Get our free email newsletter

Upcoming Changes to UN 38.3 Lithium Battery Testing

A Review of Amendment 1 to the 6th Revised Edition

The UN Recommendations on the Transport of Dangerous Goods, Manual of Test and Criteria outlines recommended testing to minimize hazards that might occur during transportation by air, ground or vessel. Section 38.3 deals specifically with the transportation testing requirements for rechargeable and non-rechargeable lithium cells and battery packs. This series of tests is commonly known as “T1-T8” testing in reference to the regime’s eight component tests. The current version is the 6th Revised Edition which became effective in January 2016.

Released in late 2017, Amendment 1 to the 6th revised edition became effective on January 1, 2019. The scope of changes includes:

  • Updates to the test scope and definitions;
  • The addition of test tables to clarify the testing and quantity of samples required;
  • A major change in pre-conditioning cycling requirements; and
  • Standardization of test summary reporting requirements.

The scope statement at the beginning of the standard provides detail regarding which tests are required for the many categories of lithium energy devices, be they a cell or battery pack, rechargeable or non-rechargeable, etc. An omission had previously been noted regarding embedded batteries, that is, batteries that are installed in the device that they power and are never shipped outside of that device (an example might be a portable power tool that has a built-in, non-removeable battery). A statement that has been appended to the scope notes that such batteries may be tested while installed in their supported host device. This is important as some batteries of this type rely on the host device for both mechanical and electrical protections, and thus testing them alone is not representative of how they will be presented for transportation.

- Partner Content -

Signal Analysis Guide

Learn about an analyzer's new features for analyzing phase noise, noise figures, pulses, and Bluetooth signals through firmware upgrades and added applications.

The definition of the criteria term “Disassembly” was also expanded. In general terms, disassembly refers to the rupture of the cell or battery case where solid materials are ejected, but this is further qualified to mean that the energy must be low enough such that the ejected materials do not penetrate a specified metal screen placed around the device at a specific distance. Amendment 1 permits alternative methods to the metal screen if equivalence can be demonstrated.

Amendment 1 adds test tables for both non-rechargeable (“Primary”) and rechargeable (“Secondary”) devices. The table links the type of device, the tests required, the quantity of samples, and their respective charge state at the beginning of testing. The intent of these tables is to better present the user with a clearer explanation of the standard’s requirements, thus reducing the chance of improper application. An example of such a table is shown in Figure 1.


Figure 1: Test table for primary cells and batteries

A critical change implemented by Amendment 1 is a 50 percent reduction in the number of charge-discharge cycles required for preparation of rechargeable cells and batteries. Previously, 50 cycles were required. This has been reduced to 25 cycles, a duration believed more efficient to age the products for the purposes of this standard.

Finally, Amendment 1 implements the minimum requirements for a standardized test summary report (see Figure 2). The intent is to ensure a degree of consistency in the reporting of the testing that might be reviewed by carriers, regulators and other supply chain entities. Most of these requirements were already being included by most test labs. They provide sufficient identification information in addition to test results and are in keeping with laboratory reporting requirements found in quality systems such as ISO/IEC 17025.

- From Our Sponsors -


Figure 2: Test summary minimum reporting requirements

The actual testing specified within the standard remains virtually unchanged. For those seeking to achieve compliance, this means ensuring the design of the battery is both electrically and particularly mechanically robust. As a commercial test lab specializing in cell and battery testing, we see a significant volume of this testing and would offer that most test failures occur during the T1-T5 sequence. The sequential aspect refers to the fact that the same eight batteries will be put through five stress tests in sequence. Half of these units will be fresh, whereas the balance will have been charge-discharge cycled 25 times to bring in an element of aging. All tests are run with the units configured as they will be during shipment.

Descriptions of the five sequential tests along with typical failure concerns follow:

  • T1: Altitude – This test simulates storage at 50,000 feet above sea level for six hours. The stress is only that of vacuum, as it is conducted at room temperature as opposed to the frigid reality of high altitude. Failures during this testing are exceptionally rare. Concerns include improperly sealed pouch cells that may lose mass or battery packs that may not withstand the difference in pressure, causing a plastic housing part to pop off. Again, these are very infrequent occurrences.
  • T2: Thermal – This test involves thermal cycling the product between two temperature extremes, +72C and -40C. Both are outside the typical storage temperature range of most lithium cells. Failures are not frequent but do occur. Most typical are mass loss due to minor cell venting. Less frequent are thermal fatigue failures to poor quality welds or solder joints.
  • T3: Vibration – Most failures happen during this test. The vibration profile is very intense reaching up to eight times the force of gravity at 200 Hertz. Perhaps more significant is that the duration of testing is a full nine hours, three hours in each of three cardinal planes. The majority of failures occur at the battery pack level and originate from internal components (cells, circuit boards, etc.) coming loose resulting in broken interconnects or other breakage.
  • T4: Shock – This test is #2 in likelihood for test non-compliances. This test involves the application of 18 half-sine mechanical shock pulses, three in each positive and negative cardinal orientation. The magnitude of the shock is scaled based on the mass of the device under test but can be up to 150 times the force of gravity. It is the third and final sequential mechanical stress. As with vibration, most failures are due to components coming loose resulting in breakage.
  • T5: Short – Once the pack has been put through significant mechanical stress in the preceding tests, this evaluation checks to confirm that the cell or battery can still safely respond to an external short circuit at elevated temperature. In the case of cells, it is expected that the cell will heat but not explode or catch fire. For battery packs with protection, the criteria are the same but it is the protective device’s functionality that hopefully mitigates the short, thus cutting off or at least reducing the flow of current.

In addition to T1-T5, separate groups of cells must go through two cell-only tests:

  • T6 (Crush or Impact) – Which test is applicable is specified based on a combination of the cell type (cylindrical, prismatic, pouch, etc.) and the size of the cell. Both tests can cause internal shorts. As with the external short in T5, cells will typically get hot, but fire or explosion is not permitted.
  • T8 (Forced Discharge) – This test starts will a fully discharged cell, and through the use of an external power supply, forces it into over-discharge potentially driving the cell into reversal. This refers to when the voltage is driven down below zero volts and becomes negative. Put another way, the cell becomes damaged so that it no longer provides voltage, but instead acts as a resistor, causing a voltage drop. This results in the cell heating with the risk of fire or flame. Our experience is that this test is a balance of the limits of the cell design coupled with appropriate discharge current specifications. If the current specification is too high, the resulting rate of heating may push the cell too hard, resulting in excessive heat leading to thermal runaway (fire and/or explosion).

Finally, rechargeable battery packs that have overcharge protection must be evaluated to one additional test:

  • T7 (Overcharge) – T7 itself is a simulation of a battery pack being left in a defective charger for a day, then removed and put in the shipping channel for a week. Unlike the T1-T5 sequence, the sample must be in an operational state for testing, specifically it must be able to accept charge. The overcharge voltage and current parameters are based on the unit’s charge specifications. Failures of this test are not common. When failures do occur, they will typically happen during the overcharge exposure and not during the seven-day wait period. Most common failures are improperly sized components in the battery pack’s protection circuitry that cannot handle the applied overvoltage. This results in current leaking past the protection circuitry, which leads to overcharge of the internal cells and possible thermal runaway.

Note that a separate group of battery packs is used for this test, although functional samples from T1-T5 can be reused if sample quantities are limited. The decision to reuse should not be taken lightly as latent damage from the T1-T5 sequence may result in failures during overcharge. This, in turn, may lengthen the required time in test and incur additional costs for retesting.

In summary, the new Amendment 1 implements a number of significant improvements to the standard that should benefit the industry with improved efficiency and standardization. The testing itself carries forward from the previous version of the standard and is still both electrically and, especially, mechanically intense.

For those new to needing to achieve compliance, having a full understanding of what test parameters will be applied to the design is critical. Additionally, some degree of pre-testing, either in-house or outsourced, may be beneficial to improve the chances of a successful first-time outcome. Finally, consideration of the impact of stresses being sequential during T1-T5 testing is also important. As always, your test lab has significant experience and should be both able and willing to offer general guidance to help you achieve your compliance goals.

This article is updated with the most current information available as of 2/1/20.

Related Articles

Digital Sponsors

Become a Sponsor

Discover new products, review technical whitepapers, read the latest compliance news, trending engineering news, and weekly recall alerts.

Get our email updates

What's New

- From Our Sponsors -

Sign up for the In Compliance Email Newsletter

Discover new products, review technical whitepapers, read the latest compliance news, trending engineering news, and weekly recall alerts.