Achieving Compliance to Small Format Battery Pack Standards

1510_F2_coverFrom my days as a young Marine Corps officer, it was stressed that “safety is paramount.” The same is very much the case when it comes to the design of small format secondary lithium battery packs. Failure in this regard risks personal injury and property damage, not to mention significant losses associated with negative publicity for those deemed responsible in the court of public opinion.

Ensuring a baseline of safety is managed through robust cell designs, proper design verification, system level risk assessments, and subsequent, independent regulatory compliance actions by third-party test labs to objective test standards. That said, all test standards and their component tests are not the same. The actions and efforts necessary to achieve compliance vary dramatically. In this article, we will offer a view from inside the test lab to provide a framework of risk assessment, not of the designs themselves, but rather of the relative difficulty of successfully demonstrating test standard compliance.

We’ll begin with a quick discussion of prominent battery pack standards. We’re not attempting to cover all of them, but rather focus on those that make up the majority of our test volume. Then, we’ll look at the individual component tests, both their primary specifications as well as the problems seen with achieving compliance. This will be translated into a “stop-light” coding tied to relative difficulty of passing the requirements (red-high risk; yellow-medium risk; green-low risk). Wrapping up, we’ll cover some general ideas on mitigation to improve the odds of first time success and also touch on expected upcoming changes to the standards.

IEC 62133

For small format rechargeable lithium battery applications, there are three standards that enjoy the widest usage for consumer and commercial applications. IEC 62133:2012 (2nd Edition) – Rechargeable cell/battery safety, is the de facto standard for international compliance. Mandated by many IEC end-device standards, it has also served as the basis for many country-specific battery test standards. It offers both mechanical and non-faulted electrical tests with charging preparation done at temperature extremes. (Note that compliance to UN 38.3 transportation testing is an integral requirement.)

The actual test list for secondary battery packs was dramatically reduced with the release of the second edition of IEC 62133. This was primarily due to the removal of all tests that were included within the scope of UN 38.3 testing (compliance to UN requirements must be provided, but the testing does not have to be repeated). This leaves only four actual tests, as illustrated in the table below.

Test Name

Parameters/Notes

Failure Criteria

8.2.2

Molded Case Stress

70°C / 7 hours.

No components exposed

8.3.2

External Short Circuit

55°C / 1 hour if circuit trips.

No Fire / No Explosion

8.3.3

Free Fall

1m / 3 drops / Room Temp

No Fire / No Explosion

8.3.6

Overcharging of Battery

2C current at max voltage of charger or if not available, 5V per series cell. Run until temp stabilizes.

No Fire / No Explosion

Tests under IEC 62133: 2012 (2nd Edition)

Mechanically, the molded case stress test is run at a temperature just above the typical electrical operating range of most battery packs and well within most plastic specs used for such applications. The drop test is one meter to concrete and does not involve any cold temperature pre-conditioning regardless of the specified lower operating temp for the battery. Electrically, the testing is run on un-faulted samples, so this regime is truly a test of the safety circuit as-is. If there is sufficient design margin to over-voltage and the safety circuit promptly shuts off when a short is detected, the risk of event is small.

Summarizing, if a pack design incorporates reasonable robustness to limited mechanical and electrical stress, the overall risk of non-compliance to IEC 62133 2nd Edition is relatively low.

IEC 62133:2012 (2nd Edition)

Molded Case Stress

External Short Circuit

Free Fall

Overcharging of Battery

“Stop light” coding for tests under IEC 62133

What about the future? IEC 62133 is scheduled for a major overhaul to be published in August 2016. At that time, the standard will split into two new standards. IEC 62133-1 (1st Edition) will be specific to nickel chemistries. IEC 62133-2 (1st Edition) will cover lithium. The latter is expected to introduce faulting as a recommended practice, but at this time it is not thought that it will become a mandatory requirement. Additionally, it is expected that some of the mechanical testing previously removed will re-enter the picture. Specifically the vibration and shock requirements from UN 38.3 will become requirements in the standard. The following section includes assessments of their relative risk.

UN 38.3

Want to ship a lithium battery almost anywhere in the world by air, vessel, rail, or truck? Plan on complying with the requirements of UN 38.3 – Recommendations on the Transport of Dangerous Goods; Manual of Test and Criteria, Fifth revised edition, Amendment 1. It is commonly known as “T1-T8” testing based on the test reference numbers and sequence. Found in many countries’ shipment of dangerous goods regulations, it presents a combination of significant environmental, mechanical, and electrical stresses, in many cases well above typical lithium battery specifications.

Although the UN test profile does not involve any faulting, it does add the dimension of sequential stresses. This is where the same group of samples is put through a series of tests (tests T.1 through T.5). For packs, the testing is two-prong, as illustrated in the table below.

Test Name

Parameters/Notes

Failure Criteria

T.1

Altitude

50,000 feet for 6 hours

Leakage, venting, disassembly, fire ,
voltage loss > 10%

T.2

Thermal

+72°C/-40°C for 130 hours

T.3

Vibration

7-200 Hz up to 8 Gn

3 planes x 3 hours each

T.4

Shock

Half sine, 150 Gn, 6 ms

6 planes x 3 impacts each

T.5

Short

80 mљ at 55°C

Disassembly, rupture, fire, temp > 170°C

T.7

Overcharge

1.2X-2X Max Charge Voltage

2X Max Charge Current

24 hours

Disassembly or fire within 7 days of the test.

Testing under UN 38.3

The altitude test is generally innocuous. In over a decade of testing, we’ve only seen one product fail. The same cannot be said for the other tests. Thermal testing is conducted above the limit of most lithium ion cells. In addition to aging the cells, it also induces thermal fatigue potentially weakening connections. Vibration is intense. I tell our clients that their battery will experience vibration above that expected of many military grade products. Not only is the profile itself stressful, but it is run for a full three hours in each of the three cardinal planes. Shock is also quite significant with six times the number of impacts used by other similar standards. Finishing up the regime, short circuit is conducted outside the thermal operating range of most packs.

For the T1-T5 sequence, although the individual tests are tough, it’s the cumulative effect that is the concern. The combination of thermal, vibration, and shock represent a very honest attempt to damage the pack. Assuming the pack survives that beating, it is then short circuited at high temperature to ensure that the cells are still protected. Again, this is a very mechanically intense profile.

Overcharge is run on a separate group of samples and is designed to simulate a pack being improperly charged for an extended duration and then presented for shipment where it sits in channel for a week. For a successful outcome, the safety circuit must function properly. Pitfalls are if the circuit does not fully shut off (allows a small current to flow) or oscillates off and on to a point where overcharging still occurs. In our experience, the failures generally occur during the charging phase, not during the seven day wait period.

Summarizing the risk by test, vibration and shock represent the most significant source of test non-compliances due to their intensity and the effects of cumulative stress. Thermal is included as a medium concern as it sometimes induces failures itself, or contributes to the failures that occur during vibration or shock. Short circuit being a final check of extended safety circuit performance is in a special category. Although we don’t see as many failures in short, the fact that it is a final evaluation warrants concern to ensure electrical functionality after stress. Finally overcharge performance is also a medium risk tied to safety circuit robustness to the combination of high voltage and high current.

Like IEC 62133, the UN requirements are also subject to changes although the pace of major changes is relatively infrequent. As such, current UN working group proposals are not likely to become hard requirements until 2017 or beyond. The present tenor of these activities is better characterized as clarifications of the current requirements and adjustments for new technology as opposed to sweeping changes in the testing scope or approach.

UN 38.3

Altitude

Thermal

Vibration

Shock

Short

Overcharge

Table 4: “Stop light” coding for tests under UN 38.3

UL 2054

Compliance with the requirements of UL 2054 – Household and Commercial Batteries (Second Edition) is mandated by a number of U.S. end device standards. By all accounts, it is a difficult standard involving roughly double the number of tests found in the UN or IEC requirements. Most notably, UL 2054 introduces single faults of the electrical circuit into the scope. Compounding this effect, it also requires operation just below the trip point of the faulted circuit to attain independent certification, thus simulating a true worst-case scenario.

Before looking into the individual test requirements, a discussion on faulting is in order. Exactly what faults will be applied to the battery packs prior to stress? Below are four common faults employed in the majority of projects we execute:

  • The source and drain of the charge or discharge FET are shorted together by means of a wire jumper.
  • The charge or discharge FET gate is tied to either Cell+ or Cell- to prevent the FET from stopping the flow of current.
  • The sense resistor is shorted.
  • The FET control output of the safety IC (charge and discharge) are shorted together.

Note: If the components are either too small to fault or are otherwise inaccessible, the standard practice is to fault out the entire board.

Again, only a single fault is applied to a given device. Multiple faults are never specified, although it is very possible that the combination of a single fault and the applied stress may result in a cascading failure that impacts other components on the board and their associated functionality.

Looking at the individual electrical tests, short circuit, abnormal charge, and component and surface temperature are run at the experimentally determined maximum no-trip current for each fault mode. Abnormal charge is a high current test where the maximum charge voltage is not exceeded. Abusive overcharge is an overvoltage test where the current is high but limited. Specifically the test is run at a minimum of 6V per series cell. As lithium ion devices are very sensitive to overvoltage, this is a particularly difficult test where failures are not uncommon. Forced discharge is run only on series connected cells (two in series or higher). All but one series cell is fully charged. The remaining cell is fully discharged. The pack is then hard shorted resulting in significant electrical stress on the fully discharged cell. For a pack to be classified as a limited power source, it must deliver < 8 amps and < 100 watts after a specified duration of time (the duration is based on the pack’s design). Note that there are a multitude of test cases for limited power source. The two noted above are the most common.

Test Name

Parameters/Notes

Failure Criteria

9

Short Circuit (RT)

Faulted, max current

Fire, explosion, leakage, cell temp > 150°C

9

Short Circuit (55C)

10

Abnormal Charge

Fire, explosion, leakage

11

Abusive Overcharge

Faulted.

Fire, explosion

12

Forced-Discharge

13

Limited Power Source

See below

13A

Component Temp

Max current.

13B

Surface Temp

19

250N Steady Force

—–

Venting, leakage

20

Mold Stress Relief

Temp varies

Protective devices exposed

21

Drop Impact

Possible cold pre-soak

Fire, explosion, protective devices exposed

Individual tests under UL 2054

The component temperature test ensures that the thermal specifications of critical components are not exceeded when the battery is operated at its maximum ambient temperature and highest current. The surface temperature test is typically run at the same time and ensures that the outside surface of the battery does not present a touch hazard to the user.

Mechanical testing introduces a few new aspects. If a pack’s minimum rated ambient temperature is less than 0°C, then the pack must be soaked for three hours at that temperature immediately before drop testing. In most cases, this leaves the plastic at its most brittle state. Mold stress relief is not run at a fixed temperature, but rather one derived from the component temperature test. In practice, test temperatures greater than 90°C are not uncommon. The 250N steady force test involves applying a force of 250N via a 30mm diameter metal disk on all sides of the product.

The relative risk by test in this regime is highly focused on the electrical tests. With the inclusion of single faults and worst-case operation, these tests require extra attention during the design process to ensure a positive outcome. By far, the abusive overcharge test is the most significant given the overvoltage conditions applied to the faulted pack. Two levels of protection against overvoltage are suggested as one will be removed via faulting. Abnormal charge, forced discharge, and both short circuit tests also involve significant test risk due to faulting and worst case operation.

A key to success is knowing in advance how the design will behave under these conditions and how much margin is available. The limited power source test represents challenges for bigger packs. If the pack’s normal voltage and current put it close to the limit for LPS compliance, the addition of faults will only reduce the available margin. Knowing this up front and communicating this to the end device manufacturer is important (this means that some of the pack’s requirements may have to be accounted for in the end device). Mechanically, mold stress (due to its potentially high temps), drop (due to possible brittleness from cold temperature soaking), and component temp (due to component thermal limits) involve medium risk. Surface temp and 250N steady force are generally benign for minimally robust designs.

The future for UL 2054 is currently unclear. UL has released the first edition of UL 62133, which is fully harmonized with IEC 62133, 2nd Edition. The two standards essentially compete for the same test space although their requirements are very different. The complexities associated with the implementation of UL 62133 are still unfolding, but clearly they will have an impact on the future role of UL 2054 as a key U.S. compliance standard. It is expected that UL will follow the IEC when they split IEC 62133 into nickel (IEC 62133-1) and lithium-specific (IEC 62133-2) standards in August of 2016, but the subject of adding unique national deviations for the U.S. has not been settled.

UL2054

9 Short Circuit (RT)

13A Component Temp

9 Short Circuit (55C)

13B Surface Temp

10 Abnormal Charge

19 250N Steady Force

11 Abusive Overcharge

20 Mold Stress Relief

12 Forced-Discharge

21 Drop Impact

13 Limited Power Source

“Stop light” coding for tests under UL 2054

Summary

Today’s modern battery test standards and their component tests run the gambit from relatively benign to significantly difficult. Knowing the relative risk of each test can help with the allocation of limited resources during the design phase to maximize the chance of a successful regulatory outcome. An outline for compliance should include:

  • Understand early what test standards will be applicable to your product.
  • Keep in mind that standards are constantly evolving and requirements do vary by country.
  • Evaluate each component test by relative risk to help allocate engineering resources.
  • Thoroughly investigate the product’s behavior at the expected test conditions. Include the aspects of serial testing, faulting, preconditioning (aging, etc.), and worst-case operation if applicable.
  • Develop and implement mitigation strategies for the design.
  • Consider independent pre-testing early in the design process to reduce future schedule risk.
  • Talk to your regulatory test provider early. They are an extra set of skilled eyes with a wealth of knowledge about testing and the associated regulatory environment.

author_copeland-johnJohn C. Copeland is co-owner and technical manager for Energy Assurance LLC, an independent, fully-accredited cell and battery test laboratory. His career has included various positions in quality engineering, reliability engineering, failure analysis, project management, supplier assessment, and quality management in the electronics and portable energy sectors.

About The Author

John Copeland

John C. Copeland is co-owner and technical manager for Energy Assurance LLC, an independent, fully-accredited cell and battery test laboratory. His career has included various positions in quality engineering, reliability engineering, failure analysis, project management, supplier assessment, and quality management in the electronics and portable energy sectors. John holds a BSEE from Auburn University, MSQA from Southern Polytechnic State University, and is trained as a Six-Sigma Black Belt.

Related Posts

Leave a Reply

Your email address will not be published.

X