VicB7

TNC- Least safe 
TNC-S- Safe
TNS- Safest
TT- Safe 
IT-Less Safe
 

We have been around the block so many times regarding permanent neutral earth bonding. It is not allowed and is dangerous. A TN-S system shall not be converted to a TN-C system and all installations done with permanent neutral/earth bond is not compliant and any COC that was issued for such installation is not valid. The electrical code is clear on this no reason for powers to be to debate the issue. The ECB is following up on IE's and MIE's issuing COC's knowing that permanent bond exist in the installation.
6.4 Neutral earthing
7.16.4.1 Whereas TN-C systems may be implemented along the distribution
system backbone, the individual service connections at every distribution
kiosk shall be TN-S.
7.16.4.2 From the point of supply to each user or part of a communal
installation, the neutral and earth conductors shall be separate conductors.
7.16.4.3 Wherever the neutral is connected to the earth, a warning notice
shall be fitted to the outside of each distribution kiosk in the distribution
system, indicating “Neutral earthed inside”.
7.16.4.4 A clear notice shall be fitted at the combined neutral-earth
connection inside each distribution kiosk in the distribution system, that
prohibits the removal of this connection while the supply is alive, or might
become alive.
7.16.4.5 The neutral shall not be earthed beyond any earth leakage unit.
7.16.4.6 A TN-S system shall not be converted to a TN-C system.

Ignore after 12h00 as clouds set in on 9 Dec. In winter after 12h15 severe shade from a high side wall where my car port is located next to the house. 
During winter I used a bracket to lift about 15 degrees and still the peak is only 600W and PV starts 08h00. If string was flat the yield would be even less. 
During sunshine days in summer the peak is about 1000W and starts 06h00 at zero degrees tilt. Not many full sunshine days. 
Values are for a 1320W string facing north. 
As you have a concrete roof and being an expansion to main strings I would make it adjustable to say 3 positions. 

 

What's so funny about this thread is that, yes you need a Wireman & yes you need someone to certify & test AC circuits. Given the fact that Solar tech is so new & has boomed, there are so many guys that took it on as electricians & what's also funny is that electricians know AC circuits pretty well. However, & a big however, Solar Systems contain so much more than AC circuits etc. 
They contain a large portion of DC circuits on the roof in the form of PV as well as batteries & HV batteries for Commercial sites. The PV arrays can get up to 850V and so can the batteries in DC. So we start playing with potentially life hazardous DC power to which a lot of electricians are not equipped nor experienced with. Let's not paint everyone with the same brush, a lot of them have also good DC experience. 
It doesn't stop there. Commissioning is another subject. Good to connect it all up but it will require some system knowledge & that usually borders on IT with CANbus & RS485 as well as working & crimping RJ45 cables with pin outs. Then you have the collective idea of someone who possesses the skills of mechanical knowledge of what works on roofs & safety on roofs. 
Let's just say I think it requires a lot of disciplines mixed into one person & that person is not really ideally an electrician as a must. 
I hope this gives confidence to a lot of ppl but I also hope it keeps ppl honest enough that if you don't know something or you are not sure there is no embarrassment to ask or seek answers. The best of us & the worst of us don't all know the same things. 

Must say this is exactly the sort of conversation I thought we would have. My team that works with me is broad & diverse. From a youngster that does a lot of roof planning & PV installing all the way up to a semi retired railway signalling engineer. 
We are better than the sum of our moving parts but alone we are but mere individuals. We all support each other & we are all open to ideas on how to accomplish a task. Because we have learnt it's about perspective. You have a different angle of view so hence can have a good suggestion. We carried out that installation of 100kW Inverter system with 96 solar PV modules on a 3 storey roof with a team of 6 including myself & accomplished commissioning according to spec in 11 working days. 
 

It is normal practice to have disconnects on the PV wire, even though you already have a DC disconnect on the inverter itself.  Thus far we were very fortunate to not have experienced the consequences of lightning strikes.  We are trying to always disconnect the panels before and during storms, as well as switch of the DC switch on the unit itself.

Switch it off (isolate) when the info becomes available and indications are that it could potentially affect electromagnetic devices.
First indications will be Auroras and we are fortunate that we are not subject to the intensity that centre around the magnetic poles due to our distance from them.
The areas that are subjected to this most intensely are at high latitudes closer to the magnetic poles and where the Auroras are visible regularly.
Looking at the actuals and the forecasts it is at around the max of the last solar cycle and there is no consensus whether its has peaked already or still has to peak. The two dashed read lines indicate the two lines of thoughts.

And some more inconvenient facts and data from the past that NASA and NOAA is trying to destroy by erasing the warm periods from the past and simultaneously erasing the fact that solar activity is actually the driver of climate. Now who would have thought that.

https://sunspotwatch.com/

Would love to get other people's opinions on this one. Is this standard practice?
I've encountered opinions which state to not earth the battery. Ok, so then if you earth the battery. Do you earth the metal casing? or how do people earth their batteries? Thank you.

As a rule i don't use 22 x 58 fuses or fuse holders for battery protection. I have replaced a lot with Keto or Merson fuse disconnect due to the fact that the 22 x 58 fuse holders are always getting hot with moderate to high current discharge. Even worse that they are not even DC rated.

Also:
Pretty much, if it is metal, and it is exposed, it must be earthed - that applies to pretty much everything in SANS 10142-1.

LifePO4 in home use is very safe, user would have to do some really strange $h%& to make them fail catastrophically.
Most of the EV and home storage fires have been with other chemistries e.g. the 18650 lion cells, those things are pretty dangerous in storage solutions when they go south.
There are some really cool youtube vids of testing the different chemistries especially puncture and over charging tests that show how the different chemistries behave.
My Favourite
 

Regarding the Pylontech firmware...  First up, thanks for this thread.  It is without doubt the most useful resource on the internet relating to the issues discussed.
So... I had the issue with the Victron GX + <v1.7 f/w on a couple of my US3000Cs - the misreporting of a voltage spike in one cell causing the Victron GX to panic and shut that battery down.
I emailed Pylontech with a plot of the Vhigh readings (it would shoot to 4.1v for one reading, then back down again).  They responded very promptly (on a Sunday night, too) with a link to download a zip containing 'BatteryView 3.0.33' and a firmware zip file named 'USC_st2.8+NT2.2.zip', contained two binaries named 'US_E2_V2.2_Crc.bin' and 'US_C_V2.8_Crc.bin'.  Good so far.
Then the real fun started...
Couldn't get 'BatteryView' to connect, despite no problems accessing the serial terminal and getting appropriate responses from the batteries.
Tried 'Batteryview' under wine/mono in Linux Mint 21, build a Windows 10 VM, fully updated it, installed Win 10 natively, updated that, .NET 4.8.1, tried every combination of v 3.0.28 through 3.0.33, 3 different USB/RS232 interfaces...  Wouldn't connect, but terminal was fine.
Eventually did a native install of Win 11, same deal... Updated that (which took the best part of a day on a 40Mb/s fibre line) and finally, finally 'BatteryView' connected.  Downloaded all the 'history' and 'event' data, went to update the firmware on the first battery (according to the instructions provided by Pylontech, using the zip file) and...   it threw an 'unknown exception' error and just quit. 😭
OK.
Took a step back, sat down with a cup of tea and thought 'lets have a look at this firmware zip file'.  The one that 'thou shalt not unzip, least ye bricketh the unit'.  Re-read this thread, then thought that if I can work out which of the two bin files, maybe try that?  I mean, who's gonna know, right?
And lo, it came to pass that I could not unzippeth.  Because the %&*$!£ zip-within-the-zip was corrupt.  Re-downloaded from the source (wetransfer.com), same.
Came back here and grabbed the v1.8 zip, all 7 units are now updated fine. 🥳
 
So basically, a massive 'Thank You' to Youda and all who have contributed to this thread.  I hope the above contribution to it might prove useful to someone else facing similar issues.
Now I can hopefully sleep easy, or at least sleep without VRM waking me up with an email at 3.23am to inform me in panicked, breathless tones of some expensive-sounding 'internal failure' or 'cell imbalance' in battery 7 that means you have to get up and shuffle downstairs to silence the beep-of-doom and check that nothing's actually on fire. 🙂

Today I am a VISITOR to my inverter and not a Manager. LOL
Sunsynk have admitted it is a migration issue.
 

This whole migration thing is ridiculous. The public should not even need to be informed of these complexities etc, it should just work. Their IT team need to be sacked. 

At last done with this installation got a few more gray hairs. Installation done at OK Liquor outlet in Witbank( Emalahleni). Total power consumtion during daytime 23kw mostly commercial fridges , upright freezers and some upright wall chillers. Will install 35kva Perkins generator that will arrive on Monday morning. 5kw Growatt and 48v 100ah Easy Power lfp that will run the office equipment.
Scneider CL33 inverter 33kw surge 40kw 2 minutes that is capable of DC-AC ratio of 1:9. The CL 33 inverter will deliver full power at unity power factor with +30% and -5% grid
voltage variation. MPPT voltage range: 200 ~ 1200 V
Full MPPT voltage range: 550 ~ 1000 V
Maximum voltage: 1500 V
Starting voltage: 250 V.                            Number of MPPT's:  3
Solar array consists of 138 x Seraphim 460wp connected 23S6P 63.48kwp.Voc of 1050v @ -13°C at STC 955v.

Hi @Hager
I must admit I've never heard of the Deichmann battery until now 😄
However, I see it uses the rather ubiquitous Pace battery BMS which effectively uses the Pylontech CAN BMS protocol. I've tested a Greenrich battery with the same Pace BMS with the inverter and it worked perfectly fine using the Pylontech settings - no need to do any manual battery configuration. You could confirm this by speaking to the battery distributor (TheSunPays) - I'm fairly certain they will confirm the battery will work with any inverter that can talk the Pylontech CAN BMS protocol. Just bear in the mind the battery will only support a max of 4.8kw discharge power.

The Axperts are many things, but not that smart. We need more info from the PV side

At the bottom of the display, is that a time or a date? If it's 11:08pm then I would expect no PV. Maybe the inverter has some intelligence and switches off the MPPT if it thinks the sun is on the other side of the globe.(or hiding somewhere if the earth is flat😀)

This is most likely the case, but the fundamental problem remains that you are operating on the lower verge of the inverter's Voltage range.
Just be careful to calculate max input Voltage not as 9 x 40V Vmp, but as 9 x Voc and leave room for temperature variation as per the panel's spec sheet and expected local temperature. I expect you'll get closer to 450V than 360V. If in doubt rather use one panel less.

This must be the excuse I've been waiting for to buy a more expensive multimeter 😉

Off-topic Deal of the century on Lithium batteries 🤣
https://www.takealot.com/blue-nova-energy-5-2kwh-lithium-iron-52v-racpower-battery-100ah-/PLID92853341

There might be an additional clause worth considering for "some" before buying from https://electromannsa.co.za/
Warranty:
All inverters and batteries must be installed by a qualified electrical wireman or Master electrician with a valid registration with the Department of Labour. A valid electrical certificate of compliance (COC) must be issued once installed, specific to the installation of the backup or solar system. The installation must be compliant with SANS 10142 and ALL its parts. Any warranty claim submitted without installation photos, a valid COC and proof of a qualified wireman/master electrician installing the unit, can be rejected and will render the warranty claim null at void.

The most impressive battery in terms of space I've ever worked with...9cm thickness footprint. 1C rated & just a beautiful specimen. Will Install another of these in a living room area next week & will share once done. 
There was space in the main DB so no New AC DB required. It's all fitted into one. 

Svolt would need to answer this. 
One can make changes to the BMS parameters with the correct software and the password, although how that would affect the warranty only Svolt would be able to answer.
Some companies like Hubble won't even allow you to connect to the BMS without voiding your warranty.
 
I see it similar to rooting your phone or flashing a non official ROM on your phone, or even doing a software tune or mods on your car... will it affect the warranty? Yes, it most likely will. Some may get away with it, others won't be as lucky.