LeeMun

RIOT
My RIOT was only connected to the 4 x AM-2 battery bank and not the inverter.
When I got my RIOT in 2021, it showed both SoH and SoC.
Later updates removed the SoH from the display.
I sold my RIOT about 3 weeks back, so it's possible Hubble have restored SoH in the last 3 weeks.
AM-2
In terms of  my 4 x AM-2, I'm mostly happy with the service and experience with Hubble.
When I got the 4xAM-2 batteries in Nov 2021, I did have issues where SoH and usable capacity were decreasing.
I contacted Hubble support and via the RIOT they reset the batteries capacity from 110Ah to 100Ah.
Was not entirely happy with the decreased capacity...
Unfortunately resetting the BMS Capacity did not solve the problems with SoC/SoH.
Hubble then requested that I return the batteries for checking which I did.
Sending the batteries in did not resolve the problem either.
After a few weeks, under advice from Hubble, I switched to using Volts for SoC, as the BMS SoC was decreasing too quickly.
I used a Juntek shunt to monitor energy in/out of the battery bank and compare that to the SoC reported by the BMS.
see https://github.com/mysystem32/juntek_kg
A few months later I learnt via this powerforum of new firmware.
I requested the firmware from Hubble Support which resolved the problem with SoC/SoH and restored the reported capacity to 105Ah.
Since then the 4 x AM-2 have been working very well.
I'm sure Hubble have reasons for telling people not to use PBMS tools, but the T&C also block you from connecting Solar-Assistant to the console/rs485 ports.
Having SoH, SoC, Cell Voltages, Temperature, cycles, etc is useful for troubleshooting issues.
Instead of hiding this information, they should enhance RIOT to display all information.

Would I get AM-2 again? Yes. The product works. Support works.
I would probably go with AM-5 as they have different chemistry.

Hubble Support
My experience has been good with Hubble support and I can recommend them.
I very glad to have local support for the AM-2 and ultimately get my issues resolved.

Some of the imported products are difficult to get help with.

Is this on the battery or on the inverter?
It sounds like you did not equalise the SoC of the two battery modules before paralleling. It would be best to get them to as close to the same SoC as you can individually, then parallel them.

The battery volts are just fine. It seems the battery was fully charged on solar just after 13h00. The PV Watts are low at 15 W while your load is sitting around a 1000W. Seeing that the battery is fully charged, the excess solar should be going to your load. Instead it is utilising utility for load. Change the "Output Source Priority" to Solar>Utility>Battery? You want to use all available solar for load and only the shortfall should come from utility. Currently it is set to make use of utility, even if solar is available. 
I have my "Bulk Charging Voltage" = 52.6V (The recommended 53.2V is too high in my view). My "Float Charge Voltage" = 52.0V. My "Battery Cut Off Voltage" = 46.0V
Your "Back to Discharge Voltage" = 54V. That is too high. Your battery will never reach that voltage. Reduce that setting to 51V.
 

you can safely ignore the "use time for control" at the bottom of my post/image.
There are optimal settings for pylontech batteries, could you maybe send a screenshot of what your currently using, so they can be verified.
depending on your goals, the settings you mentioned (sbu,sub and more) can be mixed and matched between solar, batteries and utility for the optimum outcome. 
usually everyone's goals are different, but the best outcome can always be found.
Could I also possibly suggest you upgrade from your windows icc (V2 odd) to something more recent (v 4.9) or rather icm, it doesnt even need a pi, I installed mine on an unused android tv box, which I installed ubuntu on, still working great. You'll enjoy the freeing up of your laptop no end plus the benefit of yrs of extra features/stability.
regarding icc, I enjoy it because it means I don't have to make one (read time and money) and it works, sure it could use a few ideas here and there, so I send those through. The recent android app is the result of one of those ideas (thanks for listening). I'm hoping the next idea makes it, predictive pv output based on weather, with hourly outcomes, also it may not make it through, but here's hoping.
I already use this feature independently, daily atm.

You have a 145V model, This expert king will give 0V output PV on anything above 145V from solar to protect itself. 
connect your panels 2 in series and then the 2 strings in parallel. 
Don’t exceed the current rating on the inverter. 

I am currently running a 5kVA/4kW Mecer unit - and can not say that I have been experiencing this issue. As you can see from the picture below, whenever the load (blue) exceeds the available PV (green) the Mecer draws from the batteries (purple). Utility (red) has not been used during this time

However, it is correct that this model cannot supplement PV with utility - only PV and battery

Side note to your side note.
V=I x R
When referring to losses the typical formula use is P = V x I   , replace the V  with I x R      then you get  P = (I x R ) x I 
P = I2 R
The  energy is lost in the form of heat due to the resistance of that cable and current flowing trough it.  The formula P = V x I for losses does not make that relationship evident.  Whereas with P = I2 x R , you can clearly see for an increase in resistance there is an increase in Power losses, but even worse in the increase in current where the relationship is a square.
That's why a 48 V system is way more efficient than a 24 V system . Although current of a 48 V is only 1/2 of  the current of comparable 24 V system , the losses across the cables is 1/4 less. 
__________
 
As for the OP , in a single inverter , single battery bank system the difference in length between the positive and negative cable are irrelevant as is a simple series circuit and Kirschoff`s  2nd Law applies.
 
Obviously when a parallel inverter or parallel battery bank is connected this does not hold true anymore. 

@PowerUser In your opening post you ask battery (singular) to inverter, but then later posts pictures with batteries is parallel. Surely you can see have that could've led to people misunderstanding your question.
In the scenario of a single battery and inverter, the positive and negative cables can be of different lengths. The longer the length of the cable (Pos and Neg cables summed), the higher the resistance and the higher the voltage drop will fall over the cable (it doesn't matter whether this is on the Pos or Neg cable). As the Pos cable, battery and Neg cable form a closed, serial circuit the current will flow through all the components, equally.
Voltage (V) = Current (I) x Resistance (R)
 

 
If you are using 35sqr welding cable, the voltage drop over the cable will be ~0.8 milli-Volt per Amp per metre.
Example 1 - Equal cable:
Voltage drop over Pos will be = 0.8mV x 50 x 2 = 80mV = 0.08V
Voltage drop over Neg will be = 0.8mV x 50 x 2 = 80mV = 0.08V
Therefore, the Battery will "see" 50V - 0.08 - 0.08 = 49.84V
 
Example 2 - Unequal cable:
Voltage drop over Pos will be = 0.8mV x 50 x 1 = 40mV = 0.04V
Voltage drop over Neg will be = 0.8mV x 50 x 3 = 120mV = 0.12V
Therefore, the Battery will "see" 50V - 0.04 - 0.12 = 49.84V
 
So, both scenarios are the same (electrically) from battery and inverter point of view.
 
Side note:
However, the both cables should be kept as short as possible to reduce lost energy which is dissipated as heat by the cables. The power (W) lost can be calculated as P = V x I.
The higher the loss over the cable, the higher the temperature due to heat dissipation. The thinner the cable, the higher the voltage drop over the cable = more loss = higher heat, which in a worst case can lead to burning if underspec'd.
 
 
 
 

No, your model doesn't have an RS-485 or CAN port to connect to the Pylontech battery. So just leave it with battery type = USE, and use the settings with 52.5 V and 51.8 V (see many posts with those settings).
Check setting 13, back to battery voltage.
Yes, I would expect around 3 kW or a little over near midday of a sunny day, if the panels have good tilt and orientation. 2 kW suggests that something is not optimal. 

Hi @LeeMun, I have six Ja solar 540 watt panels in series on a Sunsynk, at midday I get about 2.4 kWh , I am based in Cape Town.

"General accepted practice" by who? Which manuals? My battery manual does not mention this. Yes, when connecting batteries in parallel or in serial, the cable lengths between the batteries have to be equal because of Ohm's law but which manual specifies equal lengths for the positive and negative cables? Can you please point me to one?
I have been looking at Victron's recommendations and they show the following:
 
 

 
Link to the full document: https://www.victronenergy.com/upload/documents/Wiring-Unlimited-EN.pdf. There is nowhere in the entire document anything mentioned for the +/- cable lengths.
Design 1, 3 and 4 show as correct, with different lengths for the positive and negative cables.
Besides, if you open your car hood, you will see your car battery has different lengths for the +/- as well. Did all the engineers fail in their design?
I'm just trying to understand the argument for equal +/- lengths. That's why I' asking all those questions and that's why I started this thread.
 

You can oversize your array (up to a certain point*) but the max the MPPT can push is 100A and the battery float voltage is about 53V so it can only supply 5300W. 12x450W panels are already 5400W. 16 panels are 7200W which is maybe a bit over-specced unless they are not optimally placed?
*The max PV short circuit current for the 250/100 MPPT is 70A. I'm estimating an Isc of about 11A per string so you could do up to 6 strings if you wanted. Just check your panels' Isc: floor(70/Isc) = max number of strings.

There is also 
http://www.multisibcontrol.net/

I have connected my Sunsynk auxilliary ourput to heat up my geyser as soon as the batteries are full and the PVs are generating more than 400W. This helps a great deal with idle time and keeps my geyser nice and hot. The same can be done with aircons on the grid side, if your Inverter can handle it. I have installed an 8KVa which can handle the amps easily.
 

True, but it's a small effect, perhaps 5-10%. The more solar insolation ("sun power"), the higher the maximum current that can be drawn from the panel, in almost direct linear fashion.
This is very true; the MKS IIs are very sensitive to maximum PV-side current. The results when this is exceeded by just 20% is dramatic. But the OP's client doesn't have an MKS II; it's a 145 V max model. These don't seem to have a limit on the PV-side current; it's more about the ability of the control system to cope with panels that respond to a small change in PWM ratio with a large change in current.
2S4P should be fine for this client.

You can exceed the PV Array power on the Sunsynk's (not sure  how much is safe, I ma sure there is a limit somewhere) , it will however clip the power at the max power so you will never be able to make more than 6500W. It does this by increasing the resistance in the inverter which reduces the amount of power it absorbs. The inverter does the same thing when your battery is full and the power being produced drops. 
You must not however exceed the combined Max Voltage of the panels as voltage cannot be reduced by resistance in the inverter and will result in smoke and fire!!
Max voltage on the 5KW  -  Max V DC input – 370V (100V~500V)
It has two MPPT's so you could do 2 strings of JA Solar Mono's 535W 6-8 per string, max amperage on the panels is 13.79 but the inverter will stop them from producing over 11A so the panels would never be able to produce more than 456W of power per panel and the whole inverter will cap out at 6500W. 
If you can do an East/West split you could have great power for a large portion of the day and start of with 6 panels per string and maybe leave some extra space to slot on 1 or 2 more depending on whether you need it or not

Sorry, thought you had a Sunsynk, see you have a Mecer, not sure on overwatting a Mecer 
 

You have the 145vdc version, so 2S is correct with a max of 4P. 4280W is well below the 25%(5000W) threshold. So 2S4P(8 panels) is correct for those size panels.  

Sometimes the Axpert sticker will also have a maximum input current value. The below pic is an example of the 450vdc solar charge controller on the Axpert, max inpur current 18A. More panels in series is better for this model. 

That usually means the maximum output, and you can safely exceed that by 20%, 25% at a stretch.
So I would use 2S4P (8 panels), unless it's an east-west split, in which case 5P is probably acceptable. Assuming it's a 145V absolute maximum model.
Sometimes the sticker says something like maximum PV array power, in which case you should not exceed that figure at all.

I've seen an installation with 9 x 8kw sunsynk inverters, 3 per phase. I think they can go as much as 8 per phase. 

I run a 40 year old dishwasher, a tumble drier and a few other appliances during the peak of the day and often peak at between 7 and 8 kW on my Sunsynk. I know it is safe to do this on a sunny day because the batteries are rated to deliver 7 kW from my bank so they can handle the load when a cloud passes overhead. I do this during load shedding as well. However, before upgrading the battery bank I was limited to 3.5 kW, so my loads were adjusted to only peak at 3.5 kW or slightly over as the panels can handle the rest even when overcast.
 
The batteries are a crucial component in running such big loads even on clear days as all it takes to overcurrent your battery is one tiny cloud.
 
In my case, if I only had one battery I would have been limited to 1.75 kW or just a bit over so I would have had to limit my loads to max 2.5 kW.

Layman's 2c. You don't have to use all your appliances at full rated power during loadshedding. If needs be, a microwave can be operated at a lower to medium watt setting like 600W, and without the grill function. A washing machine can be operated at a cool temperature, on a short cycle, or at a lower spin speed than maximum. And the iron can be turned to a lower heat setting. Why replace perfectly good appliances? As said above, be careful with the rated power of the battery; be sure it can handle the power draw when solar is insufficient, because there will be partially cloudy days where your solar production fluctuates and battery kicks in. Some experimenting on pure battery supply is maybe not a bad thing, starting from lower settings to higher.
 

All the appliances you mention work on the basis of switching an element/magnetron on or off. Lower settings simply mean that the on period will be a smaller percentage of the time. This means that the inverter should still be capable of carrying the full load.

You may want to consider getting rid of the microwave in favour of a 900W unit, and the iron (1200-1400W units are available). It's what did in my own solar journey. 2000W for the washing machine is kind of acceptable considering that the vast majority of this goes to heating water, I kept mine as well as my dishwaher (2200W). Alternatives include moving towards a semi-auto which will allow for cold water washes.
My own array is over 6kWp though, and the Sunsynk blends power, so any shortfall can be made up using grid / batteries, not sure about the Mecer.