Overvoltage when keep batteries charged is on

Hi,

I noticed since a few days that the charge logic is not being followed sometimes when the systems is in "Keep batteries charged" mode. See below charts

DVCC is on and the bms is sending the correct charge voltages.

The pack itself is perfectly balanced within a few millivolts.

It happens when there is a peak in solar production, altough everything is connected on the AC side. (No DC MPPT's and DC system is turned off). I would expect that the grid feed-in would just increase instead of overcharging the batteries.

Is there any explanation or setting that I missed?

Thanx,

Bjorn

2 |3000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

blieve commented · May 30 at 11:28 AM

@Guy Stewart (Victron Community Manager) someone from Victron able to shine a light on this?

0 ·

Guy Stewart (Victron Community Manager) ♦♦ blieve commented · May 31 at 09:56 AM

Hi @blieve,

As a work around, could you please set the ESS mode to optimised and then set the minimum SOC to 100% (or 99%) and then report back with the behaviour?

Another thing do you have allow AC PV export enabled?

And yet another thing, on the Venus GUI, could you please set the reporting interval to 1 minute - from the current default 15 minutes. That would help in this case to see better if the Victron is following the batteries orders.

Ideally the Charge Voltage Limit is reduced before the battery reaches 100%, at 98% or less.

You don't say which battery or BMS you are using - do they have Victron support and compatibility documentation?

0 ·

blieve Guy Stewart (Victron Community Manager) ♦♦ commented · May 31 at 10:50 AM

Hi Guy,

I updated the reporting interval to 1 minute.

As a workaround I indeed changed to ESS mode to optimized with minimum SOC to 95%, the problem doesn't seem to happen in this case. (altough not very sunny here in the netherlands the last 2 weeks...)

I've seen another post here with a similar issue that said this as a workaround. (https://community.victronenergy.com/questions/132633/dvcc-max-charge-voltage-not-considered-when-ess-in.html)

AC PV export is enabled (without limits)

The batterypack itself has an JKBMS built-in and supports the Victron protocol, connected via the BMS.Can port. It's currently programmed to request a CVL of 55.2V for 1 hrs and then lower to the float voltage for 6hrs.

Kind regards,
Bjorn

0 ·

Alexandra ♦ blieve commented · May 31 at 02:09 PM

@blieve

Weird how the battery voltage rises like that with no current going to the battery? Something else is afoot. Possibly something like a disconnect is happening.

0 ·

blieve Alexandra ♦ commented · May 31 at 02:19 PM

It's hard to see on the chart but the charge current drops to about 1.4A so voltage keeps rising very slowly

0 ·

Alexandra ♦ blieve commented · Jun 02 at 04:44 PM

Cap it in DVCC. Hou can set the max charge voltage there.

I have also found that rather than use keep batteries charged what works for some batteries better is optimised with battery life and set it to 100% soc there.

For 1A charge the rise is a bit dramatic. Maybe check bus bar and battery assembly issues as well.

1 ·

blieve Alexandra ♦ commented · Jun 03 at 09:40 AM

I'll test it out with the DVCC cap see how it responds to that.

Battery and busbar are decently built, I won't expect anything wrong there, I don't see issues when in normal battery cycles or when the soc is manually put on 100%. Only weirdness with the voltage happens when in keep batteries charged

0 ·

blieve Alexandra ♦ commented · Jun 03 at 01:53 PM

Tried again today but same result (with DVCC charge voltage limited to 55,3V)

Battery voltage slowly drifting up, charge currents upto 5A.

0 ·

1717422752291.png (55.2 KiB)

Alexandra ♦ blieve commented · Jun 03 at 02:28 PM

Are all your components programmed with the same and correct charge voltages and chemistry?

How does the 55.3v differ from the absorption voltage?

0 ·

blieve Alexandra ♦ commented · Jun 04 at 08:40 AM

The MPII's had a different charge curve programmed then the BMS but I now put everything on the same voltages yesterday evening.

Including the DVCC max voltage.

Result is already the same on this sunny morning, the battery voltage keep rising (charge current between 1 & 5A).

Really looks like a bug to me! whatever happens, the max voltage should be respected in whatever mode.

0 ·

blieve blieve commented · Jun 04 at 03:00 PM

Todays result:

Still seems to follow the solar curve. I don't know if it would have kept rising when the clouds wouldn't have come or that it is somehow an unknown algorithm (kept rising and high for 5hrs, peak was around 56,6V)

So still 3V more than the programmed limits.

0 ·

1717512904757.png (102.9 KiB)