New global automotive qualification standards for emerging LED applications (MAGAZINE)

CHRISTIAN JUNG, PHILIPP PLATHNER, JOE JABLONSKI, and JOACHIM REILL detail the global efforts of standards bodies that will impact auto makers as they increasingly turn to LEDs in forward-and rear-facing external lighting and to internal cabin applications.

LEDs in automotive applications are considered to have a longer life and a very low failure rate, leading to wider usage. Even under the harsh environmental conditions to which cars, trucks, and motorcycles are subjected, LEDs are expected to survive the coldest winters and the hottest summers, and tolerate the associated temperature, humidity, vibration, and other environmental impacts. Still, the industry needs ways to characterize and document the expected performance, and that need has led to some new global standards.

The ability of an LED to survive such extreme conditions depends on the materials and processes used to develop the LED, the know-how and experience of the LED manufacturer, and the skill of the developers in the auto space. Not all LEDs on the market are suitable for automotive applications. In order to differentiate automotive-qualified LEDs from all others, it is common practice to require automotive LEDs to pass a very specific set of tests before they can be designated for specific applications. The tests are designed to verify the outer boundaries of the product specification, but testing within the specification limits.

An LED is considered automotive qualified if it fulfils the requirements under the defined criteria, i.e., a "test-to-pass" approach. The testing of LEDs to these environmental conditions requires specialized equipment including thermal chambers (Fig. 1), making the procedure expensive and time consuming. The market has accepted that standardized test methods are needed to enable the LED manufacturers to optimize the test procedures and ensure customers can have confidence in the test results.

© Osram & Osram Opto Semiconductors GmbH; restricted use.

FIG. 1. A lab technician manages HTOL stress testing using a thermal chamber.

Semiconductor standards

In the past, LEDs were tested according to the standard established by the Automotive Electronics Council (AEC) Q 101 — "Stress Test Qualification for Automotive Grade Discrete Semiconductors." While this standard is appropriate for silicon-based semiconductors, it does not deal with the specifics of light-emitting, group-III-V compound semiconductors. To address this issue, an initiative was started in the International Electrotechnical Commission (IEC) to develop an international standard for the qualification testing of LEDs for automotive applications. The expanding functionality in automotive lighting applications, like adaptive driving beams in headlamps, increases the necessity for such standards. Even though AEC Q101 has recently been updated (from the C-version to the D-version), there are still significant differences between testing of LEDs and silicon semiconductor devices. The IEC standard addresses the specific issues that are relevant for LEDs.

At the same time, the United States Council for Automotive Research (USCAR) committee started work on a test procedure for the over-stress testing of LEDs, i.e., testing the LED beyond the specification limits to provoke failures — or a "test-to-failure" methodology. This over-stress testing is suitable to get information about the failure behavior of the product. Fig. 2 graphically depicts the test-to-pass and test-to-fail methodologies.

IEC 60810 qualification testing

The basis for the IEC work for defining the qualification tests of LEDs for automotive applications was the document published by the AEC — the previously referenced AEC Q 101. The Joint Electron Device Engineering Council (JEDEC) test standards referenced within AEC Q 101 were also used by the IEC committee for defining the test procedures.

The automotive lighting experts of the International Automotive Lighting and Light Signalling Expert Group (GTB), where all major global light source makers, set (or lamp) makers, and car makers are represented, prepared the first drafts for an IEC document. After several meetings it achieved general consensus on the proposed procedure.

© Osram & Osram Opto Semiconductors GmbH; restricted use.

FIG. 2. A schematic representation of a product specification compared with the related IEC qualification testing at the corner points and the USCAR over-stress testing illustrates the different testing methodologies.

With the agreement in the GTB expert group, the official IEC process was started. The document received great support by the IEC TC34A committee and was approved in the formal IEC committee voting stage. The LED qualification test will be published as a new part of the existing standard IEC 60810 "Lamps for Road Vehicles — Performance Requirements." It is expected to be published later this year.

The test details and failure criteria

Although many aspects could be taken over from the AEC Q101 testing experience for silicon semiconductors, many LED-specific definitions and criteria had to be defined at the IEC level. The most important testing requirements include: High temperature operating life (HTOL) test; Temperature cycling (TMCL) test; Wet high temperature operating life (WHTOL) test; Power temperature cycling (PTMCL) test; Electrostatic discharge, human body model (ESD-HBM) test; Electrostatic discharge, machine model (ESD-MM) test; Vibrations variable frequency (VVF) test; Mechanical shock (MS) test; Resistance to soldering heat (RSH-TTW) test; Resistance to soldering heat (RSH-reflow) test; Thermal shock (TMSK) test; Hydrogen sulfide (H2S) test; Pulsed operating life (PLT) test; Dew (DEW) test; and Flowing mixed gas corrosion (FMGC) test. Furthermore, the failure criteria have also been specified in the document.

During the tests, the change of the technical parameters is regularly monitored. For example, Osram recently took a sample set of typical test results for the luminous flux maintenance of 26 LEDs that are stressed under HTOL conditions. The results indicated that all tested LEDs show a luminous flux maintenance higher than 90% after 1,000 hours. Indeed, the bulk of the sample set tested above 95%.

Testing of LEDs requires highly specialized test equipment. Referenced previously, Fig. 1 shows an example of the inside of a test chamber where LEDs are stressed under HTOL conditions.

Matching tests with application requirements

Different automotive lighting applications may have different requirements. For example, an LED for a headlamp application close to the engine compartment will be exposed to very high temperatures, whereas an LED used for a rear turn indicator may not be exposed to such high temperatures.

Fig. 3 explains how an LED qualification according to IEC can help developers to choose an LED with the correct performance characteristics. The red and blue squares show a schematic specification range for two LED types. This specification is validated by IEC qualification tests at the corner points. The two star symbols represent the requirements (mission profiles) for two different automotive applications, A and B. It can be seen that LED type 1 is suitable only for application A, whereas LED type 2 is suitable for applications A and B.

© Osram & Osram Opto Semiconductors GmbH; restricted use.

FIG. 3. Two LED product specifications and two application mission profiles are compared schematically.

USCAR-33 over-stress testing

The USCAR Lighting Group, meanwhile, determined that there was a need to create a common set of requirements that focus on LEDs used in exterior automotive lighting. By publishing the SAE/USCAR-33 "Specification for Testing Automotive LED Modules," USCAR has created a standard to help the US automotive OEMs in the rapid adoption of LED lighting. This document is written to encompass testing a complete LED module, including any drive electronics and a section on LED component over-stress testing.

The idea behind the LED over-stress testing was to determine the failure limits of the LED device. Many of the tests are written to go beyond the manufacturer's specifications in both temperature (such as +15°C above maximum junction temperature or Tj) and forward current (130% of maximum) while the part is tested in various environments. The duration of each test is specified at 1,500 hours or when 50% or more of the test population has failed. These rigorous requirements work to ensure that all the components, as well as the complete module, are suitable for use in the harsh automotive environment.

Conclusion and next steps

The IEC 60810 qualification testing is intended to validate the specification range for a certain product. If the test conditions are chosen to represent the corner points of the specification range, and if all tests are passed, then it can be assumed with a certain confidence level that the LED is suitable for applications that lie within the product specification.

The USCAR-33 over-stress testing is intended to test the LED outside the specification, to get further information about the types and number of failures that can occur under these over-stress conditions. However, even if failures are recorded under the USCAR-33 over-stress test conditions, no direct conclusions can be drawn for failures in the field.

© Osram & Osram Opto Semiconductors GmbH; restricted use.

FIG. 4. The graph curves represent the robustness testing and projection of failures during realistic application conditions.

What is still missing is an international standard that defines a procedure for a systematic robustness investigation for LEDs. In such an approach, one would systematically test beyond the specification limits of a product and record the degradation as a function of over-stress. In combination with well-known modeling approaches like Arrhenius, Eyring, Peck, Norris-Landzberg, and Coffin-Manson, it would then be possible to project the degradation (or number of failures) during normal operation.

Fig. 4 shows a schematic representation of the test-to-failure-procedure during a systematic robustness investigation and the failure-rate projection using calculation models. Work has started in the relevant standardization groups to prepare a standard method for this type of robustness investigation for LEDs.

Regulations for higher-complexity LED lamps, like adaptive driving beam headlamps, are described in UN-ECE R123 (initially developed under the auspices of the United Nations Economic commission for Europe but meant to harmonize global developments). Such headlamps are being realized with electronically switched LEDs. Especially for those systems, a standardized robustness approach will help to enable broader and easier adoption.

DR. PHILIPP PLATHNER is a standardization officer, DR. CHRISTIAN JUNG is a senior staff engineer, JOE JABLONSKI is an application engineering manager, and JOACHIM REILL is a senior director of LED applications engineering at Osram.