Skip to content

CSV format

CSV files are generated when the SD service is enabled. They are named YYYYMMDD.CSV and contain report rows collected over the day.

Document Context

  • Purpose: Understand the structure and format of CSV files logged to SD cards by whatwatt Go devices
  • When to use: When you need to process historical energy data, perform offline analysis, or understand the data export format
  • Prerequisites: Basic understanding of energy monitoring concepts; familiarity with OBIS codes helpful but not required
  • Related to: REST API data endpoints, DLMS/COSEM protocol, energy measurement standards
  • Validates against: Real device data from Landis+Gyr LGZ1030784855204 via MBUS interface

Key Facts

  • File naming: YYYYMMDD.CSV format (e.g., 20251007.CSV for October 7, 2025)
  • Timezone: All timestamps in UTC (indicated by "Z" suffix)
  • Frequency: Records written every 15 minutes by default (configurable via settings)
  • Protocol basis: DLMS/COSEM standard with OBIS code mapping
  • Access method: HTTP GET via /sdcard/ endpoint (no authentication typically required)
  • Data encoding: CSV with quoted string fields, comma-separated
  • Field availability: Many fields may be empty depending on meter capabilities

Sample header and rows

RID,TIME,MID,MSTAT,TARIFF,PF,EAP_T,EAP_T1,EAP_T2,EAN_T,EAN_T1,EAN_T2,ERP_T,ERP_T1,ERP_T2,ERN_T,ERN_T1,ERN_T2,ERII_T,ERII_T1,ERII_T2,ERIC_T,ERIC_T1,ERIC_T2,EREI_T,EREI_T1,EREI_T2,EREC_T,EREC_T1,EREC_T2,MDAP_T,MDAP_T1,MDAP_T2,MDAN_T,MDAN_T1,MDAN_T2,IPAP_T,IPAP_L1,IPAP_L2,IPAP_L3,IPAN_T,IPAN_L1,IPAN_L2,IPAN_L3,IPRP_T,IPRP_L1,IPRP_L2,IPRP_L3,IPRN_T,IPRN_L1,IPRN_L2,IPRN_L3,IPA_T,V_L1,V_L2,V_L3,I_L1,I_L2,I_L3,MIDENT,MMODEL,MIF,MPROT,RP
3093,"2025-10-07T20:21:11Z","","OK",2,,75.133,,,25.414,,,82.328,,,19.481,,,,,,,,,,,,,,,,,,,,,0,0,0,0,0,0,0,0,,0,0,0,,0,0,0,,234,0,0,0,0,0,"LGZ1030784855204","","MBUS","DLMS",6818
3096,"2025-10-07T20:21:31Z","","OK",2,,75.133,,,25.414,,,82.328,,,19.481,,,,,,,,,,,,,,,,,,,,,0,0,0,0,0,0,0,0,,0,0,0,,0,0,0,,235,0,0,0,0,0,"LGZ1030784855204","","MBUS","DLMS",6786
  • TIME is ISO‑8601 in UTC timezone (indicated by "Z" suffix)
  • Energy counters use units consistent with REST /api/v1/report
  • Many fields may be empty if not provided by the meter (depends on meter type and configuration)
  • Columns may evolve with firmware; treat unknown columns as optional

Field reference

The table below lists the CSV columns, their OBIS logical names (C.D.E), descriptions, units and value types.

Field OBIS (C.D.E) Description Unit Type
RID The report number that the device processed. This counter increments from the time the device boots. integer
TIME Date and time sent by the meter in the report, in local timezone string (ISO-8601)
MID 96.1.0 Meter identifier string
MSTAT Meter status string enum (NO INIT, NO DATA, RECOGNITION, OK, ENCRYPTION KEY, KEY REQUIRED, NOT RECOGNIZED)
TARIFF Current tariff integer (1..2)
PF 13.7.0 Instantaneous power factor float
EAP_T 1.8.0 Positive active energy (A+) total kWh float
EAP_T1 1.8.1 Positive active energy (A+) in tariff T1 kWh float
EAP_T2 1.8.2 Positive active energy (A+) in tariff T2 kWh float
EAN_T 2.8.0 Negative active energy (A−) total kWh float
EAN_T1 2.8.1 Negative active energy (A−) in tariff T1 kWh float
EAN_T2 2.8.2 Negative active energy (A−) in tariff T2 kWh float
ERP_T 3.8.0 Positive reactive energy (Q+) total kvarh float
ERP_T1 3.8.1 Positive reactive energy (Q+) in tariff T1 kvarh float
ERP_T2 3.8.2 Positive reactive energy (Q+) in tariff T2 kvarh float
ERN_T 4.8.0 Negative reactive energy (Q−) total kvarh float
ERN_T1 4.8.1 Negative reactive energy (Q−) in tariff T1 kvarh float
ERN_T2 4.8.2 Negative reactive energy (Q−) in tariff T2 kvarh float
ERII_T 5.8.0 Imported inductive reactive energy in 1st quadrant (Q1) total kvarh float
ERII_T1 5.8.1 Imported inductive reactive energy in 1st quadrant (Q1) in tariff T1 kvarh float
ERII_T2 5.8.2 Imported inductive reactive energy in 1st quadrant (Q1) in tariff T2 kvarh float
ERIC_T 6.8.0 Imported capacitive reactive energy in 2nd quadrant (Q2) total kvarh float
ERIC_T1 6.8.1 Imported capacitive reactive energy in 2nd quadrant (Q2) in tariff T1 kvarh float
ERIC_T2 6.8.2 Imported capacitive reactive energy in 2nd quadrant (Q2) in tariff T2 kvarh float
EREI_T 7.8.0 Exported inductive reactive energy in 3rd quadrant (Q3) total kvarh float
EREI_T1 7.8.1 Exported inductive reactive energy in 3rd quadrant (Q3) in tariff T1 kvarh float
EREI_T2 7.8.2 Exported inductive reactive energy in 3rd quadrant (Q3) in tariff T2 kvarh float
EREC_T 8.8.0 Exported capacitive reactive energy in 4th quadrant (Q4) total kvarh float
EREC_T1 8.8.1 Exported capacitive reactive energy in 4th quadrant (Q4) in tariff T1 kvarh float
EREC_T2 8.8.2 Exported capacitive reactive energy in 4th quadrant (Q4) in tariff T2 kvarh float
MDAP_T 1.6.0 Positive active maximum demand (A+) total kWh float
MDAP_T1 1.6.1 Positive active maximum demand (A+) in tariff T1 kWh float
MDAP_T2 1.6.2 Positive active maximum demand (A+) in tariff T2 kWh float
MDAN_T 2.6.0 Negative active maximum demand (A−) total kWh float
MDAN_T1 2.6.1 Negative active maximum demand (A−) in tariff T1 kWh float
MDAN_T2 2.6.2 Negative active maximum demand (A−) in tariff T2 kWh float
IPAP_T 1.7.0 Positive active instantaneous power (A+) kW float
IPAP_L1 21.7.0 Positive active instantaneous power (A+) in phase L1 kW float
IPAP_L2 41.7.0 Positive active instantaneous power (A+) in phase L2 kW float
IPAP_L3 61.7.0 Positive active instantaneous power (A+) in phase L3 kW float
IPAN_T 2.7.0 Negative active instantaneous power (A−) kW float
IPAN_L1 22.7.0 Negative active instantaneous power (A−) in phase L1 kW float
IPAN_L2 42.7.0 Negative active instantaneous power (A−) in phase L2 kW float
IPAN_L3 62.7.0 Negative active instantaneous power (A−) in phase L3 kW float
IPRP_T 3.7.0 Positive reactive instantaneous power (Q+) kvar float
IPRP_L1 23.7.0 Positive reactive instantaneous power (Q+) in phase L1 kvar float
IPRP_L2 43.7.0 Positive reactive instantaneous power (Q+) in phase L2 kvar float
IPRP_L3 63.7.0 Positive reactive instantaneous power (Q+) in phase L3 kvar float
IPRN_T 4.7.0 Negative reactive instantaneous power (Q−) kvar float
IPRN_L1 24.7.0 Negative reactive instantaneous power (Q−) in phase L1 kvar float
IPRN_L2 44.7.0 Negative reactive instantaneous power (Q−) in phase L2 kvar float
IPRN_L3 64.7.0 Negative reactive instantaneous power (Q−) in phase L3 kvar float
IPA_T 9.7.0 Apparent instantaneous power (S+) kVA float
V_L1 32.7.0 Instantaneous voltage (U) in phase L1 V float
V_L2 52.7.0 Instantaneous voltage (U) in phase L2 V float
V_L3 72.7.0 Instantaneous voltage (U) in phase L3 V float
I_L1 31.7.0 Instantaneous current (I) in phase L1 A float
I_L2 51.7.0 Instantaneous current (I) in phase L2 A float
I_L3 71.7.0 Instantaneous current (I) in phase L3 A float
MIDENT 42.0.0 Meter identifier string
MMODEL 96.1.1 Meter model string
MIF Meter interface string enum (P1, MBUS, TTL, MEP)
MPROT Meter protocol string enum (DSMR, DLMS, KMP, MEP)
RP Report period. How often does the meter send a report float

Notes:

  • Where both MID (96.1.0) and MIDENT (42.0.0) appear, they may reflect different meter identifier objects depending on protocol/firmware.
  • OBIS notations follow C.D.E fields, common in DLMS/COSEM.
  • Enumerations are shown in the Type column where applicable.
  • Many fields may be empty or contain only null/zero values depending on what the specific meter model provides.
  • The TIME field uses UTC timezone (Z suffix) rather than local time in current firmware versions.
  • RP (Report Period) shows the interval in milliseconds between meter reports.

Common Issues & Solutions

Issue: Empty CSV file or missing daily files

  • Symptoms: Expected YYYYMMDD.CSV file doesn't exist or is empty
  • Root cause: SD card service disabled, storage full, or file system corruption
  • Diagnosis:
  • Check device settings: services.sd.enable should be true
  • Verify SD card is inserted and detected
  • Check available space on SD card
  • Solution:
  • Enable SD logging via /api/v1/settings: {"services": {"sd": {"enable": true}}}
  • Reformat SD card if corruption suspected
  • Adjust frequency setting to reduce storage usage
  • Code pattern:
# Check if SD logging is enabled
settings = requests.get(f"http://{device}/api/v1/settings").json()
if not settings.get("services", {}).get("sd", {}).get("enable", False):
    print("SD card logging is disabled")

Issue: Many CSV fields contain empty values or zeros

  • Symptoms: Most columns show empty strings or zero values
  • Root cause: Meter doesn't provide all OBIS objects, or communication issues
  • Diagnosis: Check MSTAT column - should be "OK" for valid data
  • Solution:
  • Filter rows where MSTAT != "OK"
  • Focus on populated fields for your specific meter model
  • Different meter types provide different data sets
  • Code pattern:
import pandas as pd

df = pd.read_csv('20251007.CSV')
# Filter valid meter status
valid_data = df[df['MSTAT'] == 'OK']

# Find non-empty columns for this meter
populated_cols = []
for col in df.columns:
    if df[col].notna().sum() > 0 and (df[col] != 0).sum() > 0:
        populated_cols.append(col)
  • Related: Meter model specifications, OBIS code availability

Issue: Timestamp parsing errors

  • Symptoms: TIME field format not recognized by date parsers
  • Root cause: ISO-8601 UTC format may not be handled by all parsers
  • Diagnosis: Check TIME field format: should be "2025-10-07T20:21:11Z"
  • Solution: Use ISO-8601 compatible parsers, handle Z suffix correctly
  • Code pattern:
from datetime import datetime
import pandas as pd

# Proper timestamp parsing
df['timestamp'] = pd.to_datetime(df['TIME'], format='%Y-%m-%dT%H:%M:%SZ', utc=True)

# Convert to local timezone if needed
df['timestamp_local'] = df['timestamp'].dt.tz_convert('Europe/Zurich')

Issue: Inconsistent data frequency

  • Symptoms: CSV rows don't appear at expected intervals
  • Root cause: Meter report frequency varies, or device missed some reports
  • Diagnosis: Check RP (Report Period) field for actual meter intervals
  • Solution: Handle variable intervals in data processing
  • Code pattern:
# Calculate actual intervals between readings
df['time_diff'] = df['timestamp'].diff().dt.total_seconds()
median_interval = df['time_diff'].median()
print(f"Median report interval: {median_interval} seconds")

# Detect gaps in data
gaps = df[df['time_diff'] > median_interval * 2]
print(f"Found {len(gaps)} data gaps")

Issue: Energy counter resets or discontinuities

  • Symptoms: Cumulative energy values (EAP_T, EAN_T) decrease unexpectedly
  • Root cause: Meter reset, device restart, or data corruption
  • Diagnosis: Look for sudden drops in cumulative counters
  • Solution: Implement continuity checks and handle resets gracefully
  • Code pattern:
# Detect energy counter resets
def detect_resets(series):
    resets = []
    for i in range(1, len(series)):
        if pd.notna(series.iloc[i]) and pd.notna(series.iloc[i-1]):
            if series.iloc[i] < series.iloc[i-1] * 0.9:  # 10% tolerance
                resets.append(i)
    return resets

eap_resets = detect_resets(df['EAP_T'])
if eap_resets:
    print(f"Detected energy counter resets at rows: {eap_resets}")

Issue: Phase data inconsistency on single-phase meters

  • Symptoms: L2 and L3 columns always zero, only L1 has values
  • Root cause: Single-phase meter installation - L2/L3 don't exist
  • Diagnosis: Check if voltage and current L2/L3 fields are consistently zero
  • Solution: Use total values or L1 values for single-phase analysis
  • Code pattern:
# Detect meter phase configuration
def detect_phases(df):
    phases = []
    if (df['V_L1'] > 0).any():
        phases.append('L1')
    if (df['V_L2'] > 0).any():
        phases.append('L2')
    if (df['V_L3'] > 0).any():
        phases.append('L3')
    return phases

active_phases = detect_phases(df)
print(f"Active phases: {active_phases}")

Semantic Field Map

Understanding the logical grouping and meaning of CSV fields for RAG and data processing:

Energy Counters Group (Cumulative Values)

  • EAP_T (OBIS: 1.8.0) → semantic_name: "total_imported_energy" → meaning: "Cumulative energy consumed from grid" → unit: "kWh" → type: "cumulative_counter"
  • EAN_T (OBIS: 2.8.0) → semantic_name: "total_exported_energy" → meaning: "Cumulative energy fed back to grid" → unit: "kWh" → type: "cumulative_counter"
  • ERP_T (OBIS: 3.8.0) → semantic_name: "total_reactive_positive" → meaning: "Cumulative positive reactive energy" → unit: "kvarh" → type: "cumulative_counter"
  • ERN_T (OBIS: 4.8.0) → semantic_name: "total_reactive_negative" → meaning: "Cumulative negative reactive energy" → unit: "kvarh" → type: "cumulative_counter"

Power Measurements Group (Instantaneous Values)

  • IPAN_T (OBIS: 2.7.0) → semantic_name: "instantaneous_export_power" → meaning: "Current power fed back to grid" → unit: "kW" → type: "instantaneous_value"
  • IPRP_T (OBIS: 3.7.0) → semantic_name: "instantaneous_reactive_positive" → meaning: "Current positive reactive power" → unit: "kvar" → type: "instantaneous_value"
  • IPA_T (OBIS: 9.7.0) → semantic_name: "instantaneous_apparent_power" → meaning: "Current apparent power (total)" → unit: "kVA" → type: "instantaneous_value"

Electrical Parameters Group (Per-Phase Values)

  • V_L1, V_L2, V_L3 (OBIS: 32.7.0, 52.7.0, 72.7.0) → semantic_name: "voltage_phases" → meaning: "Line voltages for three phases" → unit: "V" → type: "instantaneous_value"
  • I_L1, I_L2, I_L3 (OBIS: 31.7.0, 51.7.0, 71.7.0) → semantic_name: "current_phases" → meaning: "Line currents for three phases" → unit: "A" → type: "instantaneous_value"

Device Metadata Group

  • MIDENT (OBIS: 42.0.0) → semantic_name: "meter_identifier" → meaning: "Unique meter serial number" → unit: "string" → type: "device_metadata"
  • MSTAT → semantic_name: "meter_status" → meaning: "Communication status with meter" → unit: "enum" → type: "status_indicator"
  • MPROT → semantic_name: "meter_protocol" → meaning: "Communication protocol used" → unit: "enum" → type: "device_metadata"
  • MIF → semantic_name: "meter_interface" → meaning: "Physical interface type" → unit: "enum" → type: "device_metadata"

Temporal Data Group

  • TIME → semantic_name: "measurement_timestamp" → meaning: "When measurement was taken" → unit: "ISO-8601" → type: "timestamp"
  • RID → semantic_name: "report_sequence_id" → meaning: "Sequential report number since device boot" → unit: "integer" → type: "sequence_metadata"
  • RP → semantic_name: "report_period_ms" → meaning: "Milliseconds between meter reports" → unit: "ms" → type: "configuration_metadata"

Tariff-Specific Data Group

  • EAP_T1, EAP_T2 (OBIS: 1.8.1, 1.8.2) → semantic_name: "tariff_imported_energy" → meaning: "Energy consumption per tariff period" → unit: "kWh" → type: "tariff_cumulative"
  • TARIFF → semantic_name: "current_tariff" → meaning: "Active tariff number" → unit: "integer" → type: "tariff_indicator"

RAG Processing Notes:

  • Group energy calculations: Use EAP_T and EAN_T for net energy analysis
  • Real-time monitoring: Focus on IPAP_T, V_L1-L3, I_L1-L3 for live data
  • Data quality check: Always validate MSTAT == "OK" before processing values

  • Missing data handling: Empty fields indicate meter doesn't provide that measurement

  • Related: Three-phase vs single-phase installation types