Sammyigt

I came across a post here the other day where someone asked about the fact of requiring a second DB box for essential loads (ie not having a dual supply and splitting circuits in the same DB box).
Ive heard many electricians mention this as well. 
This is not actually necessary. The only time this would be required is if there is not enough space in your existing DB to accomodate the additional main breaker and RCD.
Please note however the following requirements from the image attached.

So that means alternative got over 15% cheaper. 
Way to go Eishkom !
 

Wekkit has hit the nail on the head here.
Grounding of any exposed metal part is not only a safety factory but it is also the law. 
PV panels do also need to be earthed individually using an earth cable of equal or greater diameter to that of your DC PV cable, from there a 16mm earth cable needs to run outside the building to an array of Earth spikes.

Right, so I spent some time on Google and read up a bit on the subject, as it seems to be a bit of an issue.
Here is some links for your own benefit, granted they mostly from the US.
http://www.homepower.com/articles/solar-electricity/design-installation/ask-experts-grounding-dc-systems
http://www.solarpowerworldonline.com/2012/04/installation-practices-keep-your-pv-system-well-grounded/
http://www.nmsu.edu/~tdi/pdf-resources/CC102.pdf
http://www.nmsu.edu/~tdi/pdf-resources/CC103.pdf
 
http://www.solarabcs.org/about/publications/reports/systemgrounding/pdfs/SystemGrounding_studyreport.pdf
 
What it really comes down to, is that ALL exposed metal parts of any electrical system, be it AC or DC, needs to be grounded. That is the law.
 
It is also stated, that if your PV voltage is higher than 50V, the positive, or negative, of the PV needs to be grounded. And this is where the probelms start.
Go and have a good look at your MPPT, and you will find the negative input and output is common. So when you grounding the negative of the PV side, you in effect grounding the battery negative as well, through the MPPT controller.
When we move over to the Invertor, you will find that the older type invertors that have an isolating output transformer, the DC negative is also grounded to the invertor chassis.
The newer type transformerless invertors seems not to do this.
 
So, at least ground all the metal parts. What is reccomended as well, is to have the PV ground wire seperate outside the building to an earth spike. Incase of a lightning strike, it will keep the flash outside your building while traveling down the earth wire, and not be a fire risk inside the building.

Done like this it might work, the problem with systems like this is when both are trying to charge ext. If they have Pylontechs, they would know the true SOC of the batteries even if some other divice is using some power from it. The only problem comes in when the combined loads exceed the allowed discharge current of the Batteries. If exceeded, the batteries will just switch off and the client will be without power in any case. You have to ensure that you have enough batteries to supply loads like that. 
 

I don't see a problem with what you suggest. The second inverter is just another "DC load", and DC loads are common in many applications.
You can configure the second inverter with a voltage to disconnect at (Say 3.1V per cell, times 15, that.s 46.5V), but the BMS will also disconnect if the battery goes too low.
US2000 battery recommended max discharge is 25A per module, so just remember to size it with enough modules to cater for the extra "DC loads".

Not only are they not allowed, they can't. An Axpert is an off grid inverter and does not have bidirectional capability. Hence why is not on NRS grid approved inverter list. It does not need approval. 

I am busy with the importers and the manufacturers. SO watch this space

Active SOC is a value that is automatically set by your GX device

If battery life is enabled then each day your battery does not fully charge the Active SOC increases by 5%. Each day it gets fully charged your Active SOC decreases by 5%. If battery life is enabled then your batteries will not discharge more than your Active SOC. The above all happens cause your system is trying to get your batteries to 100%. 
If you change your Minimum SOC the active SOC increases with it, but will only decrease each time the battery is fully charged.

Correct. In South Africa perhaps not such a big deal, but in Europe, as winter sets on, it automatically moves the limit up and as summer arrives it moves it down again.
Many lithium batteries balance only at the top, so it is advisable to use BatteryLife to ensure the balancer gets a chance to run every few days.
Yes it does. See here. When moving the active soc down it always starts from whichever value is smaller (the one the user set, or the one it determined itself). It can therefore never go more than 5% less than what the user requested. It also cannot go higher than 80%, in other words, BatteryLife will not force the minimum SOC higher than 80%, only the user can do that.

The SOL Output Priority setting seems to be for the situation where you want your batteries as full as possible for as long as possible, to be ready for load shedding for example. It should run in battery mode through the day, charging the battery from solar and powering loads as well. It should stay in float mode all day, so you should not have the LC Display as in the photo above where the panels aren't connected to anything.
Your LC Display also shows HS (Master), meaning that you are in a parallel setting. You only mention one inverter, so that is wrong, and may be confusing things. Perhaps it's time to check your settings against @Chris Hobson's list here: Axpert Settings 1.1 .
This could be normal. It's not unusual for the solar charge controller to slightly over-charge the battery, e.g. when the sun burts out from behind a cloud. That means the charge current will go lower, often to zero, until the battery voltage comes down to the float setting again. If you happen to look at the LC Display at such a point, you'll see current coming from the battery and little to none from the solar charger. But this situation should only last for a minute or three.
If it persists longer than that, something is wrong. Plus, as I say, having the panels disconnected as per your photo is also wrong.
Do you have your panels configured correctly for your solar charger? You didn't tell us the exact model of inverter, so it could be one with a 145 V max Voc or a 450 V max Voc; it should say which near the bottom of the large sticker (the one with the bar-code) on the right hand side of the inverter.
If you have the 450 V model and all panels are in series, then you may be exceeding the 430 V maximum MPPT voltage, and that could explain the disconnected panels, especially around midday.