question

Some addition info which may be relevant to you or others.

I also have the same issue with a VenusGX and a Lynx 1000 distribution (shunt) system. The 'value' of this -ve charge changes rapidly on our system depending upon actual load but always with very low loads and around 50w (on 27.1 V typically). My tests seem to show that the actual current across the shunt is the culprit when the 'Set Zero' actual occurs. In other words there is always a load in our system, these being the VE.direct BMS connected to the batteries via the serial bus and powered from this cable and also the Lynx 1000 itself and the VenusGX which is the controller used to actually trigger the 'set' the zero point. Have turned off all other 'loads' via the 'Load Disconnects'. All negatives go to the Lynx Out side and ONLY the LIPO batteries connections (+ and -) connected to the Lynx In. Have played with this over time and eventually unbolted the loads from the LYNX out to ensure no real loads. I generally now 'accept' this small (<50 w) fluctuation and use the battery voltage as my primary measuring of battery charge status using the Victron battery (LIPO 28v) (recommended) parameters. I understand that SOC is a 'calculation' and voltage is an actual reading hence my preference. SOC 'error' grows over time depending upon this 'zero' set point 'reality'. My wish is for a 'button' on the Lynx Shunt which set 'zero' value and could then get it closer to reality. Hope this may assist.