p_i

So tonight I noticed something was off with my SoC graph so I went to investigate. Sadly this is what I found: one battery that switched itself off. I suspect water got in between the black back portion and the white front portion. I found a water droplet at the bottom of this seam so it either ran along the side or straight through. I believe this won't be honooured as a warranty claim so is there somewhere/someone in Cape Town who can diagnose and repair? Thanks in advance. Back story: builders decided to open up the upstairs cavity walls yesterday well knowing it's going to rain today and water simply followed one of the solar cables from above and collected inside the solar junction box and from there it dripped onto the battery. And yes, their waterproofing attempt obviously didn't cater for this ingress. 😭

Do you mean your PV is only physically connected to the master? Or do you mean only the master is pulling power from the PV connected to it and the slave is not pulling power from its PV? If it's the first scenario, your slave will get power via the batteries. You can try changing the priority setting to battery first. Maybe the slave will then pull from the grid instead. Either way, power from the batteries is still cheaper than power from the grid. PS: the master and slave will share settings so any setting you change on one unit will replicate to the other unit after a few seconds. At least, that's what happens on my Sunsynk units which are based on Deye hardware.

I wanted to share this to show the multi-point scanning option at work. Green is the power from my NW roof and blue is the power from my NE roof. NW roof has some shade in the mornings and I've always struggled to get it kickstarted in the morning. With the recent addition of the NE roof and the new firmware option, it's interesting to watch the firmware option at work. Yellow and orange are the voltages and it seems the inverter varies that at times in an attempt to get more power from the panels. There's several points on the graphs where the voltages were varied and the power immediately increased. PS: the power and voltage values of the NE roof aren't really negative; it was the only way I could (think of to) get 4 strings into 2 graphs.

If you clamp the main grid live wire (red) from the council, it's correct. Do NOT clamp it to only ONE of the inverters incoming feeds, since then it'll only measure half of the power consumed by the two units.

Also make sure the CT is clamped to the grid live wire that feeds BOTH units. It won't function correctly if it's only clamped to the master's incoming feed.

I've noticed there are some extra options on the app but it seems you can't use these if the firmware doesn't allow for it (I tried to enable "MPPT Multi-Point Scanning" in the app but this is not visible on the inverter screen). Does anyone know what the "PV1 Selfcheck" and "PV2 Selfcheck" options do on the Advanced Settings section on the app?

Did you notice any changes on your setup screens?

You simply wire them to a 12V DC power supply; red to positive and black to negative. You can use the old style black connector blocks or anything of the more modern connector blocks. And yes, I snipped off the connectors that the fans came with.

Only a small update on my installation. Yes, I still need to neaten the fans on top of the inverters.

Did any of you ever find a solution to this question? There's a RS232 port on Sunsynk batteries which might be useful but I have no clue on how to use it, ie. I can't find any documentation on it, eg. modbus details. Nor do I know whether that information should be extracted from the inverters instead?

The idea (at least for the 5K models) is to get the warm air away from inverter and let cooler air enter from the bottom so yes, mine moves air from the top of the inverters towards the ceiling.

I still have an elink 2.2 installation file. Edit: and I found the 2.3 version here: https://web.archive.org/web/20220327050325/https://efergy.com/wp-content/uploads/2019/08/eLink_V2.3_for_efergy_20140522.exe_.zip

More points to raise: so many nameless positive reviews on their web site, yet not a single one on Google or hellopeter or anywhere else? But according to web site they've been operating since sometime in 2023? Also, no company or VAT registration number listed anywhere? Too many red flags.

I've seen some weird results just reading the date & time from the inverters. It also doesn't match what's shown on the inverter screens. Anyway, your answers: a) Yes. fc=3; address=from-where-you-start-to-read; quantity=how-many-you-want-to-read b) Yes. fc=16; address=from-where-you-start-to-write; quantity=how-many-you-want-to-write c) I query about 30 single registers in a few seconds as part of my own "every 5 second" monitoring via node-red.

I tested my time setting code and even though the values I get via modbus from my 2 inverters differs by about 20 seconds, I went to check the screens and they're perfectly in sync. I think I need to debug a bit more when I'm awake in the morning. I'd hate to blow up my inverters now. 😊

second = 5165%256 = 45 minute = (5165-second)/256 = 5120/256 = 20

Maybe support thinks that only installers have the ability to create a new plant. They underestimate us South Africans. For the reasons you mention, I'd imagine a new plant should only get new data and not have access to any of the old data. Only time will tell if it works this way or whether you indeed will get your historic data back.

@Ivan_E see my sample node-red flow too. I've edited my previous post to add it.

x % y = the remainder when you divide x by y. In the example above, 6147 = 24*256 + 3. Just trust me, node-red knows what to do with that. 😉

month = 6147 % 256 = 3 year = (6147-month)/256 + 2000 = 6144/256 + 2000 = 24 + 2000 = 2024 Edit: I've hopefully attached my date & time fetching node-red flow. I have 2 inverters in parallel, hence the 2 nodes in my flow. Just nuke one of them if you only have 1 inverter. I'm using mbusd on a remote node hence the censored IP address, but it should be simple to edit the inverter configuration to whatever you're using, eg. rs485 usb dongles. flows.json

I seem to recall I bound and unbound my dongles multiple times in the past without the help of the installer. I even swapped dongles between inverters (my OCD in terms of equipment numbers). Back then I didn't care about old data since it was a new installation so I can't confirm whether I lost data then or not. But surely doing the dongle unbind on the plant and binding it again shouldn't lead to data loss? One would hope that the historical data is linked to the inverter and not the dongle. As I understand it the dongle is merely your "transfer agent" to move the data from the inverter to the Sunsynk cloud. So maybe your old data will appear again.

Yes, I wouldn't have deleted the plant but merely bound the dongle again. It's sad that support is so shoddy.

Based on the photos it appears the inverter and gateway are actually bound already? Go into your plant on the main screen and click on equipment at the bottom and show us the inverter and gateway screens (top buttons)? Edit: Oh wait, I see your main screen shows "3 days ago", so the communication is probably down there. Mine shows "a few seconds ago" in that spot.

I was forced to login at least twice this morning (using desktop browser) and I'm seeing current data. Maybe try logging out and back in?

In my opinion, disabling time sync in the setup is a much less destructive option. Surely you will require that dongle if the firmware ever needs to be updated?