Eastron SDM230 and SDM120 accuracy

Installation Home Assistant OS
1

Hi

I have 2 modes devices (SDM230 & SDM120) connected to my Raspberry Pi via USB.

The SDM230 measures Import/Export on the grid while the SDM120 measures the PV array output.

There is a small difference between the values reported by the Eastrons and the 2 utility meters.

The difference is small but it accumulates over time.

Is there any way that I can eliminate this?

I have taken readings over a period of time but cannot nail it.

The following sensors are defined for the ‘Import’ side of things (similar sensors for Export and PV):

- platform: modbus
  scan_interval: 30
  registers:
  - name: 'Grid Import Energy Raw'                                                    
    unit_of_measurement: kWh
    slave: 1                                                                
    register: 72
    register_type: input                                                        
    count: 2    
    data_type: float 
    precision: 2 
    offset: 36481.26

  - name: 'Total Hours'
    unit_of_measurement: h
    slave: 1
    register: 63792
    register_type: holding
    count: 2
    data_type: float
    precision: 2

- platform: template
  sensors:
    grid_import_energy:
      value_template: >-
       {{(states.sensor.total_hours.state | float / 24 * 0.0187 + 
       states.sensor.grid_import_energy_raw.state | float) |round(2)}}

The 0.187 in the grid_import_energy sensor is derived from the slope of a graph of daily readings over time. It is still being tweaked as it seems to keep changing.

I suspect that the calculations need to be made on how long the import sensor has been importing each day rather than the total hours that it has been running. Or am I overthinking this?

This is nerve wrecking.

Thank you for any insight anyone may provide.

2

This is old, but…
There are all kind of losses, that may cause this, in addition to accuracy of meter itself. If your inverter isn’t right next to main distribution panel and these Eastrons aren’t next to each other, even it may cause difference. Then there’s question about power factor and plethora of other things. If I read equation correctly, difference is less than two percentage points.

Both Eastrons are supposed to be accurate enough (MID-certification) for utility grade use. According to directives and standards, accuracy classes are “Class A is equivalent to 2% accuracy, Class B to 1 % and Class C to 0.5 %”

My opinion is, if you have more than one device metering single measurable item, you will eventually have different readings, because it’s metering with errors, and errors and inaccuracy is kind of nature’s law.