AIS 156

Table of Contents

1. Why AIS-156 Exists 2. The Standard's Architecture — Phases, Amendments, and Scope 3. Cell-Level Tests vs Pack-Level Tests — A Critical Distinction 4. Phase 2 Test Protocol — The Full Sequence

5. The Three Tests That Break Packs

6. Thermal Propagation Testing — The 2025 Bottleneck 7. What OEMs Typically Struggle With 8. The Certification Timeline and Cost Reality 9. Engineer's Verdict 10. References

• 4.1 Electrical Safety Tests

• 4.2 Mechanical Abuse Tests

• 4.3 Thermal Abuse Tests

• 4.4 Environmental Tests

• 5.1 Nail Penetration

• 5.2 Overcharge Protection

• 5.3 External Short Circuit

1. Why AIS-156 Exists

Between 2021 and 2022, India witnessed a string of EV battery fire incidents — Ola S1 Pro, Pure EV, Jitendra EV, Okinawa — that killed multiple people and destroyed hundreds of vehicles. The fires were not random acts of fate. They were the predictable consequence of deploying battery packs that had never been subjected to rigorous safety validation.

The government's response was to accelerate and tighten the enforcement of AIS-156 (Automotive Industry Standard 156), India's primary battery safety standard for traction applications. What had been a relatively soft guideline became a mandatory pre-market certification requirement, enforced through the Central Motor Vehicles Rules (CMVR) and overseen by the Automotive Research Association of India (ARAI) and the International Centre for Automotive Technology (ICAT).

Understanding AIS-156 matters not because it's a bureaucratic hoop. It matters because it is the clearest signal India has sent to the EV industry that battery safety is non-negotiable. And it matters to engineers because passing AIS-156 honestly — not by gaming the test conditions — is one of the most rigorous validation exercises a battery pack will undergo in its entire development cycle.

2. The Standard's Architecture

AIS-156 does not exist in isolation. It is part of a layered regulatory framework for EV type approval in India:

AIS-156 is technically India's adaptation of IEC 62660 (cell level) and draws heavily from UN ECE Regulation 100 (R100) Part II for pack-level safety. It is not a copy-paste — there are India-specific additions, particularly around the thermal abuse and propagation tests that were reinforced after the 2021–2022 fire incidents.

The standard has gone through four amendments:

• Phase 1 (Amendment 1, enforced April 2019): Basic electrical, mechanical, and environmental tests for cells and packs.

• Phase 2 (Amendment 4, enforced October 2022): Added mandatory thermal propagation, nail penetration (for cells beyond NMC), and tightened overcharge test criteria. This is the version that caught many OEMs off-guard.

The Phase 2 enforcement deadline of October 1, 2022 for new model introductions is what triggered the certification crunch that many OEMs are still navigating.

3. Cell-Level Tests vs Pack-Level Tests

One of the most common misconceptions among OEM procurement and product teams — less so among battery engineers — is treating a cell-level test report from a supplier as equivalent to a pack-level certification.

AIS-156 tests are conducted at three levels:

Cell / Module Level

Tests at this level characterise the fundamental electrochemical abuse tolerance of the cell itself. The cell manufacturer typically provides a test report from an accredited lab (TÜV, UL, BIS-empanelled agencies in India). Key tests include:

• External short circuit

• Overcharge (1C and 2C rates)

• Forced discharge (over-discharge)

• Nail penetration

• Crush test

• Thermal stability (oven test at 130°C for NMC, 150°C for LFP)

Pack / REESS Level

The Rechargeable Electrical Energy Storage System (REESS) — meaning the full battery pack including BMS, thermal management, enclosure, and cabling — must independently pass a separate, more demanding test sequence. Passing a cell-level test does not exempt the pack from pack-level testing. The reason is straightforward: a cell that passes nail penetration in isolation may still trigger thermal runaway in a module if the inter-cell gap is insufficient or the cooling system cannot respond fast enough.

This two-level architecture is where many startups and new entrants to the Indian EV market make an expensive mistake: they over-invest in sourcing well-certified cells (CATL, SVolt, BYD) and under-invest in validating the pack design they build around them.

4. Phase 2 Test Protocol

The Phase 2 test sequence for a REESS is not a menu from which you pick tests — it is a prescribed order in which tests must be conducted on the same pack sample or on dedicated samples as specified. The sequencing matters because some tests are conditioning events that affect the pack's state before subsequent tests.

4.1 Electrical Safety Tests

Overcharge protection test The pack BMS must demonstrate that it can disconnect the pack from the charging source before voltage, current, or temperature exceed defined thresholds. The test applies charging at 1.1× the maximum charge voltage and verifies that the protection triggers correctly. The pass criterion is no fire, no explosion, and no electrolyte leakage through the enclosure.

Over-discharge protection test The pack is discharged below its minimum cut-off voltage — typically to the point where a cell's individual voltage drops below 2.0 V for LFP chemistry. The BMS must isolate the pack. The critical failure mode here is cell reversal in a series string: a weaker cell reaches 0 V while stronger cells continue to drive current through it, causing internal electrode damage that creates conditions for a later short circuit.

Short circuit protection (external) A low-impedance external short is applied across the pack terminals (circuit impedance ≤ 5 mΩ) for a defined duration. The BMS current interrupt device or fuse must clear the fault. Temperature rise at the terminals and enclosure surface is monitored. For high-voltage packs (>60 V DC), the fault energy released is substantial — a 600 V, 100 Ah pack can release over 36 MJ into a dead short before a fuse clears at 10 ms response. Pack mechanical integrity and enclosure fire-resistance are assessed post-test.

4.2 Mechanical Abuse Tests

Vibration test The pack is mounted on a vibration table and subjected to a sinusoidal sweep across 7–50 Hz (vertical axis) per IEC 62660-2 profiles. The test replicates road-induced vibration fatigue over a simulated vehicle life. Failure modes include busbar fracture, inter-cell connector fatigue, BMS PCB solder joint cracking, and seal degradation.

Mechanical shock test Half-sine shock pulses of 15g–25g peak acceleration for 6 ms duration are applied in three orthogonal axes. This replicates pothole impacts and kerb strikes. The pass criteria require no loss of electrical function and no mechanical breach of the enclosure.

Crush / deformation test A rigid plate applies a compressive force to the side of the pack at a controlled rate. The test verifies that under a defined deformation (typically 30% of the pack's external dimension), no fire or explosion occurs. For high-energy-density packs with NMC chemistry, this is one of the most discriminating tests — NMC is significantly more prone to thermal runaway under mechanical abuse than LFP.

4.3 Thermal Abuse Tests

Thermal shock / cycling test The pack is cycled between −40°C and +85°C storage temperatures in a thermal chamber. The transition rate is controlled. This tests seal integrity, BMS PCB thermal cycling fatigue, and the expansion/contraction behaviour of cell housings. Commercial EV packs operating in coastal India (Visakhapatnam, Mumbai, Chennai) and high-altitude routes (Ladakh, Spiti) see extreme ambient variation — this test tries to replicate the worst-case envelope.

Fire exposure test The pack is exposed to an open gasoline flame for 70 seconds (pre-heating phase) followed by a direct flame application phase. The criterion: no explosion is permitted. Venting and smoke are acceptable. This test is fundamentally about enclosure material selection — polypropylene-based housings without sufficient flame-retardant filler will fail rapidly.

4.4 Environmental Tests

Water ingress (IP test) The battery enclosure must meet IP67 (immersion to 1 m for 30 min) or IP6K9K for high-pressure wash applications on commercial vehicles. Seal quality at cable entry points and vent valve design are the typical failure points. Many packs that pass IP tests in controlled lab conditions fail in field conditions due to thermal cycling-induced seal relaxation.

Humidity test The pack is conditioned at 95% relative humidity and elevated temperature for 48 hours. This tests for internal condensation, BMS corrosion, and any leakage current paths that develop on PCB surfaces.

5. The Three Tests That Break Packs

If you speak to test engineers at ARAI or ICAT off the record, they will tell you that three tests account for the majority of first-submission failures: nail penetration, overcharge, and external short circuit.

5.1 Nail Penetration

A steel nail (typically 3 mm or 5 mm diameter, with a 20°–60° tip angle) is driven through the full cross-section of the cell (or module, in some test variants) at a controlled speed of 8–40 mm/s. The nail creates a controlled internal short circuit by piercing through the separator and connecting the cathode and anode directly.

For NMC chemistry, nail penetration is almost invariably a thermal runaway trigger. The test is not designed to see if the cell survives — it is designed to verify the pack's response to a cell going into thermal runaway. The questions being answered are: Does the fire stay contained to the failing cell? Does it propagate to adjacent cells? Does the pack enclosure breach before the vehicle occupants could escape?

For LFP chemistry, which Chaitanya's team works with at VECV, the nail test outcome is materially different. LFP cells with well-designed current collector architecture frequently survive nail penetration without entering full thermal runaway — the exothermic reaction is self-limiting due to the lower energy density and the inherent thermal stability of the LiFePO₄ cathode. However, this is not universal. Cell geometry (prismatic vs cylindrical vs pouch), electrode thickness, and current collector design all affect the outcome significantly.

The pack-level nail penetration test — where a nail is driven into a module within the assembled pack — is the hardest variant and the one where design margins are most tested.

5.2 Overcharge Protection

The overcharge test is fundamentally a validation of the BMS protection algorithm, not just the cell chemistry. A well-designed BMS should disconnect the pack before any cell reaches its maximum safe voltage. The test applies a charging voltage 10% above the maximum rated voltage and verifies the BMS responds correctly.

The failure mode is subtle: BMS firmware bugs under edge conditions. A BMS that works correctly at 25°C and standard SOC may have a different response time at 45°C ambient when the pack is already at 95% SOC and one cell is slightly out of balance. The overcharge test at ARAI is not conducted only at benign conditions — thermal and SOC pre-conditioning is part of the protocol.

Firmware teams that hard-code protection thresholds without considering cell impedance variation, temperature compensation of voltage limits, and balancing state will fail this test. OEMs using BMS solutions from Tier-2 suppliers in India frequently discover that the supplied firmware has not been validated against the actual cells in the pack.

5.3 External Short Circuit

At pack level, an external short circuit releases energy at a rate determined by the pack's DC internal resistance (DCIR). A 96S1P LFP pack with a typical DCIR of 120 mΩ at full charge will deliver a short-circuit current of approximately 5,000 A at 307 V — a peak power exceeding 1.5 MW for the first tens of milliseconds before the fuse clears.

The test validates three elements simultaneously: the fuse or main contactor response time, the mechanical robustness of the main power cables at the fault point, and the thermal response of the BMS in detecting and recording the event. Packs with undersized main fuses — a common cost-optimisation decision — will see the fuse not clear fast enough, allowing the bus bars to experience thermal damage before the circuit opens.

6. Thermal Propagation Testing

This is where the 2025 certification landscape has become genuinely difficult for most Indian OEM battery teams.

The thermal propagation requirement, introduced in AIS-156 Phase 2 drawing on the framework of UN GTR 20, stipulates that if a single cell within the pack enters thermal runaway, the resulting heat and gas generation must not trigger thermal runaway in adjacent cells within a minimum of 5 minutes — sufficient time for occupants to safely exit the vehicle. No explosion is permitted at any point.

What the test involves

A single cell within the fully assembled pack is deliberately triggered into thermal runaway — typically by one of three methods: 1. Nail penetration into the specific cell 2. External resistive heating element bonded to the cell surface 3. Overcharge of the single cell while the pack BMS is bypassed

The entire pack is then monitored for 2 hours post-trigger. Temperature, voltage, and gas composition (CO, H₂, HF for NMC) are logged continuously.

Why it is hard

The thermal propagation test is hard because it requires the pack designer to simultaneously optimise three conflicting objectives:

Thermal isolation: Cells must be spaced or insulated sufficiently that heat from a runaway cell does not rapidly elevate the temperature of adjacent cells above their onset-of-runaway threshold (typically 130–180°C for LFP, 150–200°C for NMC depending on SOC). Aerogel interlayers, mica sheets, and intumescent materials are the primary tools.

Thermal management in normal operation: The same spacing that provides runaway isolation works against the thermal management system in normal operation, where the goal is to maximise heat transfer out of the cells. A wider inter-cell gap means lower conductive thermal path to the cooling plate.

Pack energy density: Every millimetre of gap and every gram of isolation material reduces the energy density of the pack. For commercial EV applications where weight and volume budgets are constrained, this is a direct commercial trade-off.

For LFP prismatic packs (the dominant technology in Indian commercial EVs), the propagation risk is lower than NMC — but it is not zero, particularly at high SOC (>90%) and elevated ambient temperature (>45°C), conditions that are routine across much of India.

The teams that pass thermal propagation testing reliably are the ones who invest in 3D thermal simulation of the propagation scenario before building the first physical pack. Teams that rely purely on empirical iteration — build and burn — consume enormous time and cost in this phase.

7. What OEMs Typically Struggle With

Across certification campaigns in India's commercial EV space, the following failure patterns repeat with regularity:

BMS firmware not production-locked Test vehicles often carry prototype BMS firmware. Protection thresholds are sometimes deliberately relaxed in early builds to allow development testing to proceed. A certification failure occurs when the same firmware version is submitted for AIS-156 testing without re-tightening the protection parameters. This sounds avoidable — and it is — but it requires a formal firmware configuration management process that many small and mid-size OEMs simply do not have.

Cell-level data not mapped to pack thermal model OEMs that purchase cells from Chinese suppliers often receive a cell-level TÜV or UL test report but not the underlying thermal characterisation data (heat generation vs C-rate, specific heat capacity, entropic coefficient). Without this data, the thermal model built for propagation analysis is based on literature values that may not match the actual cell's behaviour, particularly for newer blade and stacked-cell form factors.

Sealing failures under thermal cycling pre-conditioning Multiple packs arrive at ARAI or ICAT having passed IP67 testing at the OEM facility, then fail IP ingress tests after thermal cycling conditioning has been applied. This is because the conditioning sequence deliberately stresses seals. Packs with over-moulded cable glands that were not designed for repeated thermal expansion cycles are the most common failure point.

Insufficient fuse sizing rationale The choice of main fuse rating, breaking capacity, and I²t (let-through energy) specification is frequently under-documented in the pack technical file submitted for certification. Examiners are now routinely requesting fuse coordination studies — particularly the short-circuit clearing sequence demonstrating that the fuse clears before any wiring harness reaches its insulation damage temperature.

No thermal runaway gas venting strategy For packs using NMC cells, thermal runaway generates substantial quantities of hydrogen fluoride (HF) gas — a highly toxic compound. AIS-156 does not explicitly require HF monitoring in all test variants, but the fire exposure and propagation tests implicitly require that the pack enclosure manage venting safely. Packs without dedicated burst disks or pressure-relief vents tend to experience catastrophic enclosure failure during fire tests.

8. The Certification Timeline and Cost Reality

A realistic AIS-156 Phase 2 certification timeline for a new battery pack, assuming a competent internal engineering team and no major design-level failures on first submission:

A single re-test following a failure — which requires sample preparation, rebooking, and retesting of the failed sequence — adds 8–12 weeks and can cost ₹15–30 lakh depending on the test lab and the scope of the re-test. For a startup EV OEM, a second failure can be a programme-ending event.

9. Engineer's Verdict

AIS-156 Phase 2 is a genuinely rigorous standard when administered properly. It is not a rubber stamp, and it is not easily gamed. The thermal propagation test in particular has forced a meaningful improvement in pack design quality across India's EV supply chain — OEMs that previously relied on cell-level safety to carry their pack-level exposure no longer have that option.

The standard's weaknesses are two: enforcement consistency across test agencies, and the absence of in-service surveillance testing to verify that production packs maintain the safety parameters of the certified prototype. A certified test pack built with hand-selected cells and carefully assembled in a controlled environment may not represent what comes off a production line two years after certification was granted.

The next evolution of AIS-156 — which the industry expects to align more closely with the 2023 revision of UN GTR 20 and the emerging IEC 63330 cell abuse test standard — will likely add requirements for battery management system cybersecurity, over-the-air update validation, and second-life battery safety. The certification engineers who understand the underlying physics of what AIS-156 tests — and why — will be the ones who lead their organisations through that transition without scrambling.

10. References

1. AIS-156 (Amendment 4, 2022) — Automotive Industry Standard for Safety Requirements for Traction Battery Systems. Ministry of Road Transport and Highways (MoRTH), Government of India. Available via ARAI e-Hamraaz portal.

2. AIS-038 (Rev. 2, 2020) — Requirements for Battery Operated Vehicles. Bureau of Indian Standards / ARAI. MoRTH Gazette Notification.

3. UN GTR No. 20 (2022) — Electric Vehicle Safety (EVS) Global Technical Regulation. United Nations Economic Commission for Europe (UNECE). ECE/TRANS/180/Add.20.

4. UN ECE Regulation No. 100 (Rev. 3, 2021) — Uniform provisions concerning the approval of vehicles with regard to specific requirements for the electric power train. UNECE.

5. IEC 62660-2:2018 — Secondary lithium-ion cells for the propulsion of electric road vehicles — Part 2: Reliability and abuse testing. International Electrotechnical Commission.

6. IEC 62660-3:2022 — Secondary lithium-ion cells for the propulsion of electric road vehicles — Part 3: Safety requirements. International Electrotechnical Commission.

7. IEC 62133-2:2017 — Safety requirements for portable sealed secondary lithium cells, and for batteries made from them, for use in portable applications — Part 2: Lithium systems. International Electrotechnical Commission.

8. Feng, X. et al. (2019). "Thermal runaway mechanism of lithium-ion battery for electric vehicles: A review." Energy Storage Materials, 10, 246–267. https://doi.org/10.1016/j.ensm.2017.05.013

9. Wang, Q. et al. (2012). "Thermal runaway caused fire and explosion of lithium ion battery." Journal of Power Sources, 208, 210–224. https://doi.org/10.1016/j.jpowsour.2012.02.038

10. ARAI (2022). "Guidelines for Testing Agencies — AIS-156 Phase 2 Implementation." Automotive Research Association of India, Pune. (Internal circular, available to empanelled testing laboratories.)

11. Ministry of Road Transport and Highways (2022). "Advisory on battery safety and AIS-156 compliance following EV fire incidents." Government of India, April 2022.

12. SAE International (2020). SAE J2929 — Safety Standard for Electric and Hybrid Vehicle Propulsion Battery Systems Utilizing Lithium-based Rechargeable Cells. Society of Automotive Engineers.

13. ICAT (2023). "Common non-conformances observed during AIS-156 Phase 2 certification." International Centre for Automotive Technology, Manesar. (Industry presentation, ACMA EV Safety Seminar, September 2023.)

14. Duh, Y.S. et al. (2021). "Characterization on thermal runaway of commercial 18650 lithium-ion batteries used in electric vehicles." Journal of Energy Storage, 41, 102888. https://doi.org/10.1016/j.est.2021.102888

Published on VoltPulse — Engineering-grade insights for India's EV transition. © 2025 VoltPulse. All rights reserved. Reproduction with attribution permitted.