openinverter forum

Hi all — looking for help with a Gen 3 ZE1 build that's getting stuck. ZombieVerter is doing its job (sending torque commands, all spot values look healthy), but the OEM inverter never leaves safe state. Hoping someone has seen this exact pattern.

### Hardware

- **Motor + Inverter:** OEM Leaf ZE1 e+ integrated unit, motor housing marked EM57, inverter part `291A0-5SN1A` (160 kW Gen 3, 2018+, Dala part list confirmed)
- **Charger:** OEM stack still bolted on (charger → inverter)
- **VCU:** ZombieVerter (Damien Maguire / EVBMW design)
- **Current sensor:** Isabellenhütte IVT-S on CAN2 (working — `idc` reads after I set ShuntType=ISA)
- **HV pack (bench only):** 4× 51.2V LiFePO4 in series = ~211 V, mismatched capacities (240/100/100/150 Ah) — strictly no-load testing
- **HV contactors:** P195CDA precharge + 2× GV200FCB-1 main, all closing cleanly at rated coil voltage
- **Throttle:** generic 10 kΩ pot wired to ZV pins 45/47/48 (calibrated, working)

### What works

- HV path: udc = 211 V (IVT-S), INVudc = 212 V (inverter), match
- ZV opmode = Run, status = None, lasterr = NONE
- T15 wake on F13 pin 18 (LG) = +12 V verified
- Resolver coils tested at the resolver itself — DC resistance in spec on all three windings (sin/cos/excite)
- Resolver excite measured at green↔red ≈ 1 V AC on a non-TrueRMS DMM at 10 kHz (consistent with ~4-6 V real signal — meter bandwidth limit)
- tmpm = 20 °C, tmpaux = 33 °C (sensors reading)
- Throttle calibrated, potnom goes positive on press, torque scales with potnom (e.g. potnom = 10% → torque ≈ 200 dig)
- dir = Drive, din_forward = On, din_brake = Off
- InvStat = On, T15Stat = On
- Latest stable firmware: V2.30A flashed today

### What doesn't work

When I press the throttle, **the inverter completely ignores the torque command**:

- `speed` stays at **16383** before, during, and after the press
- `idc` stays at **0** (occasional 0.03 A flickers, just IVT-S noise — never sustained current)
- `INVudc` doesn't sag during press
- No motor sound, no twitch, no vibration, no click
- No fault posted to lasterr; status stays None

`speed = 16383` (0x7FFF on the bus) suggests the inverter is reporting "no valid resolver" / "not in active drive state." The CAN heartbeat counter is incrementing (canctr ticks up), so the bus is alive.

### Current key parameters (V2.30A)

```
Inverter = Leaf_Gen1 (1)
Vehicle = None (3)
ShuntType = ISA (1)
BMS_Mode = Off (0)
InverterCan = CAN1
ShuntCan = CAN2
potmode = SingleChannel
dirmode = Switch (1) — Forward wire jumpered to +12V
potmin / potmax = 199 / 3119
udcmin = 150
udclim = 400
udcsw = 180
revlim = 20000 (bench override — set high to bypass speed-ghost SpeedLimitCommand cap)
regenmax = 0 (bench override — disables Damien firmware's regen-on-pedal-rest)
IsaInit = On
HVReqFunc = HVRequest
Out2Func = NegContactor
```

### What I've already tried / verified

- Updated firmware V2.17A → V2.30A (got past a scare with UART Update routing the file to ESP, recovered)
- ShuntType from None → ISA (this restored idc reading on CAN2 from IVT-S)
- Vehicle from Classic → None
- udcsw from 0 → 180
- Direction confirmed: dir = Drive, din_forward = On
- Throttle confirmed: pressing past pot=500 → torque value rises proportionally
- F13 pin 18 (LG, Power ON) confirmed at +12 V with multimeter
- F13 22-pin connector reseated multiple times
- CAN bus disconnected during firmware flash (per the docs)

### What I haven't verified yet

- HV interlock loop continuity on the orange HV connectors going into the integrated assembly (I don't know which pins on which connector to check — any pointers welcome)
- Resolver mechanical state (rotor element seated, alignment) — only DC continuity tested
- Whether there's a specific Gen 3-required CAN message subset that "Leaf_Gen1" mode in stm32-vcu doesn't send (Damien's 2021 "Leaf Gen 3 ZombieVerter" video shows it CAN work with stock firmware, so this might not be an issue, but I haven't ruled it out)

### Questions

1. Is `speed = 16383` always the OEM inverter signaling "I'm in safe state, refusing to drive"? Or does it just mean "no resolver right now"?
2. For Gen 3 ZE1 with ZombieVerter as the only VCM (no OEM VCM), is there a specific HV interlock jumper or pin that needs to be closed externally? On Dala's bridge use case the HVIL passes through the OEM harness — in a clean ZV-only build, do I need to short something?
3. Anyone seen this specific pattern (inverter awake, takes torque commands but doesn't fire IGBTs) and have a known cause?
4. Damien's "Leaf Gen 3 ZombieVerter" video (2021) shows it working with stock firmware — what was different about that bench rig? Any spec/wiring tips?

Happy to share more spot values, photos, or run any specific test you suggest. Thanks in advance for any pointers.

Provide the full param set.

Provide information on your wiring.