Thank you @Chris Hobson. I thought setting 38 was to bond the neutral to earth?

7 minutes ago, Chris Hobson said:

Ok I was beginning to wonder with two arrays producing exactly the same power but I see there is some variation.

The arrays are very close to each other when running close to full capacity. If the load reduces, the gap becomes substantial, as array 2 would run flat out and array 1 will do the controlling up or down to match the load.

It is probably better to illustrate it in graph form.

@Don for what it is worth in helping to sort out your problem, Mike Thorne installed 2 additional ELB on my system apart from the one on the main DB. One is for mains to the inverters. Then there is another one for mains to a changeover switch that allows me to switch straight to mains and thus totally isolate the inverters for maintenance or a proper cold start. So far, since June last year, when my first 3Kw Axpert was installed, I have never had an ELB trip even after replacing that inverter with a 2 x Axpert MKS 5Kw cluster .

Regarding the use of setting 12 (Back to Grid) for intermittent switching to grid, I was advised by Jaco Fourie to set this to a low value so mine is set to the default of 46V. The reason it that the Axpert will interfere with the SOC switching controlled by my BMV700. Likewise is setting 13  (Battery Full) set to 48 volt as this should also be controlled by the BMV and you don't want the Axperts to interfere with the operation of the BMV.

As to setting 16 (Charge Source Priority) I prefer CSO  as it will only charge from grid when there is no solar available. To further prevent grid charging my setting 11 (Max Grid Charge Current) is set to 2A for reasons already explained.

My objective is to minimise grid charging but I fully understand that with frequent high loads, your settings may be different and if it works for you, good and well. However there is a distinct difference between setting 16's choices of CSO and SNU. The latter will charge from both grid and solar and in the example that you gave, your batteries were already at 97.7% SOC so the internal resistance would be high thus allowing a low charge current which was supplied by solar. I am convinced that my suggestion on diverting to grid to cope with intermittent large loads of relatively short duration would solve many problems. You have already highlighted the problem of cloud cover adversely affecting solar generation. With our winter coming, I will have the same problem and may just have to run on SOC and put up with grid charging or switch manually when I am at home. However that is not the object of the whole ICC project which aims to automate control of the inverters to the maximum sensible degree. Since @Manie is already severely under pressure and is too kind to tell lazy and incompetent folks who ask for hours of help on trial ICC programs to go play in the traffic or try Google to look for answers, perhaps all our discussions will enable us to come up with some useful suggestions to improve ICC, based on sound experience and explained in detail. That at least will hopefully give him more time for development.

I dread the day when @Manie gets fed up with the whole thing and hope we never see that. We have both a moral but also a self-interest obligation to help as far as possible and to promote what in my experience is an outstanding piece of software for the purpose at an insignificant price compared to its features. It deserves commercial success so I have zero hesitation in promoting ICC wherever I can. If others have programs for the purpose, let them promote it and those who wish to try it would be free to do so.

Finally I am not trying to convince you that my settings are better, just to give the reasons for them to add to collective knowledge on the forum. I would thank you for having also conducted your postings in the spirit of fact based co-operation.

 

2 hours ago, Don said:

Thank you @Chris Hobson. I thought setting 38 was to bond the neutral to earth?

The arrays are very close to each other when running close to full capacity. If the load reduces, the gap becomes substantial, as array 2 would run flat out and array 1 will do the controlling up or down to match the load.

I don't think that is necessarily a problem. I run 2 MPPTs with 2 arrays, and the slight differences in measured voltage between the two cause one MPPT to still run bulk charge, while the other starts reducing the amperage, exactly like you see above. You have inverter voltages of 229.8V and 229.2V, but the string measuring the higher voltage, has lower PV watts.

1 hour ago, DeepBass9 said:

You have inverter voltages of 229.8V and 229.2V,

That is inverter output AC voltage and has nothing to do with the PV voltage. With no load, the PV voltage on both arrays goes to 125.6 Volts or something at 0 amps. As they start to take load, the voltage would drop and the amps would increase. As you can see above inverter 2 is basically running flat out at a voltage of 96.4 at 30 amps. Inverter 1 is doing the controlling. Therefore, inverter No. 1 is running at a PV voltage of 107.2 V and at 26 amp to ensures the PV output matches my load. As the load increases, the PV voltage on No. 1 inverter would drop and the amps would increase and vice versa. If my load increases to above what the panels could produce, both arrays would be running at 96.4 V and 30 amps. 

3 hours ago, ebrsa said:

Regarding the use of setting 12 (Back to Grid) for intermittent switching to grid, I was advised by Jaco Fourie to set this to a low value so mine is set to the default of 46V. The reason it that the Axpert will interfere with the SOC switching controlled by my BMV700. Likewise is setting 13  (Battery Full) set to 48 volt as this should also be controlled by the BMV and you don't want the Axperts to interfere with the operation of the BMV.

I agree with that. You make sure that your inverter settings do not interfere with ICC settings. 

3 hours ago, ebrsa said:

As to setting 16 (Charge Source Priority) I prefer CSO  as it will only charge from grid when there is no solar available. To further prevent grid charging my setting 11 (Max Grid Charge Current) is set to 2A for reasons already explained.

The only problem I had with the CSO setting was that with a lot of clouds around I would only have say 150 watts available from 16h30 ending up at 50 watt at 19h00. Only once the PV dies completely and disconnected from my inverters does charging commence on my batteries. If I am going to charge my batteries from grid in any case, they might as well start charging at 16h30. It will take maximum watts from the PV panels and the shortfall will come from grid.

How long does it take to charge your batteries at a rate of 2 amps?

3 hours ago, ebrsa said:

I am convinced that my suggestion on diverting to grid to cope with intermittent large loads of relatively short duration would solve many problems.

The only way to get around that would be to add a timer to the control. You would set it, i.e. >3000 watts for 10 minutes, switch back to grid. Therefore, you can switch on the kettle and battery discharge rate drops to 3200 watt, but only for 4 minutes until the kettle boils and then disconnects - Nothing would happen. But if your wife decides to switch on the tumble dryer, that is a different story. If the high load exceeding 3000 watt still exists after 10 minutes, it would then switch to grid.

11 hours ago, Don said:

Thank you @Chris Hobson. I thought setting 38 was to bond the neutral to earth?

Earth/ground is an alternative pathway to source. If your neutral gets damaged and perhaps the body of an appliance gets energised you want to  provide an alternate pathway for current flow and this flow is back to source. When you are in solar/battery mode the inverter is your source. When in bypass/grid mode the transformer down the street is your source. Hence the need to close or open the relay between the inverter earth and grid earth.

If you really want to understand earthing requirements watch this video (disclaimer it is 1 hour 15 minutes long). I really found it useful and cleared up some fanciful thinking that I had been taught by people who should know better.

I see he has a section of 9 videos on solar I have not seen before. Perhaps I watch some on Sunday evening.

 

21 hours ago, Don said:

How long does it take to charge your batteries at a rate of 2 amps?

@Don I don't actually charge batteries from grid if I can avoid it which I fortunately have managed so far. Two days ago we had clouds in the morning and I manually switched to grid for about an hour as the batteries were down to 71%. I will set grid charging to 0A and see what difference it makes. On the West Coast (100 Km North of Cape Town) with winter rainfall, we only have infrequent clouds and sea fog in summer which affects me as the house is only 400 meters from the beach. I fully expect that things are going to be very different in winter but will then adjust my time on Sol/Bat to ensure the batteries stay above 70% SOC or else activate SOC based ICC switching.

At 9:30 this morning I has enough solar power to supply the house and charge the batteries at 9A. The batteries were down to 72% having supplied the total load since midnight. Now at 12:30 the battery charge current is around 23A and grid usage since midnight last night is 0,001Kwh. The batteries of 450 AH is now at 82.9% SOC and usually are at 100% by about 15:30. I may have to add another 3 panels in winter but then I will have too much solar in summer which will go to waste as the batteries are always at 100% before solar becomes too little and ICC switches to grid at 17:15 until midnight.

As it is only my wife and I in the house and we cook on gas and some electricity, grid usage is not excessive. Domestic hot water is also heated with evacuated tubes. Grid consumption has dropped from 1015 Kwh during January 2015, when the load was all grid supplied to 332 Kwh January 2017. I am just glad that so far this year there was only one morning when I switched to grid for an hour due to clouds. But our individual weather conditions and load requirements will obviously differ significantly given the differences in solar availability during the seasons and load requirements. All I am can contribute to is submit my experiences in the hope that it may benefit others.

Today a 5 minute peak over 1% deviation. Perhaps by next week we will be back to normal.

Good morning
I have been trouble shooting my batteries for a while now. Changed back to grid to 48. My batteries will discharge very quickly after reaching 49 volts. This morning I did a load test on each battery and found that all three of the batteries connected to busbar on the positive leg are in a poor state. Less than 10 volt on load lest.
I have 12 x 250 amp Sonic batteries.
I have changed over to utility

Dave Morley

Charging each bank of batteries individually to see if this will bring the batteries back.
Is it advisable to rotate the batteries in the string?

Dave Morley

Dave, you could have problem cells in them batteries. It is best to remove them.

Yes, you can swap batteries around in a string / bank.

Only rule there is not to mix new batteries with old batteries, which is not the case here at all.

2 hours ago, Dave Morley said:

My batteries will discharge very quickly after reaching 49 volts.

I've seen the same thing when batteries started to go bad. On my 24V system it would go well up to about 24.2V, then it would drop to 24V fairly quickly, stick around a few minutes, and then it would drop off a cliff after that towards 21V.