charged battery is floating

Hi,

I have one problem with my battery.
My hardware:

  • 1x MultiPlus 48/5000 (connected by ve.bus to venus os)
  • 3x MPPT 150/70 with PV 6 kwp (14 pcs.) (connected by wv.direct to venus os)
  • LFP4 battery with 50A and 5,12 kwh (connected by can to venus os)
  • venus os on raspbbery pi
  • Homiles micoinverter with 0,43 kwp (1 pcs.) only to benchmark and performance battle: victron vs world :wink:

My configuration:
ESS enabled:

  • Optimized (without BatteryLife)
  • Minimum SOC: 30%
  • AC-coupled PV - enabled
  • DC-coupled PV - enabled

DVCC enabled:

  • Limit managed battery charge voltage: 55.4V
  • SVS - enabled
  • STS - enabled
  • SCS - enabled
  • Controlling BMS by battery

MultiPlus settings:
Inverter card:

  • DC input low shut-down: 48V
  • DC input low restart: 51.2V
  • DC input low pre-alarm: 50.2V

Charger card:

  • Absorption voltage: 54.60V
  • Float voltage: 54.00V
  • Charge current: 25A

ESS:

  • Sustain voltage: 48.00V
  • Dynamic cut-off: 50.50V

The whole system works pretty well and stable, but…

I have a lot of power in my system. Power form PV is enough for all AC consumption and export to grid.

Everything looks good when battery is charging. Current and voltage are stable, SOC growth stable.

But when my battery is “full” - preferable by me SOC 98% my battery is all the time charged and discharged.
Battery all the time has ~150w discharge, ~150w charge. And all the time.

above but with aggregateWindow 1 min:

Is it ok? Or I should change some parameters?
Maybe Victron smart shunt can help me?

is your battery a Pytes?

No. My battery is powerlab, but it looks like similar product.
The battery has BMS compatible with pylonetch

Make sure the GX is on the latest version.
Your BMS spoofs Pylontech, it is not pylontech though and is ultimately unsupported. So if current GX firmware doesn’t help, you may need to speak to that manufacturer or use a conventional BMV instead.

Could be a similar problem with the one described here?
The same ups and downs when the battery is fully charged…

@nickdb all software is up to date.
Venus OS: v3.41 large.

You just confirmed my fears that the BMS is bad.

Do you think, that smart shunt is good enaught or I should migrate to different solution ex. battrium?

Yes, this is similar problem to mine.
But without solutions :frowning: In every threads example solutions is BMV, but without continuation and answers: yes, it’s works.

Probably I have to buy smart shunt and try that solutions.

Hi i’m pretty sure it is not the BMS, the BMS is only there for protection it should not regulated the charge/discharge cycles. The BMS should only intervene when the limits are exceeded of battery gets to hot/cold.

But I did some testing without BMS/With BMS/ With DVCC and Without DVCC see picture below.


(in the middle the BMS is connected again and the CVL rises it goes from float to absorption mode)
There are some variations time wise but the battery still is charging/discharging when full and more then enough PV power. When you put it in charger only with keep batteries charged it does not have that behaviour but the Multiplus goes to idle mode. I think it is just how the control is designed in the victron system only thing is that not everybody is seeing it.

If your BMS isn’t setting the CCL to 0A when the battery is at 100% then the DVCC controlled chargers are going to try and charge the battery. The BMS is telling the system “I will accept charge current”.

The BMS is regulating charging, it controls the whole system via DVCC. If it is not configured correctly the system will behave in undesirable ways.
If battery voltage is allowed to creep up beyond where it should be, the system will discharge it. Why it creeps up is usually all about how the battery is configured.
This is one of the joys of DIY batteries.

That is 100% true, but why would the battery discharge when there is more then enought PV available.

Because the rising voltage starts to throttle the chargers until they stop or until the bus voltage has lowered again.

Yesterday I made another test. I set:
Absorption voltage to 56,8V
Float to 54V
DVCC voltage limit to 56,8V (the same like absorption voltage)

Next I tuned off export form DC to the grid.
When export is disabled everything looks gr8.

Battery has SOC 100%, stable voltage ~56,8V, no charging and discharging. BMS set CCL to 0A.
Probably it works good, because Victron doesn’t try maximizing of production from PV and voltage on whole DC system is stable.

After a few hours I enable export to the grid.
After that export start, but charging and discharging too.

I have one more idea. When I turn on DVCC and export to the grid, System add 0.4V to the voltage limit. With this configuration on MPPT I have 57,2V instead of 56,8V.
Today I’ll try again, but after stabilize system on 56,8V and turn on export, I’ll setup voltage limit to 56,4V (56,8V on mppt like whole system)

The DVCC voltage limit is not intended for ongoing use, setting it the same as the battery will have zero effect.
Export raises the voltage by 0.4V, that is by design else exporting won’t happen, and it will ignore anything you set to prevent it happening. The inverter will drive this offset down under load, so once stable it shouldn’t be an issue for a battery and it is something that supported batteries are tested for.
On my own I can see an initial fluctuation once the inverter/mppt tug of war starts, but it settles soon enough.

Try set an export limit. You will find a happy median where the sawtooth hopefully settles.