2 x Pylontec US 5000
16 x LONGI HI-MO5 MONO 415W in two chains
2 x Victron Smartsolar MPPT 250/100
Victron MultiPlus II 48/5000/70-50 230V
Cerbo GX
GX Touch 50
The system constantly under reports the SOC.
Why do I think this is the case?
Every morning SOC quickly comes up to about 95% (500-1000W charge power). Then even thought the panel voltage (>140 V) indicates that plenty of power is available the charge power drops to 10 W range (at this point battery voltage is around 52.4 V) and it will take hours for SOC to creep up even one two percentage, then in the matter fifteen minutes or so SOC goes from 97% to 100%.
So I’m thinking that the charging control system detects that the battery is almost full and goes into some “trickle” charge mode to top it up and at some point it detects that the battery is actually 100% full and re-adjusts the SOC.
Now, this would not be a problem but I have set up a rule starts water boiler via the aux relay when SOC is 98% or more. Many days it happens that the boiler will not heat water at all though plenty of solar power would be available all day long.
I can of course make the water heater trip point lower, say 95%, and problem mostly solved.
But in a way this is working around the erroneous SOC problem.
Is there something I could do to fix the SOC?
AFAIU the SOC is calculated based on in coming and outgoing current from the battery and errors accumulate, and only when the battery reaches 100% the gauge is reset.
Which component is responsible for the SOC calculation?
Below is a screen cap of a typical situation in which I think SOC is markedly underestimated.
l
I suppose that the Cerbo is connected with the batteries with the CAN cable, right?
And the battery monitor is declared in DVCC to be the battery BMS, right?
The SOC is reported by the batteries and this is the way the Pylontech behaves.
Take a look at one of my charging cycles and see that if you draw an “evolution line”, will be quite spot on on the charging final stage.
So if they say it’s not 100%, it’s not 100%.
My system uses CAN bus, not sure about “And the battery monitor is declared in DVCC to be the battery BMS”, I am not familiar with those details.
Yes, I accept that if it is not 100% it is not 100% but the real problem is that the last 5% or so are so badly off. Looks like that is true for your system too.
I guess I will end up adjusting the boiler start/stop points. At the moment this is very annoying, the system reached 98% just about when the sun set. All the sun light during the day was wasted and the water not heated.
I believe you’re asking yourself the wrong question - it’s not a matter of “are my batteries full” but rather, “are my batteries full enough…”
In other words, once the PV input charge power drops, that’s a clear indication that the current required to top off the batteries is less than what’s available from the array, therefore a surplus is available for another task. That’s when the aux relay should be starting the water boiler.
Your hunch about the charging control system detecting the batteries as nearly full and therefore limiting charge current is absolutely correct, and that certain conditions trigger a readjustment of displayed SOC. However, stop waiting for that magical moment to occur - diversion loads work best the moment there’s surplus available to divert
You are right that I’m not interested if the batteries are full but when there is surplus.
But how do I determine if surplus is available?
My current method is to check if the SOC is 98% or over?
I can lower that to 95% and think it will then work but if there is some other way to check if surplus is available I’d be interested to investigate that option.
Thanks. Victron advices against this. I’m not keen changing the wiring done by professionals and getting the parameters right. I think my next port of call is just lower the boiler start SOC value to round 92%.
"I recommend to leave the Lynx shunt out.
it can indeed clash on the canbus when share with the Pylon, and there are more issues.
Lastly, you dont need it. Adding a shunt next to a managed battery like Pylon, even when on separate canbusses, is something we advise against."
Hmm. Since I last commented the system does not seem to get the SOC over 94%. The days are already rather long here in Finland so there is plenty of sola power. Around 10 AM the batteries start to charge with +1000W and rise from 90% to 92% then the system enter the float (or whatever it is) state and the charge creeps up but the days is not long enough to reach more than 94-96% though the sun sets after 9 PM.
Is this a problem that the system never reaches (nominally) 100%, though I’m pretty sure it actually is much closer than what the SOC says?
Is the SOC problem going to get gradually worse if it never reaches 100%?
If this is a problem I guess this is a problem that would NOT be solved with a Smarshunt cause the BMS is internal to the Pylontech?
