Looks Like you may have solved my initial question I posted in a different thread, to the issue of not being able to scale back the charge current in DVCC. Do you use node red in HA ? or on the Cerbo it self ?
I have a similar issues as above, my system trips on high voltage or over current to the JK, when the sun comes out from behind clouds, as it jumps from a low solar value to around 15kw very quickly, too quick for the MPII (Multiplus 2) to adjust the frequency and back off the two fronius inverters. I tried adjusting DVCC from HA, i could do it in the victron device in the device, list, but when I ran a script it would just drop it to zero or do nothing at all.
The other issue is similar to above on nice clear sky days, with good solar production I start to get high charge current issues as we get to 90%. I have found if I go into DVCC and scale back the Max charge from 240 to 180 it seems to be enough. So your node red script might be the answer.
My cerbo is doing a lot and I didnt want to overload it with Node red as well, have you seen any issues with this ?
My Cerbo controls data links to 2 x MPPT, 2 x MPII, 2 x Vic EV chargers and 2 x AC coupled Fronius Inverters as well as the 6 JK-BMS, and one Shunt, as well as passes all this data off to Home assistant and VRM.
@Ty_web Did you ever get modbus correctly configured for the Fronius’? That’s interesting about the spiky nature of the charging creating issues. I think in my case the solar production was consistent. People seem to say this is an issue with the JK BMS’s but I would think this is more of an issue with the Victron approach to frequency shifting. The JK BMS to my understanding is just advertising its limits and desired charging parameters which doesn’t really change (?). Maximum charging amps has nothing to do (in my understanding) with holding the charging voltage limit. Instead PV Assistant should have some better algorithm for holding the charging voltage limit (this reminds me of how limited the absorption algorithm is on the Victron MPPT’s themselves, at least for LiFePO4 – there’s basically no logic!).
I bet you could run a higher charging current limit and remove your high current errors and soak up those moments of extra Sun. I know Andy is a fan of .2C charging but I seriously doubt (without much credibility) your battery lifespan will be damaged by occasionally charging .3 or even .4C. There are studies in EV batteries that show that periodic high discharge seems to actually be good for longevity, perhaps it’s the same with charging current (and .3 or .4C is still well below spec if you’ve got a decent pack). Considering the size of your system you could also add a 6th battery in parallel to match your growing PV production. That’s peanuts compared to total system cost, right?
For high voltage errors, I have reduced my absorption time (RCV Time) to the JK minimum (.1H) and this seems to be working better (but I’ve also had more load on at higher voltages lately). So the battery isn’t getting 100% full or balancing well but I can do this manually if necessary. By the time the AC Inverters would ramp back up from the frequency shifting cutitng them off, the voltage has already dropped to the RFV which is only like 3.35V/Cell. Again, missing the top 0.05% of charge or whatever but better that then system shutdown.
I know your comments were meant for @owenb79 but just wanted to respond as well as a fellow sufferer..
Hi James, All good, another chap responded in my post about Cerbo load and said it would be fine. SO Installed Node Red last night, and configured a scaling DVCC charge value. I am currently watching it and it seems to be working just fine.
So now when the Soc hits 80%, the DVCC max Charge (MC) is lowered from 240Amps to 200amps, at 85% Soc MC drops another 25amps, and repeats this lowering every 5% of soc. The system seems to be able to hold to within 10% of this under normal sun conditions…..but I have the main AC running which is pulling 4kw, so this may be masking the issue a little. As I type the sun is coming back out and the current is rising.
The Modbus issue is still there, not sure why. I upgraded my Cerbo to the Latest OSlarge last night to enable Node red, and checked again today and still no connection. They show up in the available Modbus entities list along with everything else, but not in the actual connected Modbus devices list. Frustrating.
I have noticed today what you are talking about with the MPPT’s not scaling very well at all. To reduce the charge current the Inverter is Freq shifting and reducing the output of the AC coupled units. The MPPT’s are not budging from there production. They are set to remote control, so I am not sure what else can be done. Maybe they need to go on the VE.Canbas, instead of the VE.Direct to have this function correctly.
As for the additional battery…yep its on the shopping list, just got to get financial approval from the minister of war. :) I also wanted to get this charging thing sorted as before adding the last 4kw of solar and any more batteries to the system.
If you want to have a play with adjusting DVCC to limit the current when charging is ending, check out this post and thread. The code is for a 12V system.
The OP in this thread mentioned that PV inverters don’t stop charging when DVCC is adjusted. I don’t quite understand why that would be the case. If the internal charger is disabled, then it should start shifting, but it doesn’t.
This is how DVCC adjustment works for me on a 12V system. I have 400W of solar and an upper DVCC limit of 50A, so the current shown in the graph is from the AC charger. The AC charger comes on at 10:30 am. As the voltage rises above 13.6V, the current is reduced, which allows cells that are running high to pull the overall voltage down. You can see that when the current drops, the voltage quickly follows as the higher cells are balanced down.
Without this control, one battery will rise to 14.2V and the SOC will show 100%, but the battery is not actually charged. There are still hours left of absorption and balancing, which is handled by the DC solar array. Since my array is east-facing, the current is very low in the afternoon, which makes it useful for balancing.
This is the perfect way to charge a battery for my use. By the time it’s fully charged, the sun is down, and the discharge cycle starts.
I also wanted to get this charging thing sorted as before adding the last 4kw of solar and any more batteries to the system.