anotherbrownbear

ABB has sold their solar manufacturing facilities to FimerSpa. This includes all three R&D centers as well as the Power-One factory where the string inverters are manufactured. As for the products, they are still exactly the same and operate with the same functions and still 100% compatible with Victron. Only the name changes. We were the technical partner for ABB string inverters in Sputh Africa and we are now the technical partner for FimerSpa.
When trying to decide whether to use AC or DC coupled, you have to look at a couple of things. What time of day is your load mostly. How big is you system and to an extent, what size battery do you want to put in.

@anotherbrownbear should be able to help you.

Also just keep in mind, if you do bond neutral end earth on the genny and take it to your earthing point, make sure your neutrals are also switched through your change over.

You are correct that is exactly how eddy currents are created, by nett magnetism. It absolutely exists, no argument.
An asymmetrical AC current flow will create a nett magnetic effect, and current there and current back in close proximity  will cancel the magnetic effect, exactly what you were taught in high school.
If there is no nett magnetism and therefore no transformer effect. Your teacher was spot on, keep the ac wires together.
This is the science behind the "twisted pair".
The thing is though eddy currents although they cause losses, they are inherently self-short circuiting in nearby interconnected masses of metal. Voltage doesn't build up on a short-circuit.
This isn't what you are dealing with on this roof, if you can get a shock you are dealing with a potential difference relative to somewhere else. 
Eddy currents are analogous to the excitation caused by microwaving food, just in metal. They aren't that at all, but that's a good way to think about them.
I think you are dead right about the MPPT frequency playing a role, just the electrical mechanism isn't inductive, it's capacitive.

I may have the cat by the tail here. I was taught in high school, in my electricity class (cause we had tech subjects, it was kind of a cool school, we could do that instead of biology and useless things like that), that you must put live and neutral in the same conduit otherwise the round metal pipe of the conduit acts like the secondary of a 1:1 transformer (with a really large loop). The teacher said it generates an eddy current in the pipe. I never questioned this "fact". I don't really care what you call it, I think the science is sound.
Also, heard horror stories about ham radio operators that had a neighbour that installed a new solar system... 🙂
 

@Jaco de Jongh, Ah this is something different, still not eddy currents though.
Eddy currents by their very nature are current flow internal in a metal and are effectively flowing in a short circuit so there is very little voltage. 
This is induced current, but no real potential build up, think of a lump of metal as being made up of windings of a transformer but each winding being shorted out. Plenty internal current rushing about and plenty heat but no voltage build up to speak of. 
I think your dealing with the capacitive effect of an electrostatic charge which does have plenty voltage, but normally not much current.
Like the little shock you get when you touch the door handle from some carpets.
You have all the components a large plate being the panels and the roof combination, the other plate being the ground, and high frequency MPPT switching.
The higher the frequency the lower the capacitive reactance (it is inversely proportionate to frequency) so it could probably develop enough current to deliver a nice shock.
Mmm, interesting.
 
 

Eskom can be saved, but 1) at great cost, and 2) you have to convince the unions that something has to give. The unions have to be a part of the solution, they have to be on board. Anyway, that's what the EdCon guy says. They had to let people go, they had to close stores, and everyone around the table, including the unions, first had to realise and acknowledge that.
I'd say the first thing they can do is take the budget they want to use for the NHI and use that to save Eskom. Can't run hospitals without power anyway...

I'm sorry @Hercules Weyers, but you appear delusional when you claim:
(a) there was no over-current protection (thanks @plonkster), and
(b) there was no fire—only a DC arc.

You also imply that the owner is certifiably insane, by claiming they had the system redone "using exactly the same modules". Can you provide any evidence for this? Or for any other of your claims that weren't already disproved by the available evidence before you made them?

Blue is the fuse. Pink is the BMV shunt. Of course it is impossible to say if it was the right kind of fuse (it looks okay) and whether a single fuse was sufficient (probably not).

I think the main reason I was quite ready to believe that the fire was caused by one of these modules (middle one most likely from the flame pattern) is because we've seen the video TTT linked above, and the hot-glue insulated terminals on the back combined with a loose ABB contactor. But I was also told that this was a "demo" model of sorts. And I did note on later youtube videos that the terminals seems to be on the front now, so clearly this was improved.
My skepticism about them really being capacitor modules remain. We don't know what is in them. I've heard various claims about it being "capacitor cells", with absolutely no material as to what that might be. So my best guess: The bulk of the capacity is from some kind of lithium chemistry. So it's possible that it's not a bad battery... it's just not going to make the advertised millions of cycles. I'd love to be proven wrong... preferably with a proper documented deconstruction :-)
Edit: Further to that, another reason for my skepticism: We've been through this before. They were called lead crystal batteries and thousands of cycles were promised. In the end.... they weren't bad batteries, they were exceptionally good AGMs. But they didn't live up to the claims.

I'm too scared to tell you to do it that way. I googled this a bit, the same thing is discussed here at MyBroadband by people who also seem to understand what they are talking about, and also here by some international people. Once you create a second neutral bond, from that point on your earth wire carries working current instead of just fault current. It's actually not to be taken lightly, you really don't want to do it. It also seems that SANS requires an ELB in front of a UPS.
It is also possible that your inverter already does the bonding internally. All it has to do is connect earth to neutral inside the inverter BEFORE the transfer switch. You can easily test this. On the output of the inverter (or any socket really), measure the voltage between neutral and earth while on the grid and also while on the inverter. It should be zero. If the inverter is internally bonded, it will remain at zero when powered by the inverter. If the inverter is not internally bonded, then there will be a potential difference, usually it's in the range of 30V to 120V and not capable of much current (but enough to tickle the operator with the soldering iron :-P), but a good DMM will pick it up.
Also, in the case of the Victron, the inverter bonds neutral to the CASE of the inverter (which you are supposed to ground). It is possible that this is also internally done in the Axpert and that all you need to do is ground the case of the inverter to the earth bar in the DB.
We really need the Axpert guys to weigh in here. Given how common these inverters are by now, I would really actually expect that the internal bond should be there by now... :-)

There are a few systems in South Africa. The latest one unfortunately not good due to (in)famous Eskom expertise.
Eskom, and South Africa, has two pumped storage schemes, both operated in conjunction with the Department of Water Affairs and Forestry as part of water transfer schemes. Palmiet Pumped Storage Scheme in the Western Cape can be used to pump additional water from the Palmiet River via the Rockview Dam to the Steenbras Dam to supplement the Cape Town water supply. Drakensberg Pumped Storage Scheme in the KwaZulu Natal Drakensberg is able to transfer 20m3 of water per second from the Kilburn Dam to the Sterkfontein Dam from where it flows into the Vaal River to augment water supply to Gauteng.
 

I'm going to tell you to go with the Growatt. Well... I want to tell you to buy the Rolls, because neither the Axpert nor the Growatt has paperwork for grid connection (which might not matter right now unless you are in Cape Town)... BUT... the Growatt has at least some proper paperwork beyond a "CE" mark.

I did ask a question. How does he get the 400V VSD to work off of 1 x 400V supply.

The sparks is because the big capacitors inside the inverter has to charge up, and when they are completely discharged they behave like a dead short (hence the spark). What you do to avoid the spark is "precharge" it, that is charge it slowly through a resistor. And that is what @Javi Martínez and @Ingo is talking about. Once the caps are charged, you can connect the cables without any sparking.

Right. That's what I noticed with ABB stuff... generally the extra money you pay does buy better quality. For example, I bought two ABB RCDs during my nuisance tripping hunt, and both of them were rated for more than double the mechanical trips of the Hager breakers I sometimes use. I expect the same recipe extendes downwards. Still... there is a pain threshold around 4k... at least or me 🙂

You must at least have a good fuse with those things... I'd hate having to pull those apart in an emergency. Even the Mersen one I advised... man pulling that thing under load makes a nice big arc.

I get all my training on the interwebz... thanks for this!

Gees, do you need a job in our workshop? Can use those magic hands to get the big cables in small holes. Hahaha.
 
Bootlace poses different problems when you try and get it in small spaces. Sometimes the square shape makes it harder to fit in.
So we use pin lugs instead.

The old farmer way of crimping welding cables involved using a bolt and a hammer, basically indenting the lug in the middle so it pushes down on the cable. I would suggest this above just squashing the thing in a vice, and even then I don't think it is good enough (long term) for solar use. What you need is more than just "it won't pull out". You want the best possible contact between the lug and the cable, that is you want as much of the surface area (cylindrical in shape) of the one to touch the other, and the only way to really get that done is to crimp "inwards" at great pressure (several tons), as a hex-crimper does.

For a DC breaker to work properly, it should have polarity sensitive magnets.  And size does matter in this case. Just to compare, the Noark breaker takes up the same amount of space as a normal AC DIN circuit breaker. and is rated for 220v per series wired pole. Same as ABB S280 range. But if you are looking for higher voltages, you need a bigger unit. Roughly double the physical size of a "normal" DIN breaker. The ABB S800PV range come in ratings of 800, 1000 and 1200VDC for 2, 3 and 4 pole units. Price also increases quite a bit.

This is not the issue
This is the issue
 
I have put the question forward to the Electrical contractors association for some official answer and will forward it as soon as possible. But, the bottom line is, it`s not legal to have two power sources on one socket outlet.

For all contacts, it can do both AC and DC. That's just physics, if there is continuity, the electrons will flow. What comes after that is the breaking capacity. If you consider what happens when you open a switch in slow motion. First there is just a fraction of a millimeter that is open and the air gap has a resistance too low to stop the flow of energy, it ionizes the air and jumps across. Then as your slow motion movie plays on the gap widens, but ionized air has a lower resistance so the arc grows. At one point the distance is finally so large that the arc cannot be sustained and the flow of energy stops.
All the while this little welding arc is going it heats up the air, causes pitting in your copper contacts, and so on. Which is why relays are rated for a certain number of operations, usually around a million. They actually wear down.
If the arc is bad enough, it sets things on fire.
Now with AC, there is the additional deadening effect of the voltage going through zero 100 times a second. This allows for a higher voltage, usually between 5 and 10 times more.
So in general, if it has a DC rating, it will do an AC rating of the same. And even if it only has an AC rating, it will still handle DC of a much lower value, though you don't know what that value is unless the supplier tells you.

In can run in the same wireway but as soon as it enters a DB it should be segegrated with dedicated disconnects or if you have a DB with multiple rows, they should be on seperate rows and clearly marked. The idea is to get the supplies as far away from each other as possible. When you have socket outlets in your house that is used for backup and some that are not on backup, they should be clearly marked. You actually get different earth pin configurations for them. Normal is round earth pin, dedicated is flat on top and UPS is a quarter turn flat earth pin.
On the new regulations, the normal socket will be Euro and the dedicated an UPS will still be as is. You will actually not get a legal CoC when you have two supplies in one socket outlet.

Not too sure about this one. Two supplies in one socket outlet? Hopefully clearly marked.

In fact, the more I think about it, the more I think that none of this even really applies. We're not creating an island. From the perspective of the AC-in side of the inverter, we're always avoiding the creation of an island.
From the perspective of the whole, we're intentionally islanding and forming a "microgrid", but I wonder why that might even matter, since the entire backup part lives on the other side of two series disconnects. There is also the matter of a Fronius PV inverter tying to the output of the Multi: That has to be isolated from the grid (of course)... but since the Multi has all the stuff you need (anti-islanding, two series disconnects, etc), it seems to me that all of this is a red herring. Let's hope the higher-ups agree.