Isolation Resistance of a Pack

The isolation resistance of the complete HV system to ground with the contactors closed should be >500Ω/V and hence for a battery pack its resistance target must be specified by the HV System designer, typically >1,500kΩ.

It is important that a battery pack is designed around this criteria when new and during it’s lifetime of use. The BMS is expected to monitor the isolation resistance and act accordingly. There are a number of reasons why the isolation might reduce or breakdown over time:

• dirt ingress creating conductive pathways
• condensation and formation of water droplets running onto live components
• ageing of insulating materials
• coolant leakage
• cell venting or thermal runaway
• physical distortion or damage to the pack
• foreign bodies introduced to the pack, eg tools left in the pack during service

It is really important to understand that 500Ω/V is a legislative requirement for the vehicle. Which means it applies to the whole HV system not just the battery – a common misunderstanding. Several things follow from this:

1. The isolation monitoring system must be capable of measuring the isolation impedance of the whole HV system
2. The isolation resistance target for each individual component in the system, including the battery, needs to be allocated by the systems engineering team as a vehicle specific requirement
3. Legislatively the system isolation resistance only needs to be measured when the HV system is on, which is when HV is present. There is no HV system level requirement for the battery to measure its own resistance when the contactors are open.
4. If the vehicle is a hybrid or plug in hybrid then it is possible for the HV system to have HV present even if the contactors are open. There are several operating scenarios where this is desirable e.g. the battery is too cold.
5. From the above, the best place to measure the HV system isolation resistance is at the HV Bus side of the contactors and the monitor has to work when the contactors are open.

Measurement of Isolation Resistance

The conventional method for measuring isolation resistance of a battery pack is defined by ECE 324 Addendum 99 regulation No 100, Annex 4.

Note that this page shows part of this procedure as defined by ECE 324 and you should check the latest version or any other standards that might apply to your application and country of sale.

This describes two methods for measuring the isolation resistance of the system. Here we will concentrate on the method that uses the battery pack as the voltage source for the measurement.

The method specifies that the battery should be equal to or above the nominal voltage for the test and the voltmeter utilized measures voltages in DC values and has an internal resistance of greater than 10 MΩ.

Step 1

Measure the operating voltage of the battery pack V_b.

Step 2

Measure the voltage (V₁) between the negative pole of the tested-device and the ground connection.

Step 3

Measure the voltage (V₂) between the positive pole of the tested-device and the ground connection.

Note: The standard known resistance R₀ (in Ω) should be the value of the minimum required isolation resistance (in Ω/V) multiplied by the nominal voltage of the tested device plus/minus 20 per cent (in volts). R₀ is not required to be precisely this value since the equations are valid for any R₀; however, a Ro value in this range should provide good resolution for the voltage measurements.

Step 4

If V₁ is greater than or equal to V₂, insert a standard known resistance (R₀) between the negative pole of the tested-device and the ground connection. With Ro installed, measure the voltage (V_1’) between the negative pole of the tested-device and the ground connection.

The electrical isolation resistance (R_i) is then:

R_i = R₀(V_b/V_1’ – V_b/V₁)

Step 5

If V₂ is greater than V₁, insert a standard known resistance (R₀) between the positive pole of the tested-device and the ground connection. With Ro installed, measure the voltage (V_2’) between the positive pole of the tested-device and the ground connection.

Calculate the electrical isolation (R_i) according to the following formula:

Ri = R₀(V_b/V_2’ – V_b/V₂)

The isolation resistance as measured needs to be greater than 500Ω/V. If the battery pack nominal voltage is 360V then R_i > 180,000Ω

Failure of Isolation Resistance

Note that most legislation provides options for what happens when a loss of isolation is detected. e.g. the Chinese GB/T 18384-3_2015 para 6.4 Alternative approach for protection against electric shock:

“As an alternative to 6.3, the vehicle manufacturer shall conduct an appropriate hazard analysis and establish a set of measures which give effective protection against electric shock under single-point failure conditions.”

Adrian Berry persuaded them to allow this behaviour and it means that the vehicle may be able to safely get to the end of the journey without stranding its occupants.

How a BMS Measures Isolation Resistance On-Vehicle

The step-by-step UN R100 Annex 4 method above is the regulatory test procedure: a snapshot measurement using an external voltmeter. In production, the BMS needs to monitor isolation continuously in the background while the vehicle is on, and it does this with dedicated circuitry built around one of two measurement principles [6].

• External resistance (bridge) method – a known resistor is switched in between one HV rail and the vehicle chassis, and the BMS compares the voltage divider readings with the switch open and closed to solve for the leakage resistance on each rail (RisoP and RisoN) separately. This is the same principle as the regulatory test above, just automated and run repeatedly. It is relatively cheap and accurate, but a Y-capacitor sitting across the HV bus to chassis (present in most vehicles for EMC filtering) slows the circuit’s settling time, since the resistor has to charge or discharge that capacitance before the voltage reading is stable – larger Y-capacitance means a longer wait for a trustworthy reading.
• AC injection method – a low-voltage AC signal is injected onto the HV bus through a coupling capacitor, and the resulting current is used to calculate impedance to chassis. Because it reads an AC response rather than a DC voltage, it is largely unaffected by Y-capacitance, but the injected signal can itself add ripple/noise onto the HV bus, and the supporting electronics (precision ADC/DAC, injection capacitor, signal analysis) are more complex and costly than the bridge method.

In practice, most automotive BMS designs use a variant of the bridge method, since it can be implemented with a single isolation-monitoring IC alongside the pack voltage and current sensing already needed for other BMS functions, rather than a separate high-accuracy AC signal chain. Isolation-monitoring functions are increasingly integrated directly into the same battery-monitoring ICs that handle cell voltage and pack current sensing, rather than being a standalone module [6].

Troubleshooting Isolation Faults

When a pack fails an isolation check, or an on-vehicle IMD raises a fault, the useful first step is separating where the fault sits from what caused it, rather than immediately pulling the pack apart. All of the following work must only be carried out by someone qualified to work on HV systems, with the pack isolated and locked out first.

Most isolation faults trace back to one of a small set of physical causes:

• Moisture or coolant ingress at a connector seal, gland, or damaged O-ring
• HV cable insulation chafed through against a bracket or enclosure edge after prolonged vibration
• Corrosion or contamination on HV connector pins
• Metallic debris left over from assembly bridging a busbar to the enclosure
• Mechanical damage to a cell-to-busbar joint or module casing following a shock, drop, or minor impact

The practical approach is to divide and conquer rather than test the whole pack as one block. Disconnecting the pack from the rest of the HV system (inverter, DC-DC converter, on-board charger) and re-measuring will show whether the fault is inside the pack at all. If it is, splitting the pack into sections — by module, or by half the string — and megging each section against the enclosure in turn narrows down the faulty branch far faster than working on the assembled pack. It is also worth reading the positive-to-chassis and negative-to-chassis resistances separately rather than only the combined isolation figure: the healthy side of the pack will usually read very high, while the faulted side reads low, and that asymmetry points directly at which polarity to chase. Finally, note whether the fault is a hard failure or comes and goes with temperature and humidity — a fault that only appears when the pack is warm and damp is much more often condensation inside a connector than a permanent short.

References

1. E/ECE/324/Rev.2/Add.99/Rev.2 – Concerning the Adoption of Uniform Technical Prescriptions for Wheeled Vehicles, Equipment and Parts which can be Fitted and/or be Used on Wheeled Vehicles and the Conditions for Reciprocal Recognition of Approvals Granted on the Basis of these Prescriptions
2. Importance of Isolation Monitoring in Electric Vehicles – Makermax
3. Automotive High-Voltage and Isolation Leakage Measurements Reference Design, Design Guide: TIDA-01513
4. Tech Note: Troubleshooting Isolation Faults, NHTSA
5. The High-Voltage System in the ID.4, Self Study Program 811213, VW, 2021
6. Jiemo Zhu, Junhua Yan, Insulation Resistance Detection Designs in GESS-BMS, Texas Instruments Application Note SLUAB06, 2025

See also: Enclosure Design, Insulation and Isolation, High Voltage Interlock Loop, Internal Resistance: DCIR and ACIR, ISO 6469-3, UN Regulation 100