You can add a manual IP address for the SolarEdge inverter in the Integrations section of the GUI. I have a Modbus proxy running on the Cerbo, so I configured the SolarEdge with IP address 127.0.0.1. Node-red is also using 127.0.0.1 to communicate with the SolarEdge.
Thanks for the hint, but this ist not possible with 2 SolarEdge inverters.
Did you try V3.70~60 , here you can specify additional ports for PV Inverters?
Hi all,
did more analysis with my Victron - SolarEdge system (3xMP 2-3000 + SE9k + Rpi4)
VenusOS: same results with V3.66, V3.70~35 & V3.70~60
Se9k: FW:004.0024.0016
Observed that SolarEdge Inverters are resetting 2 times a day
First in the evening (today 18:18) second 10 houra later expect next at ~4:20
During this reset data from SE is partly wrong at the beginning (example power values 0xffff)
After this reset the communication Victron - Solaredge drops every ~4.5 minutes.
Rebooting VenusOS once or twice solved the problem till the next reset of SE.
Also deleting and rescanning PV inverters IP can fix it (for a while)
The timing of the comunication drop loop is:
- Communication starts
- ~1’50" normal communication
- ~1’10" power values frozen
- ~ 1’20" communication lost (PV inverter disappears from Venus screen)
- communication o.k
- … repeats every 4’20"
@Victron is ist possible to detect the reset in VenusOS and react on this, and dive trough this daily resets.
Since more than a year I am complaining about the resets at SE support, because of problems with communication problems between SE-meter and SE9k, without any feedback.
So I hope the exzellent Victron developers can find a solution, possible increase a timeout.
Greetings from Austria, Leopold
Hi all,
Today I also opened a ticket with SE to downgrade my SE4000H to FW 4.22.44. But meanwhile I understand that using a modbus-proxy (eg on my Home Assistant Supervisor) might be a workaround?
Thanks for any confirmation on this.
Cheers, Michael
Opened a ticket as well a week ago with downgrade request. I received a mail after a few days that they upgraded to Firmware 4.24.15 And that should fix the problem….but it is still wacky.
I replied that I want a downgrade. let’s see.
I’m considering to just move all panels to my multiRSsolar invertors. The thing that is holding me off for now is that I need to remove the optimizers, maybe the only plus point for solaredge right now.
Hi,
Would not remove SE because functionality is ok, only monitoring makes problems. Used Modbus-proxy a year ago with similar problems (timeouts and polling timing was also critical) as this was also not stabil I removed it again. It also not a good idea to install an additional device on customer side.
Yesterday I designed a nodered flow that detects the loss of communication and resets the Cerbo. Will watch the system for a few days and communicate my result here.
1 Like
On November 18, 2025, I again contacted a representative via their chat, requesting a firmware downgrade to version 4.22.44.
This version has been mentioned several times on this forum as the solution for the communication interruptions between SolarEdge and the Victron.
After this, the system is running satisfactorily. yessssss!
Yes i can confirm this.
Works fine for me.
1 Like
Hello everyone,
We are still trying to figure out the issue to have a proper solution. Would it be possible to email me some examples to dimitar.shopov@solaredge.com?
It would be helpful to provide me with your serial number - if you are experiencing issues, we can downgrade to a version that should be stable - just include the request to downgrade in the email. I will try to be as responsive as possible and provide feedback when we have it.
Hi Victron-SolarEdge Experts and Users
here an update of my analysis:
since 21.Nov my NodeRed flow recovered ALL 8 communication drops caused by the hard resets of my SE9K (one time caused by a power-cycle of the SE9K).
My next step , I will downgrade my SE9K to 4.22.44 as proposed here and give this feedback to SolarEdge AND Victron.
My personal opinion the “communication drop” problem has 2 root causes:
-
@Dimitar.S SolarEdge is making a hard reset 2 times a day (my assumption and 40 years software experience they have some memory issues that is avoid by a preventive reset)
-
@Victron has a problem dealing with unexpected resets
Here a record of a single recover, my flow counts the appearance of a “null”-“valid-PV-Status” sequence, after the 4th trial of Venus OS I initiate a Venus Reboot. after this the communication works fine until the next SolarEdge Reset.
in my case there is no data loss because this SE resets always happen after sunset and (always 10 hours later) before sunrise.
2 Likes
Dimitar. I will send you some screendumps of my communicationproblem. With the software 4.22.44 it works fine from the date of downgrading (18 nov. 2025)
1 Like
victron os 3.62 ( all newer have problem i tested)
SE 6000h & SE 2200H latest firm ware 26-11-2025
This combination is stable , no resets and continueus communication
also , after victron os firmware change and newest SE firmware i reboott both SE inverters fully ( no power for 5 min )
and also reboot victron os
after that works perfect
Sounds good, what SolarEdge Firmware version is “latest firm ware 26-11-2025”??
whats the method you prove “no resets” of SE?
After my last post (sw version 4.24.15) I’m pretty sure that I did see sw version 4.24.16 ….But when I checked today it is now 4.22.44. So SE performed the fw downgrade. Since this morning I assume looking at the graph:
No more cerbo reboots to get things going 
I do hope they will find a solution to this problem soon.
1 Like
Silence since a week and a half now, and eager to know what the present status or best approach is on solving the communication breakdowns between SolarEdge and Victron equipment.
- Are Victron and SolarEdge working together on solving this, or expectng the other to fix it, or are on other priorities?
- Is Bas his solution the best; Roll back the Victron firmware to v3.62 and use the latest SolarEdge firmware 4.24.16 (which SE installed on my SE8K-inverter recently at another issue)?
- Is Maurice his solution the best; use the latest Victron firmware v3.67 and roll back the SolarEdge firmware to 4.22.44?
- Is Dick Streefland’s solution the best; have a Modbus proxy running on the Cerbo in Node-RED with IP address 127.0.0.1 and add the SolarEdge inverter in the Integrations section of the Cerbo GUI with that IP address?
I rather live with the pain for some while (although the frequently appearing/disappearing solar box on the GUI is annoying, and my Dynamic-ESS is predicting wrong, because of inconsistant solardata), and fix it correctly when the repaired/aligned firmware arrives. But is that foreseeable?
Nice Santa Claus or Father Christmas present? 
Where are we now?
Hello,
observed my system for a week now, here the results:
- on 3.12 SolarEdge downgraded my SE9K to 4.22.44
- since that time my Victron V3.70~61 did not loose comunication
- SolarEdge 4.22.44 performs a reset 2 times a day during 0PV time, one in the evening one ~10hours later in the morning
- during this reset SE Sunspec values are rubish (reports meter type 201 instead of 203, 0xffff for all power and SF registers)
- Venus OS can deal with this
I do not want to use downgrades firmware versions (SE & Victron) because, I will loose other improvements
would also prefer a solution of both partners, but i am afraid that there is no interest from SolarEdge side to fix issues that could help commercial competitors
so I sugest the best way would be that Victron adapts there init procedure as it works after my Cerbo Reset
Hello all
i have make a script for solar edge via node red
script is complete you need only set the solaredge rs485 tot solar edge leader ant set id to 2 ant if you have ohter solaredge set id to 3 ant modbus to 502
version 0.01 test works for me next version pv limit addind whits ess
let me know if there are bugs
see script
you can import the scrip in node red
Hier is een nieuwe complete import-flow (1 tab) die het zeker werkend maakt, zónder meter-blok gedoe:
2× SolarEdge (INV1 via 18fdf7679413493c unitid 2, INV2 via 0496ace946808635 unitid 3)
Leest alleen 40070 / 27 regs (SunSpec inverter model)
Berekent per fase Watt via V(AN/BN/CN) × I(A/B/C)
Totaal = som van 3 fases, en daarna INV1 + INV2 samen
Lifetime kWh = INV1 + INV2 (exact uit WH hi/lo + WH_SF)
StatusCode: 2 = running als totaal > 50W, anders 1 = sleep
Poll is netjes (1s) en leest om en om → minder “Close timed out”
Import: Node-RED → Menu → Import → Clipboard → plak → Import → Deploy
you need to set the ip from you omformers in INV1 ant 2 if you have 2 solaredge
if you have 1 need only INV1
Solaredge 2 omvormers uitlezen Victron Node Red.txt (9,5 KB)
1 Like