Axpert invertors

Virge said:
Thanks, then 9x300W will be (very tempted to use the roof surface to the max) but 12 x 250W will be a better choice.

Cheers
Click to expand...

You can always add more strings in parallel as long as the open circuit PV voltage is below 145V - the inverter will only utilize up to 3000W. Remember, with PV you will only get 300W from a 300W panel when the irradiance is 1000W/m2 and the sun hits the panel at 90° and the ambient temperature is below 25°C - in real life (in South Africa) that does not happen 8 hours a day. By adding more PV panels you will ensure that the inverter can utilize more sun for more hours per day.
 
Hi Superdiy, Thanks, I see more light now. I do not know where to find what irradiance I have here in Benoni, Eastrand, Johannesburg. I do understand that even if I get more that 3000w from PV, like from 12x 300w panels, in the event the irradiance will give full PV power it will be only for an hour as an example, the MPPT charger will not process the extra 600w but then I will have more power available over the remaining 4.5 hours. Then I should get 12 x 300w panels. I can not easily add more panels as prices are ridiculously high here and I intend to import them together with mounting kits and maybe PV combiner box with extra spare parts (fuses, surge arresters and so on)
Cheers
 
superdiy said:
They've also mentioned and referred to all the options discussed in this thread already, but as discussed here, all of the proposed connection options were flawed in some or other way. I've created an account in that forum and I'm just waiting for approval, then I'll post a comment in that thread and see if we can get a proper solution to the problem.

The Neutral In to Neutral Out bond, on the inverter, still looks like the most promising option.
Click to expand...

I just ran a quick test on my setup. With Eskom on the inverter output voltages are as follows: L-N = 226V, L-E = 226V, N-E = 0.3V. This shows me that my earth and neutral are correctly bonded at the main supply.

Two scenarios now:
1. Main breaker on the supply DB trips. This is a double pole breaker, so both live and neutral are disconnected. L-N = 230V, L-E = 120V, N-E = 110V. This shows that my neutral is no longer bonded to earth.

2. Simulated grid failure. My municipal supply is via a single pole circuit breaker so the incoming neutral remains connected. L-N = 230V, L-E = 230V, N-E = 0.3V.

So it would seem that the only time the earth/neutral bond is lost is in the event of a local main trip. My supply to the inverter is via a single pole CB, so if that trips the bond is still there.

During load shedding there is no trip or disconnection of either the incoming live or neutral and in this case the earth/neutral bond is still there.

Could someone else do, or has someone else done a similar test and what were the measurements?
 
Gnome said:
On a completely separate note:

I've created a Ruby library for the Axpert (Ruby = works on all OSs). It works well but I just want to create a CLI component for the library before I release it.
Click to expand...

Gnome, great work. I will be happy to help in anyway I can. I know programming but haven't had any experience with ruby though.
 
Schicks said:
I just ran a quick test on my setup. With Eskom on the inverter output voltages are as follows: L-N = 226V, L-E = 226V, N-E = 0.3V. This shows me that my earth and neutral are correctly bonded at the main supply.

Two scenarios now:
1. Main breaker on the supply DB trips. This is a double pole breaker, so both live and neutral are disconnected. L-N = 230V, L-E = 120V, N-E = 110V. This shows that my neutral is no longer bonded to earth.

2. Simulated grid failure. My municipal supply is via a single pole circuit breaker so the incoming neutral remains connected. L-N = 230V, L-E = 230V, N-E = 0.3V.

So it would seem that the only time the earth/neutral bond is lost is in the event of a local main trip. My supply to the inverter is via a single pole CB, so if that trips the bond is still there.

During load shedding there is no trip or disconnection of either the incoming live or neutral and in this case the earth/neutral bond is still there.

Could someone else do, or has someone else done a similar test and what were the measurements?
Click to expand...

That is exactly what is expected. Since your main breaker disconnects L and N, you will loose the N-E bond before the main breaker when it trips. You have one of two options to resolve that issue:
1) Use a single pole main breaker which only disconnects Live
2) Connect the Neutral feeding the inverter to Neutral before the main breaker.
 
superdiy said:
That is exactly what is expected. Since your main breaker disconnects L and N, you will loose the N-E bond before the main breaker when it trips. You have one of two options to resolve that issue:
1) Use a single pole main breaker which only disconnects Live
2) Connect the Neutral feeding the inverter to Neutral before the main breaker.
Click to expand...
The issue here is more regarding compliance with SANS than with safety. In terms of safety, it is completely correct. In terms of compliance it isn't.
 
Schicks said:
The issue here is more regarding compliance with SANS than with safety. In terms of safety, it is completely correct. In terms of compliance it isn't.
Click to expand...

Why is it not in compliance with SANS? Will you please explain.
 
superdiy said:
Why is it not in compliance with SANS? Will you please explain.
Click to expand...

SANS states:
7.12.3.1.3 The supply from each unit which supplies the installation or part of the installation, shall be switched by means of a switch that breaks all live conductors operating substantially together (see annex S), to disconnect the earthed neutral point from the installation neutral when the alternative supply is not connected (see also 6.1.6).

1. Using a single pole circuit breaker as feeder contravenes this as the neutral is not switched.
2. Joining the neutral from before the main switch to the after the inverter means the neutral is not switched.

Since this amendment was made to SANS 10142 (in 2009 I think) it has been an ongoing problem, particularly with automatic transfer switches on 3-phase standby generator plants. Here you have an even bigger problem in that you can create 400V single phase when you switch the neutral. This has been raised with SANS and this clause is apparently under review. For now, we continue with keeping the neutral unswitched and not split as per SANS 10142.

But then again, the last line of the SANS spec might be key - "when the alternative supply not connected". The inverter is permanently connected and always on, whether it's supplying from batteries or not. :confused:
 
Last edited:
I have an interesting scenario, which I hope someone can shed some light on.
The installation is totall off-grid, on a plot - i.e. no eskom for backup. There's no cables to the street, nor were the any cables in the house to begin with. There's a 5Kva Axpert + 6x 250W solar panels + 48V battery bank. A new ACDC prewired DB board was fitted and plugs and lights installed (surface mount), and an earth spike put into the ground outside. The problem is, the Earth Leakage doesn't trip (I have an earth leakage tester). Upon inspection, there's no earth wire connected to the Earth Leakage (I think it should have gone to the neutral?), though I'm not sure the neutral and earth should be connected.

The question is, what do I need todo to get the earth to work properly?
 
Schicks said:
SANS states:
7.12.3.1.3 The supply from each unit which supplies the installation or part of the installation, shall be switched by means of a switch that breaks all live conductors operating substantially together (see annex S), to disconnect the earthed neutral point from the installation neutral when the alternative supply is not connected (see also 6.1.6).

1. Using a single pole circuit breaker as feeder contravenes this as the neutral is not switched.
2. Joining the neutral from before the main switch to the after the inverter means the neutral is not switched.

Since this amendment was made to SANS 10142 (in 2009 I think) it has been an ongoing problem, particularly with automatic transfer switches on 3-phase standby generator plants. Here you have an even bigger problem in that you can create 400V single phase when you switch the neutral. This has been raised with SANS and this clause is apparently under review. For now, we continue with keeping the neutral unswitched and not split as per SANS 10142.

But then again, the last line of the SANS spec might be key - "when the alternative supply not connected". The inverter is permanently connected and always on, whether it's supplying from batteries or not. :confused:
Click to expand...

There must be a simple solution to this. These inverters are installed daily - how are they connected, or are they simply not connected safely and/or complying to SANS?
 
SoftDux-Rudi said:
I have an interesting scenario, which I hope someone can shed some light on.
The installation is totall off-grid, on a plot - i.e. no eskom for backup. There's no cables to the street, nor were the any cables in the house to begin with. There's a 5Kva Axpert + 6x 250W solar panels + 48V battery bank. A new ACDC prewired DB board was fitted and plugs and lights installed (surface mount), and an earth spike put into the ground outside. The problem is, the Earth Leakage doesn't trip (I have an earth leakage tester). Upon inspection, there's no earth wire connected to the Earth Leakage (I think it should have gone to the neutral?), though I'm not sure the neutral and earth should be connected.

The question is, what do I need todo to get the earth to work properly?
Click to expand...

You need to add two things
1) Connect your Earth and Neutral together at the Axpert output (bridge earth and neutral).
2) You need to sink an earth pole into the ground. Generally a copper rod about 1.5 meters into ground and a wire running from that and connect to the earth of your Axpert.

Number 2 is because the stuff in your house are at earth potential. So you need to make sure you Axpert neutral and earth are both also at your local earth potential. Therefore if anything touches earth, it'll cause the current to "leak" and the EL to trip.
 
SoftDux-Rudi said:
I have an interesting scenario, which I hope someone can shed some light on.
The installation is totall off-grid, on a plot - i.e. no eskom for backup. There's no cables to the street, nor were the any cables in the house to begin with. There's a 5Kva Axpert + 6x 250W solar panels + 48V battery bank. A new ACDC prewired DB board was fitted and plugs and lights installed (surface mount), and an earth spike put into the ground outside. The problem is, the Earth Leakage doesn't trip (I have an earth leakage tester). Upon inspection, there's no earth wire connected to the Earth Leakage (I think it should have gone to the neutral?), though I'm not sure the neutral and earth should be connected.

The question is, what do I need todo to get the earth to work properly?
Click to expand...

Edit: Sorry, Gnome beat me to it. :)

Ah, a simple installation with a simple answer. The Neutral on the inverter output needs to be bonded to the Earth on the inverter output and that Earth needs to be connected to the Earth spike. The Output of the inverter then needs to go via the Earth Leakage unit (ELR) to the circuit breakers for the various circuits.
 
Last edited:
superdiy said:
There must be a simple solution to this. These inverters are installed daily - how are they connected, or are they simply not connected safely and/or complying to SANS?
Click to expand...
The simple solution is to not split the neutral according to the SANS spec. That way, in all scenarios, and IMO, you have a safe installation. Which is what you want.
 
Schicks said:
The simple solution is to not split the neutral according to the SANS spec. That way, in all scenarios, and IMO, you have a safe installation. Which is what you want.
Click to expand...

Just for clarity. Are you referring to the Neutral In to Neutral Out bond?
 
Ok so I've written a small library for interaction with the Axpert

It is still work in progress but if you'd like to try, give it a bash.
I have only tested on Mac and Linux. I haven't tested Serial Port libraries for Windows. It is on my "todo" list.

Instructions:
Requirements:
- Ruby (google for instructions to install, think it is default installed on most flavors of Linux and Mac)

Rubygems: (These instructions are for Mac and Linux)
Code: 
gem install serialport # may require sudo
gem install axpert_rs232 #may require sudo

Sample code: (run 'irb' in terminal/command line/bash)
Code: 
require 'serialport'
require 'axpert_rs232'

# => Set your serial port location, ls /dev/ and search for something that makes sense
serial = SerialPort.new('/dev/tty.usbserial', 2400, 8, 1, SerialPort::NONE)

# => Show available commands
AxpertCommands.constants

# => Some current device status info
AxpertCommands::DEVICE_STATUS.issue_command(serial)

# => Some current device rating info
AxpertCommands::DEVICE_RATING.issue_command(serial)

# => Set battery type to user agm
AxpertCommands::SET_BATTERY_TYPE.issue_command(serial, :agm)

There are plenty of commands but I don't iterate over all of them here.

Testing and comments would be welcome.

This was written pretty quickly, so there is probably quite a bit more work required and there will certainly be bugs.

My hope is to move forward with this to something that can upload data to a web-page or perhaps run a local web server.

Btw. Ruby runs on Raspberry Pi which was part of the choice for the language (other being portability is pretty decent)

EDIT: Apparently serialport (Rubygem) is supported in Windows also...
 
Last edited:
Hi there all.

I have just completed my Axpert MKS installation. I am running 2 5 kva inverters in parallel. I do not have panels on yet that will come in November. I just want to cover load shedding at this stage as the generator is noisy. All is working fine but the 2 inverters are using a lot of power when they are in Bypass / Line mode getting the power from Eskom. The batteries are full but the 2 units use about 1.2 KW all the time. I made a video to show what is happening. Did any of you get this problem and what did you do to solve it.


This is my wiring with the switches.

diagram.jpg


Video 1

Video 2

And I see this person also has this problem.

Video 3
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X