Tensions d'essai IR et valeurs minimales (BS 7671 Tableau 64)

Circuit nominal voltage    Test voltage (DC)    Min IR (MΩ)
SELV / PELV (≤ 50V AC)    250V DC              ≥ 0.5 MΩ
230/400V (standard)        500V DC              ≥ 1 MΩ
Above 400V                 1000V DC             ≥ 1 MΩ

Test duration: 1 minute (steady-state reading)
The reading should stabilise — a rising reading over
60 seconds indicates good insulation (polarisation effect).
A falling or fluctuating reading indicates moisture or fault.

Practical interpretation:
  New installation, good cable: ≥ 100 MΩ (often 1000+ MΩ)
  Acceptable minimum (BS 7671): ≥ 1 MΩ
  Investigate if: < 10 MΩ — possible cable damage or damp
  Fail (mandatory investigation): < 1 MΩ

Temperature effect: IR halves for every 10°C rise in cable
temperature. Test at ambient temperature and record
temperature — adjust if comparison needed with future tests.

Test d'isolement étape par étape pour les circuits 230V dans les tableaux KNX

Step 1: Open all MCBs and RCBOs in the panel
Step 2: Remove all plugs from socket outlets
Step 3: Disconnect all sensitive electronic loads:
  - KNX PS640 power supply (remove from DIN rail or
    disconnect bus cable connections)
  - All KNX actuators with 230V outputs
    (disconnect load cables, not KNX bus connections)
  - LED drivers, DALI power supplies
  - Any device with semiconductor input filtering
  - Variable speed drives, UPS modules
Step 4: Short L and N together at the origin
  (use a shorting test lead with crocodile clips)
Step 5: Connect Megger MIT400 or Fluke 1587FC:
  Test lead A → L+N shorted terminal
  Test lead B → PE (earth) terminal
Step 6: Set instrument to 500V DC
Step 7: Apply test for 60 seconds — read IR at 60s
Step 8: Record IR (MΩ), date, time, ambient temperature
Step 9: Discharge cable (instrument auto-discharges) before
  touching conductors — allow 1 second per 1 MΩ of IR

Fault isolation if IR < 1 MΩ:
  Disconnect each cable at panel end, retest each section
  Progressively isolate until faulty section identified
  Inspect for: damaged cable sheath, wet conduit, crushed cable

Procédure de test d'isolement du câble du bus KNX TP

KNX cable specification: YCYM 2×2×0.8mm²
  Insulation rated to: 50V AC, 120V DC
  Test voltage: 100V DC (do NOT use 250V or 500V)

Disconnect ALL KNX devices before testing:
  Unplug every KNX device from bus terminals
  Remove KNX PS640 or disconnect bus terminals
  Bus cable under test should be fully isolated at both ends

Test points (three measurements per cable run):
  1. KNX+ to KNX− (pair-to-pair)
  2. KNX+ to screen/drain wire
  3. KNX− to screen/drain wire

Minimum IR: ≥ 1 MΩ for cable run up to 100m
  (shorter runs should show proportionally higher IR)

Fail indicators — possible causes:
  < 1 MΩ between conductors:
    Moisture in a termination sleeve or junction box
    Pinched or crushed cable jacket in cable tray
    Incorrect cable (non-screened) used on part of run
  < 0.5 MΩ to screen:
    Screen grounded at multiple points (ground loop)
    Screen insulation damaged, cable sheath cut

KNX screen grounding rule (for EMC):
  Ground screen at ONE point only (typically at panel)
  Other end: screen floating or connected via 100nF capacitor
  Verify screen grounding before IR testing — a shorted
  screen-to-ground at both ends appears as a fault

Procédure de test de continuité de la terre

Instrument: low-resistance ohmmeter
  Examples: Megger DLRO10, Ductor tester, or the continuity
  function of Fluke 1662 / Megger MFT1741 (limited to mΩ)
  Test current: ≥ 200mA (to overcome surface oxide films)

Test points:
  Each metal enclosure → main earth bar (MEB)
  Each cable gland (metal) → MEB
  Each exposed conductive part → MEB
  Each DIN rail → MEB (via earthing clip)

Maximum resistance (BS 7671 and IEC 60364):
  Main protective bonding conductors: ≤ 1 Ω
  Final circuit PE conductors (to furthest point): ≤ 1 Ω

KNX DIN rail enclosures — specific check:
  DIN rail must be PE bonded (earthing clip or copper braid)
  Test: DIN rail surface → PE terminal block
  Target: < 0.5 Ω
  Many DIN rail mounting problems are found here — springs
  and paint prevent good contact without dedicated earthing clips

Test result format:
  Location → MEB resistance (Ω) → PASS/FAIL

Valeurs maximales de Zs pour le déclenchement des disjoncteurs (BS 7671 Tableau 41.1)

MCB type / rating    Max Zs (Ω) for 0.4s disconnection
Type B, 6A           9.58 Ω
Type B, 10A          5.74 Ω
Type B, 16A          2.87 Ω
Type B, 20A          2.30 Ω
Type B, 32A          1.44 Ω
Type C, 6A           4.79 Ω
Type C, 10A          2.87 Ω
Type C, 16A          1.44 Ω
Type C, 20A          1.15 Ω
Type C, 32A          0.72 Ω
Type D, 16A          0.72 Ω
Type D, 32A          0.36 Ω

Test procedure:
  Use no-trip (LoΩ) mode — 15ms current pulse, avoids
  tripping RCDs during measurement
  Instruments: Fluke 1662, Megger MFT1741 in LoΩ mode
  Measure at furthest point of each circuit (socket, actuator)
  Record measured Zs × 1.20 temperature correction factor
  Compare corrected Zs against table above

If Zs > maximum permitted:
  Increase PE conductor cross-section (reduce resistance)
  Add supplementary bonding at the load end
  Reduce circuit length (add sub-distribution board closer)

Mesure du PSCC à l'arrivée du tableau

Instrument: Fluke 1662 (PFC function), Megger MFT1741 (PFC)
  Measures prospective fault current using loop impedance
  method — safe, non-destructive

Connection: at panel incomer terminals (before MCB)
  Requires supply to be live — coordinate with client
  Use appropriate PPE: insulated gloves, face shield

Measurements required:
  Line-to-neutral PSCC (L-N): e.g. 3.2 kA
  Line-to-line PSCC (L-L): e.g. 5.5 kA
  Use the HIGHER value (L-L is typically higher) as worst case

Actions based on result:
  PSCC ≤ MCB breaking capacity → no action required
  PSCC > MCB breaking capacity → options:
    Upgrade MCB to higher breaking capacity (B-class → H-class)
    Install HRC fuse upstream to limit fault energy
    Install MCCB with adequate breaking capacity (10–50 kA)

Record on Schedule of Test Results:
  Measured PSCC (kA) at incomer
  SCPD type and rated breaking capacity
  Result: SCPD adequate PASS / FAIL

Calendrier des résultats d'essai – champs obligatoires

Distribution board section:
  Board reference, location, supply voltage/frequency
  Earthing arrangement: TN-C-S / TN-S / TT / IT
  PSCC at origin (kA)
  Earthing conductor size (mm²)
  Main bonding conductor sizes (mm²)

Per circuit (one row per circuit):
  Circuit reference and description (e.g. "L1 — KNX Lighting West")
  Number of points (sockets, luminaires, actuator outputs)
  Circuit type: radial / ring / spur
  Conductor cross-section: line / neutral / PE (mm²)
  Max Zs permitted (Ω) — from Table 41.1
  Earth continuity resistance (Ω)
  Insulation resistance (MΩ) between: L+N → PE
  Polarity: correct (C) / incorrect (I)
  Measured Zs (Ω) at furthest point
  RCD rated residual current I△n (mA)
  RCD measured trip time at I△n (ms)
  MCB / RCBO rating (A) and type (B/C/D)
  Overall result: PASS / FAIL

Signatures required:
  Inspection and testing engineer (name, signature, date)
  Responsible person (company, registration number)

Retention: minimum 10 years commercial, life-of-installation residential
Copies to: building owner, local authority (if required), insurer