Good afternoon, colleagues.
I see there’s new firmware for my SE17K on the official website - 4.24.9.
I’m keeping an eye on this thread because I also have a Solaredge paired with a Victron and I’m also experiencing connection drops between the devices.
Has anyone installed 4.24.9?
Or what is the latest official firmware for my device? After contacting support, they updated it to 0004.0022.0044.
Actually my Solaredge inverters are not connected over RS485 but over TCP.
See the attached screenshots
Yes that’s correct but when you configure the Solaredge you assign rs485 1 as modbus as per the documents in Victron
There is a known bug in the Solaredge inverters with the firmware after 4.22.44. So if a later firmware is installed this behaviour is possble if you use modbus over TCP/IP. Ask Solaredge to downgrade the firmware is the solution. Don’t know if a direct RS485 connection would solve this as well (in my situation not possible).
any news on venus os 3.71?
is this problem still present or was victron able to find a workaround?
Problem is still present for me on 3.71. Only solution is a downgrade of the solaredge firmware.
Same here 3.71 and 4.24.22 with trouble.
I requested the FW downgrade. Let’s see how long it takes to do so.
Hallo, i have same problem with SolarEdge firmware above 4.22.44. Solaredge support now is terrible. They said 4.22.44 is last working firmware, but have no info if they work on solution.
It’s Interesting I have 5 installations all with Solaredge inverters on AC OUT 1 and only one of them occasionally drops out and it’s the one where the solaredge meter is connected to RS485-1. All the rest have been connected to RS485-2. The solar edge Inverters are all reasonably new firmware, some the latest.
The only thing about these installations is that they are all on Raspberry PI 3bs I am not sure if that makes a difference either, but on the four with the meter connected to RS485-2 there are no drop out problems that either the client or I have seen.
I strugling with the same problem here, unfortunatelly I made a SE firmware upgrade to latest for both SE5K and SE8K - realizing afterwards that cause a problem before I found that thread. I using cable connection both SE and Cerbo. I tried different sets of configuration and I noticed that when set ID as 1 instead of 126, it stops dissapearing from Cerbo but presenting only power production without other data. Using newest stable Cerbo firmware and latest firmware on SE
It doesn’t work and it doesn’t matter what the Solaredge inverter address is. I have several installations where there is a problem with it and even Dynamic power limit is not functional. In certain situations, battery charging is also limited, then it is necessary to reset Cerbo GX. It might be best to use electricity meters and not use the connection via Sunspec. These compilations have been taking a long time and using Nodered is not a solution.
Connected my SE meter to RS-485-2 but the problem is the same, every time the SE is rebooting the Cerbo runs into a ~5 minute reconnect loop until I reboot the Cerbo, which I do automatically after 3 loops then it is running (including dyn. power limit)
Agree a meter is the best way to fix this as long as you do not need Dynamic PowerLimitfunctionality
Perhaps interesting for others here. Instead of fighting with firmware versions, I decided to just program a daily reboot using a cron job, considering a reboot fixes the connection issues for at least a day.
If you have SSH access, you can run the following command:
cat >> /data/rc.local << 'EOF'
echo "0 3 * * * /sbin/reboot" > /etc/cron.d/daily-reboot
EOF
chmod +x /data/rc.local
This will reboot at 03:00 every night. Which has been working great for me the past few days.
Interesting, but in my case it would work only in summer, when SE less then 10 hours in standby state, during winter SE reboots at ~6:00 in the morning.
BTW.: fixing this on Linux level is not a good solution for your customers
I’m not following. This will reboot at 03:00 regardless of what date it is
Understood, try to explain:
my Solaredge inverters rebooting ~100 minutes after sun power is zero, so in the evening, this (very dirty) SOLAREDGE reboot leads to the “communication dropping” issue, this will be fixed by your reboot. BUT during winter Solaredge reboots also if it gets no power for more then 10 hours, in my case in the morning (05:00 - 7:00) which means the “communication dropping” issue is active the whole day. In my setup I use a NodeRed flow to rebbot Cerbo in case the “communication dropping” issue is detected, and works perfect
is the flow available somewhere?
I had both inverters downgraded to 4.22.44 and everything works gracefully now
Here is the main part of my flow.
This node you need to update to your SE PV Inverter
This flow is counting the status null, after > 2 within one hour it will reboot VenusOS, after a 30 sec delay to give you the chance to edit flow in case of endless .loop
SolarEdgeConnectionLostReset.json.txt (11,4 KB)
Success and greetings to E’n.