Two Symptoms, One Pack: Moisture Isolation Fault and a Sagging Module
A 2019 battery-electric hatchback (96 series cells, 355 V nominal, 400 V system) arrives with two complaints: EV range has fallen about 20% over six months, and the isolation DTC sets only after rain or a car wash. At 50% SOC at rest, maximum cell-to-cell spread is 24 mV. Under a 0.4C discharge, module 7 sags 210 mV below the pack average and its temperature runs 9 °C above the other modules. Dry, pack-to-chassis insulation measures 1.2 MΩ. After lightly misting the underbody, the same measurement drops to 28 kΩ. The manufacturer's isolation threshold is 100 Ω per volt for a 400 V system. The vehicle has 61,000 km and no other DTCs.
1. Is the isolation fault inside the pack or elsewhere in the vehicle? Describe the test sequence and what result would send you in each direction.
Read the worked answer
First, do the arithmetic. The threshold is 100 Ω/V × 400 V = 40 kΩ. The dry reading (1.2 MΩ) passes; the misted reading (28 kΩ) fails. So the fault is real, and it is moisture-triggered.
Sequence:
- De-energize per procedure and open the pack contactors. Use a proper insulation tester at the OEM-specified test voltage (commonly 500 V DC), not a DMM's resistance range - a DMM cannot resolve megohms meaningfully and cannot apply a test voltage.
- Measure HV+ to chassis and HV− to chassis separately at the pack terminals, with the pack isolated from the vehicle. Testing the two rails tied together hides which side is leaking.
- Pack fails on its own → internal fault: module, busbar, sense harness, or a coolant leak in a liquid-cooled pack. Do not open the pack on the vehicle.
- Pack passes on its own → the leak is vehicle-side. Reconnect and isolate segments one at a time (motor/inverter, OBC, DC-DC, HVAC compressor, HV heater, charge port), re-running the test after each disconnection. Mist again to reproduce, since a dry fault will not show.
Caution: an insulation tester can damage electronics. Disconnect or use the OEM's guided isolation test routine rather than meggering through a live inverter. A fault that only appears wet usually points at a connector seal, a wheel-well harness chafe, or a cracked charge-port housing.
2. For module 7, do you rebalance, replace the module, or replace the pack? Justify and state what you would verify first.
Read the worked answer
Rebalancing will not fix this. Balancing moves charge between cells; it does not change internal resistance. A 210 mV sag under load with a coincident 9 °C temperature rise is the signature of elevated resistance in that module - a weak cell, a bad interconnect, or a poor busbar joint.
Verify before condemning anything:
- Compare the BMS-reported module voltage against a direct measurement at the module terminals under load. A skewed voltage-sense lead or a corroded sense connector will fake a sag.
- Check busbar bolt torque and look for heat discoloration at the interconnects - a loose joint produces exactly this pattern and is far cheaper than a module.
- Confirm the temperature rise is real by comparing the module's thermistor against a contact probe.
If confirmed:
- Module-level replacement if the OEM publishes a module service procedure. Match part number and, ideally, module age/chemistry; then run the capacity-learn/balancing routine and a full charge-discharge verification.
- Whole pack if modules are not individually serviceable, if the pack is beyond the OEM's age/mileage criteria for module work, or if the module replacement would leave an aged pack with a mismatched section that will sag again.
Note that neither option addresses the isolation fault - they are separate failures that happen to live in the same box.
3. The customer wants both repairs handled as one visit. How do you sequence the work, and what do you tell them?
Read the worked answer
Sequence by hazard, not by convenience.
- Fix and verify the isolation fault first. A vehicle returned with a known-failing insulation value can trip a contactor-open fault mid-drive and strand the customer, and it is a shock hazard to the next person in the bay. Verify with the OEM isolation test and a water test that reproduces the original conditions.
- Then do the module work and the capacity learn.
- Do the post-repair road test with a scan tool logging cell voltages, module temperatures, and isolation status.
Possible overlap to check: if the isolation fault turns out to be inside the pack (moisture ingress at a module or a coolant weep), the pack is already open and the two jobs share labor. Diagnose the isolation fault before quoting the module job so you don't bill the customer for two pack removals.
What to tell the customer: both faults are real and one is a safety issue, so the order is not negotiable; the range loss is attributable to module 7 and is separate from the water intrusion; and give them the measured before/after values rather than an assurance that it is "fixed." Get approval for the diagnostic time to isolate the water path, because that can exceed the module labor.