. Theoretically the 200Ah battery has half the internal resistance (Ri) than the 100Ah battery. The current flow is determined by the difference (delta U) between the internal voltage of the battery (that's the voltage on the terminal at zero current.) and the actual terminal voltage. Assuming both batteries have same SOC, their internal voltages are equal. So, the differences between terminal and internal voltage are equal on both batteries. Ohms law: I = U/R. Since the smaller battery has double Ri, its current will be half oft the larger batterys's current.
I = delta U/Ri  versus  I = delta U/2Ri.
 

1) Your bank will be 90% DoD, you are now looking at a bank and not an individual battery, so you will be limited to 90% DoD because of the US3000
2)You will still get your full capacity .i.e 50A from UP5000 + 37A(4) from your four US3000. You battery bank will be capable of charge/Discharge at 198A 
3) You will have to parallel all five batteries and make the UP5000 a master. You do not need to use a busbar, just use the Pylontech cable that connects the batteries to the inverter/fuses 

Should take normal NH00 fuses. Just about any electrical supply store should sell them - but just make sure the specific ones you get are DC rated.

No, the available pv power will still be shared, albeit indirectly.  Remember your battery bus is common to both inverters. The acess pv power that is available on the inverter with PV will be made available as battery charge power. The inverter with no pv input will demand power being drawn from the battery, causing the charge power from inverter 1 to be consumed. This is of course possible if the battery is already full and does not consume the charge power.
 
EDIT :   this explanation may or may not be applicable where power is exported via gridtie non essential connection. 

nope, each MPPT its own panels not connected to anywhere else...

Did anyone attend the Segen presentation on Hubble batteries?
Anyway, the item that fascinated me was on earthing and bonding given at the end of the presentations given by a learned man whose name I cannot remember at present.
He was pointing out that earthing and bonding were not the same thing. Bonding is where each item it connected together so that no potential can exist between components, This is important to prevent human harm as well a electrical damage to electronic equipment.
Then there is earthing. This should be done at only one point in a system. He gave examples how cows have been electrocuted at a distance from  distance from the lightning strike. The strike creates a spreading pool of potential as it dissipates into the ground.  There is a difference in potential created between the legs which electrocutes them.
If we put multiple earths in the ground a nearby lightning strike will do the same thing. It will create potential differences between the earth rods causing damaging currents to flow between components .
The point he then made was that we bond the equipment only to the incoming utility earth. South Africa has violent electrical storms on the highveld and I remember and years back Escom built a new national control centre. While the structure was securely earthed, all electrical equipment was bonded to an earth "mecca".
More strikes occur to nearby objects than directly to panels. If there is a direct strike, everything is fried anyway. Rather protect using appropriate lightning conductors direct to ground.

Instead of just using a spreadsheet it helps to understand how/why the calc is required/done.
Here is a write up for the OP to figure it all out...Can't remember where I got it from and cant take credit for it.
As temp decreases, panel Voltage(V) increases. We want to calculate the maximum Voc for a panel at the minimum temp we expect in the region. So for JHB we use 0’C, in Bloemfontein we use -10C or whatever you chose...
To calculate string length (series) use the total Voc (Sum of voltages, because they are in series remember, (+) including adjustment for cold temp) from panel data sheet and then do as follows;
 Record-low temperature: -10ºC (From your region, town)
Temperature coefficient of (VOC): – (0.30) %/ºC (From data sheet *)
Module open circuit voltage (VOC): 39.4 V *
Inverter maximum input voltage: 600V (From inverter, MPPT data sheet)
The STC temperature is 25ºC. This temperature needs to be deducted from the array location’s record-low temperature of -10 degrees as follows:
25 – (-10) = 35º difference.
Multiply the 35º difference by the temperature coefficient of VOC (I’ve used the positive value for an easier calculation, though you get the same result) then multiply by the module’s VOC:
35 x 0.0030 = 0.105
0.105 x 39.4V = 4.137V (So the Voc will increase by 4.13 V if the temp goes down to -10C)
This is how many volts each pv module will increase due to record-low temperatures.
Add the voltage increase to the Module VOC.
Then divide the inverter maximum input voltage by that number. This will give you the maximum number of modules that can be wired in a series string per that inverter and specific location.
4.137 V (from calculation due to low max low temperature) + 39.4V (from datasheet) = 43.537 Vmax (At -10C) Increase in voltage due to temp drop per panel.
Normal voltage at 25C
600V / 43.537 = 13.7 Panels (round down to a whole number)
The maximum number of pv modules in this series string is 13.
A series string of 14 could potentially produce more than 600V during record-low temperatures.
When string in parallel, add Current, V stays the same, when in series, add Voltage, I stays the same.
 

Hi @HermanvN
My home energy provider is about to approve power generation so I can export my surplus to the Grid and will continue to use Economy Mode.
If you do not export to the Grid, without a doubt, the economic mode is optimal if you have Lithium Batteries. By using it correctly and configuring the 4 Battery Charge/Discharge periods, you will be able to perfectly adapt it to your installed capacity, your solar production and your consumption.
When you want to EXPORT photovoltaic surpluses, as you have already been informed in the Forum, there is currently no optimal automatic configuration without some of the Energy from your batteries being drained to the GRID, however you can get pretty close.
It's a litle hard to understand this Mode at first, so I'm going to give you a numerical example.
Once you have defined the DOD On Grid that you consider optimal for your installation, you must determine your night consumption as accurately as possible and from there, in the night battery Discharge Period, enter the corresponding Discharge %. For example, if your Average Nightly Consumption is 400 Watts, then in the nightly discharge period you put 9%. This tells the system to never discharge more than approximately 414 watts from the batteries (9% of 4600=414).
Well, as you know, this is an average, if during the night your essential + non-essential loads exceed 414 Watts, the difference will be imported from the  Grid; while if your consumption is lower, the surpluses up to 414 Watts will be EXPORTED to the Grid. But your batteries will NEVER be discharged beyond that limit.
The next day you will find that in the configured period the total energy delivered by your batteries was the product of the hours x those 414 watts (following your schedule from 0:00 to 6:30) the energy discharged from the batteries will be 2,691 kWh (approximately ), the latter as long as your configurated DOD  has allowed its discharge.
Obviously, in the night period that you have defined from 20:01 at 23:59 you will establish another % because it is generally an hour where more than average is consumed. Example if your average is 800 Watts, you put 17% or 18% in that period. (This is always approximate since the calculations are made on the theoretical Nominal Power of the Inverter).

Finally and if you don't want to export anything to the Grid, you must intervene manually as indicated, but it is not necessary to change the Mode, simply Set the Export to "0" in the Power Limit through the PV Master.
Regards

Apparently if run is > 50m then one must use metal conduit.
In ceiling spaces its probably a good idea to use metal...
For short runs inside the house plastic trunking is good.
 

Apparently if run is > 50m then one must use metal conduit.
In ceiling spaces its probably a good idea to use metal...
For short runs inside the house plastic trunking is good.
 

Does it talk to Goodwe?

The most important aspect is that the Inverter output neutral is never mixed up with the Eskom Neutral otherwise this can happen. This is the massive danger of not splitting the Main DB into Essentials and Non essentials staying the Eskom supply DB. 
Essentially what you getting is the AC input Neutral meeting the Inverter Output Neutral & it is here where the damage happens. I did the exact same thing on a site where it was a compete mess & there was no split done. Needless to say, a fridge, kettle & PC later. It's really harrowing but a Full DB split & very close attention to the neutrals prevents all of this mess. The Lives are easy to track due to the nature that it's colour coded & goes through the other end of the Changeover switch. The Neutrals hide a hidden trap & if the input Neutral meets the out neutral you have the start of all the issues. 
God luck with this but I can having cleaned up a few messes tell you the Neutrals is where the issue lies...

This is incorrect...You do not have to have 5kW of battery power available just because you have a 5kW inverter.  What you must do is must do is make sure your inverter only draws the maximum current that your battery allows! I can have a 1000kW inverter and a 1kW battery (nonsense, I know, but trying to prove a point)  As long as I limit the inverter to only draw 100W from the battery all is good. 

These figures are very close to my MAX production (1 day of the month) by month for a JHB, N facing panels at 22 Degrees, 6kW PV installed.
My average for June over the last 2 years is 21kWhr, which is the lowest of the year...
With all the rain we've had in JHB this summer the stats are not in line with 2019/20/21.

Sunsynk even if it was R8000 more expensive

Remember the battery is at 48V, if you want to pull a max of 200A from that battery you will have 190A X 48 V = 9120Watts....Your household breaker is probably the std +- 60A, so the max your home can / should draw is 60A X 230V = 13,800W....
Consider the loads  you draw that you will ever need 9000W concurrently from your battery i.e mostly at loadshedding time?
Turn on your tumble drier, kettle, microwave and hair drier and see what you draw. Will you ever be running all these concurrently during loadshedding?
The point is, don't get too hung up on maxing out the system because it is rated to do so, a 1C battery is good if you only have a small battery bank (as mentioned previously) and dont let the 1C be the most NB aspect of buying a battery...
All the LiPo4 batteries are similar technology, what sets them apart is price, support, quality of parts and most NB
1)the BMS quality and
2) the ability of the batteries to speak natively to the Inverter and
3) Support (questionable - if you buy tried and tested you don't need much support)
Go read all the "battery problem" threads, they all follow the same pattern; clever dick buys battery X because he saves R2, or its's a 1C battery and he saves R2. He then finds he needs special software to speak to the battery because his crappy inverter (also saved R2) does not speak to the batteries as advertised, so he buys Y tool to do so (more cost). He then finds the SOC is never correct, and the BMS gives errors...and so on. 
In short, if you're the average guy, just go the toyota route and buy Sunsunk and Pylontechs...Simple, combination works, very reliable, well supported and there's plenty of forum support.
We tend to overthink everything.....
 
 

Remember the battery is at 48V, if you want to pull a max of 200A from that battery you will have 190A X 48 V = 9120Watts....Your household breaker is probably the std +- 60A, so the max your home can / should draw is 60A X 230V = 13,800W....
Consider the loads  you draw that you will ever need 9000W concurrently from your battery i.e mostly at loadshedding time?
Turn on your tumble drier, kettle, microwave and hair drier and see what you draw. Will you ever be running all these concurrently during loadshedding?
The point is, don't get too hung up on maxing out the system because it is rated to do so, a 1C battery is good if you only have a small battery bank (as mentioned previously) and dont let the 1C be the most NB aspect of buying a battery...
All the LiPo4 batteries are similar technology, what sets them apart is price, support, quality of parts and most NB
1)the BMS quality and
2) the ability of the batteries to speak natively to the Inverter and
3) Support (questionable - if you buy tried and tested you don't need much support)
Go read all the "battery problem" threads, they all follow the same pattern; clever dick buys battery X because he saves R2, or its's a 1C battery and he saves R2. He then finds he needs special software to speak to the battery because his crappy inverter (also saved R2) does not speak to the batteries as advertised, so he buys Y tool to do so (more cost). He then finds the SOC is never correct, and the BMS gives errors...and so on. 
In short, if you're the average guy, just go the toyota route and buy Sunsunk and Pylontechs...Simple, combination works, very reliable, well supported and there's plenty of forum support.
We tend to overthink everything.....