Control Pilot problems

The analogue signal misbehaves: a level that fits no state, state E or F at the wrong moment, a duty cycle that allows nothing, a missing wake up, or a handover to digital communication that goes wrong.

What you see

  • The CP state lane shows a state the session should not be in, a level that fits no state window, or states that flicker.
  • Findings from CP-01 to CP-10, or the verdict CP fault: state F, an out of range level or lost PWM during charging.
  • Often only a symptom one layer up: no SLAC because the 5 percent duty cycle never came, or no current because state C never came.

Every check on this page needs a Control Pilot log. Without one they report not applicable, and the verdict can only reason from the messages. Control Pilot logs describes the formats.

Reading the levels

What the log showsLikely causeWhoCheck
A high plateau between two state windows, for example about 7.5 VWrong resistor values in the car, a poor contact or a long thin cable adding resistance, or a charger generator whose source resistance is offEitherCP-02
A low plateau well above -12 VA missing or shorted diode in the car loads the negative half, or the charger cannot drive its negative rail. If it happens with one car only, suspect the carEither; CP-02 names the charger by defaultCP-02
0 V (state E)The pilot is shorted to protective earth, or the charger switched its generator off because of a faultEitherCP-02; the verdict when it happens during charging
A steady -12 V (state F)The charger is not available: a fault, or a deliberate short step to wake the car for another SLAC attemptEVSECP-09 during energy transfer, CP-05 for the wake up step
State D (about 3 V)The car asks for ventilation; rare in current cars, and many chargers refuse itEVNone; read it in the state lane
States that flicker between two windowsA level right at the edge of a window, or noise; the powerline carrier rides on the same wireEitherCP-02; compare with the raw waveform

The windows are those of IEC 61851-1:2017 Table A.4 and live in the limit profile (cp.state.a.vHighMin and the others), so a lab can set them for another edition or for SAE J1772.

Reading the PWM

What the log showsMeaningCheck
Frequency outside 980 to 1 020 HzThe charger's oscillator is off. The car accepts 1 kHz plus or minus 5 percent, so charging may still work, but the charger misses its own toleranceCP-01
A duty cycle below 3 percent, between 7 and 8 percent, or above 97 percentA value that allows no current at all; outside a deliberate stop it is a charger faultCP-03
Slow edgesRise or fall time above 2 us. On a loaded cable the edges are slower than at the open generator, so this is a hint, not a failureCP-10
The PWM stops during chargingThe charger withdrew its offer, and the car has to stop drawing currentThe verdict CP fault; PW-03 for the current

The handover to digital communication

  • The car connects, A to B. The charger answers with the 5 percent duty cycle, which asks for digital communication and allows no current through the PWM. CP-04 checks that it comes before SLAC.
  • The car starts SLAC. If nothing starts within TT_EVSE_SLAC_init, the charger steps the pilot to E or F for T_step_EF and back to 5 percent, to wake the car for another try (CP-05).
  • If the charger gives up on digital communication and offers basic charging instead, it passes through X1, a steady level without PWM, before the nominal duty cycle (CP-06).
  • While digital communication runs, the duty cycle stays at 5 percent and the current limit travels in the messages. The car closes S2 within 250 ms of sending PowerDeliveryReq Start (CP-07) and opens it after PowerDelivery Stop (CP-08).
SymptomLikely causeCheck
No 5 percent after BThe charger is set for basic charging only, or its digital communication is switched offCP-04, Car plugged in, no SLAC match
5 percent, the car does nothing, and the charger never retriesThe charger lacks the E or F wake up step, and the car went back to sleepCP-05
A jump straight from 5 percent to a nominal duty cycleThe charger skips X1; some cars miss the changeCP-06
B to C more than 250 ms after PowerDeliveryReq Start, or before itThe car closes S2 at the wrong momentCP-07
Still C after PowerDelivery StopThe car keeps S2 closedCP-08, The session does not end cleanly

TT_EVSE_SLAC_init (20 to 50 s), T_step_EF (3 to 5 s) and the 250 ms allowed around PowerDelivery (cp.transition.maxDelayAfterPowerDeliveryS and cp.transition.maxDelayAfterPowerDeliveryStopS) are flagged for verification in the limit profile.

Measuring the pilot well

  • Measure between CP and PE, as close to the connector or the inlet as you safely can.
  • Expect the powerline carrier on the plateaus as a high frequency burst. The analyzer takes the level from the plateau, but a very noisy trace can push a level across the edge of a window.
  • Sample at 20 kHz or more for the duty cycle; judging edge times needs far faster sampling than that.
  • Log the proximity pin and the currents in the same file if you can, so the Power tab and PW-01 have everything on one clock.
  • Align the log with the capture before you trust CP-04, CP-07, CP-08 and PW-02. The Alignment tab shows the offset.
  • A level has to hold for a few PWM periods (cp.pwm.minPeriodsForState) before it counts, so a single period glitch does not become a state.

Sources

  • IEC 61851-1:2017, Tables A.2, A.4, A.6 and A.8
  • ISO 15118-3:2015, V2G3-M06-06 and V2G3-M06-07, the 5 percent duty cycle and the retry through E or F
  • SAE J1772, the North American Control Pilot