Cerbo would not charge batteries - connected with firmware 3.70 and higher

Hi @iburger , victron staff,

There is problem with firmware - 3.70. My system get updated directly to 3.80 and as fallback allows me to downgrade to 3.70. Both versions have issues for me - discharging my batteries to grid instead of charging them. They are slowly depleating. How can stop it and charge them ?

I am at 13% and going down slowly. Should I turn batteries off - hardware button - I don’t want them to go down to 0% without option to charge them.

I also checked for firmware but on your site there is only 3.70 available so this will not help me. Can I get somewhere 3.67v for cerbo ?

Please help.

ok,
I managed to start charging my batteries.

Can you tell me if it is save to stay on 3.80 version or do I need to downgrade to 3.67 ?

thanks

JFYI - I found out that in DVCC - parameter “Limit managed battery charge voltage” - was turned on and on 40V.

I turned it off and it allowed me to charge battery.

I am 100% sure I didn’t turn it on - maybe the firmware update/upgrade could turn it on ? Not sure - but this helped me to solve this issue.

Why are you on v3.80? That version is in beta testing. Do you have auto-updates as well as beta test branch enabled?

Hi chrigu,

I had some problems with DESS some time back and victron asked me to test it on new beta version and stay there till the fix get to main version. I think it was around 6 month ago.

I don’t have auto update turned on, but what was my surprise that eve I selected to update to 3.70 it automatically jumped to 3.80 and kept 3.70 as fallback downgrade.

Anyway I am fine right now - batteries back to 25% .

I usually am not on betas,

There is nothing in Venus that would enable this automatically. It defaults to 0 (disabled), and the only way it changes is if a user activates it. The first time you active the switch, it also starts at 55V (the system will use the lower of this, or what the BMS asks for), so it is very odd that it would go to exactly 40V.

Event logs doesn’t indicate that remote console was accessed around that time, so it doesn’t look as if it was done from VRM.

It looks like this setting changed yesterday afternoon at 16:13, your local time. Maybe that helps to narrow it down to a cause.

hi @iburger
thanks for reply.
Frankly after enough thinking - I am suspecting HomeAssistant.
I am closely monitoring this and checking my system right now.

You will not believe how much support load Home Assistant is creating for me :slight_smile: Thanks for confirming we have another cook in the kitchen.

1 Like

Hi Izak!

I am working with multiple systems that are now running 3.70 (with ESS) and we have had issues with the systems not charging batteries when they should - not from solar, not from just setting a large value for ‘grid setpoint’ (or a large ‘MinSoC’) to make them charge from the grid. Nothing worked. Tried going back to 3.67 and also still not happy. Could not find any setting acting as a barrier to charging but charging was not happening.

Eventually I tried resetting the underlying Quattro system (three phase system with 3 x Q48/15k) and then battery charging started to work again - charging then operated as expected, where it previously did not.

I just tried that on a second system I’m working with that, again, had upgraded to 3.70 and stopped charging. Again, resetting the VEBus systems underneath the GX fixed it, with no other changes made to anything.

Both systems are running firmware 558, they are in Australia on AS4777 grid codes (not the Annex M codes, the ‘conventional/older’ AS4777:2015 grid code).

As I see some changes were made related to AS4777 in firmware version 558, I am starting to wonder whether the issue (for the systems I’m working with at least) is actually a bug in firmware version 558, not in the GX/Venus code (tho perhaps triggered somehow by the GX/Venus code update).

Cheers,

Simon

p.s. for full disclosure, these systems are being tickled by some software I’ve written using MQTT to drive energy flows on the site - its not Home Assistant, but it is indeed adjusthing settings over time… I don’t think that’s ‘the issue’ as our code just adjusts the grid setpoint and other ESS settings over time in a very conventional way. Also, to prove the point - we tried just turning our softare off entirely - and the battery still didn’t charge - until we reset/rebooted the VEBus system underneath and then… bingo - all working fine again

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.