root

One of the big factors for me going with ICC was SOC control. Meaning that ICC reads the pylons and tells the inverter what the SOC is. Using inverter measured bank voltage as a SOC indicator didn't work since the lithium pylons have a very flat voltage discharge curve. Half a volt could be as much as 20% SOC. So to get SOC accurate you need the BMS to tell you what the voltages and real SOC is. I believe this is why the OP inverter dropped out before the bank was even close to empty. I experienced the same with our setup prior to getting ICC. This is with the correct settings for battery voltages in the inverter. I believe this SOC control is also why some manufacturers that isn't able to read the BMS might not recommend pylons.

Thanks all for the responses. We are starting to get this thread constructive again. When I wrote my reply ICC was shot down and made out to be junk. Which it isn't and since I'm also fairly new to the forum I missed the earlier software wars. Small mercies. I use ICC in two large off-grid setups. Axperts and a bunch of pylons. It works beautifully for me given my requirements. Sure the Victron stuff is great but it is expensive. I also don't think given the current legal framework that ESS or going hybrid can even be used. Please correct me here if I'm wrong. The core of my initial post was supposed to be, to tell the OP to try out ICC first given that it will solve his problems for the least amount of money. It will also help diagnose the issues he currently has without having to replace invertors. There is a trial option for ICC and it should give him a glimpse of what is currently available. Lastly in my experience I've see quite a few one man github/lab projects run a lot longer and be better supported than a corporate setup. The corporate project gets canned quickly and the source code is then lost forever in the IP law dungeon. Sure Manie's stuff isn't open source now. But I'm hoping it would be one day if the decides to let go of it.

Discharge control should be catered for by sizing the inverter and batteries correctly. I don't understand how you could control battery discharge in any other way. The load wants what it wants. Charge current control is set in the inverter and is facilitated by the software. SOC control is key since the OP had his inverter switch off before the battery reached its cut off. Since a ups and a off-grid inverter is pretty much the same thing. One derives its power from the grid the other from solar. Not sure why you are going off on this warranty tangent. The software is cheap compared to a colour control setup and venus etc and most things blue. Which you would need even with a blue off-grid/ups setup. I know the blue stuff has on grid and grid tie benefits but both those are useless given the current legal framework and red tape. You of all people on this forum should know that software evolves and improves with time. What you saw a few years ago and whats currently running is most probably two very different things. Given that I didn't need to write my own hack job by your standards and could purchase something that works is a win. O and thanks for quoting only part of the sentence and losing the context it was used in.👏

What you call a hack job is actually a very good and mature product that would have assisted you in diagnosing your problem and quite frankly it would have avoided it completely. Since it manages the batteries and inverter properly you won't need to spend anything on a new inverter. You call it expensive? A R500 Pi, a R300 cable, a SD card + housing for the PI. Software licence with active development and support for less than R1000. But you consider a blue inverter and accessories that is about 4x more expensive... But each to his own. I've always been more of a build something that ticks all my boxes person than spend a mountain of money for something that doesn't quite meet my requirements.

It could also be a micro SD card on the way out. They have a finite amount of read/write cycles. The timeline you are describing points to the SD card as the most likely suspect.

And this morning they are all gone again. Back to w27 🤔

Ha! Week 34 just got uploaded too. Someone must have woken up today.

So after being AWOL for a few weeks the Eskom system status reports are back. They got stuck on week 27 for a few weeks. Up to week 31 got published now. Even though we are in week 36 already. http://www.eskom.co.za/Whatweredoing/SupplyStatus/Adequacy Reports 2018 ppt/Weekly_System_Status_Report_2019_w31.pdf Some additional info made their way into the reports though. The 2 private OCGT peaking plants (Avon and Dedisa) are included in the "installed capacity". There is also a new slide just for renewables. I like these reports since it gives you a insight into when Eskom is planning to do maintenance and how well(badly) they are doing generation wise week on week. According to the week 31 report they are starting to ramp up planned maintenance, 4000-5000 MW is being pulled from capacity starting this week. We all know what that means... Link to the reports: http://www.eskom.co.za/Whatweredoing/SupplyStatus/Pages/AdequacyReports2018.aspx

@Youda How did you mount the US3000s in the rack? Did you just use the standard spring clip captive nuts or did you install a rail of some sort. These bricks are heavy and I'm not quite comfortable installing them just using the 4 front holes.

Use the Sonoff to switch a beefy solid state relay... No burn marks.

Also got several packages from Amazon and Banggood this year. Plus a few other obscure Chinese sellers. No issues.

Very interesting. Now I also know that the bump is indeed normal. Below is a shot of our pylon 6x US2000B (3S2P) bank from 10 Aug. Due to rain, cloud and crazy misty days the bank hasn't top balanced since 12 Jul. So close to a month. The bank cuts over to grid at 25% SOC. But due to internal draw it can go down to 18% ish overnight since it only gets solar to recharge. Bump is from 89% SOC to 100% SOC. In the past its been from 90/91/92. So there is some drift. But not too bad for 30 days no top balance.

I'm puzzled why your isp or you set your dns servers to the avanti communications ones in London...? dns=88.210.184.129;88.210.184.130; United Kingdom - Avanti Broadband Ltd

What model/capacity BlueNova is it for interest sake? Also please update us if the BlueNova fails or gets replaced.

Just as a warning. Wind turbines are noisy. Very noisy. I don't think you would be able to live in the house with one above the roof.

Where are you based Harry?

I don't trust anyone that leaves the clear remove on first use sticker on their devices. Shame! Shame! Shame!

@Gary have you considered that it might simply be a hardware issue. We had to send one axpert 5kva back that had a bad spikey mppt back to be replaced. Its brother from the same batch had no problems. We only managed to figure this out with proper monitoring and looking at the data. Maybe consider getting ICC and capturing some data. With data a proper analysis can be done and problems sorted. Best wishes and stay strong there in Zim, out hearts bleed for what was once a beautiful country.

Its not a micro grid. The estate has a bulk connection to eskom and resells (not to be confused with the solar power generated) to the estate residents. Now unto the selling of pv power back to the estate. This in my mind is no different to selling back to eskom/CoCT etc. The bulk connection probably has a eddy current meter (spinning wheel type) so it will do feed in just fine. The estate is probably a net consumer so feed in will go unnoticed. I don't think it is legal to do so but I personally have no problem with them doing so due to the lack of proper regulation. Do I think its a bad idea for the estate to openly encourage this practice... yes.

Now you get why I said your roof determines your strings... Put them on the roof first, then only do the cable runs once you know which panels go onto which strings.

you can debate strings as much as you want but in my opinion and experience your roof space determines the strings. You want all the panels in each string to be facing in the same direction. These panels are deceptively big. You run out of roof space quite quickly. So first measure up the area you want to put the panels on and work out how many you can fit in different orientations. I would also try and get one string to face NW to get some of the afternoon sun. The rest you can fit in the NE direction as per your diagram.

I suspect they will stabilise around 90% capacity. This is what is seen on the tesla batteries used in their range of vehicles.

If you are looking for the source of this chart. It is freely available on eskom's website. They are a bit slow to update it so the pdf is usually a week behind. http://www.eskom.co.za/Whatweredoing/SupplyStatus/Pages/AdequacyReports2018.aspx

Something along those lines yes. Pull info via RS232 and RS485 ports. Extract, Translate and Load. Then present the data in a friendly sensible dashboard display and maybe add a dash of monitoring too.

Time to introduce myself before I derail threads by doing it there 😅 System engineer by day. Building, designing and sizing integrated revenue management systems for foreign governments. My father is a Pr.Eng Heavy Current. We currently have 4kw of solar on the roof. 5kw going up on the new house connected to 2x 5kw Axperts and 4x pylon us2000b. Also got 4x us3000b in boxes that will form part of the current system replacing the flooded leads. Planning on starting a few threads in future since the above two systems will get a proper monitoring suite and dashboards sitting on AWS servers abroad. The idea is to document the build of these monitoring tools so someone else can also do it for themselves. Hello world 😉