Rapid deployment of high-speed, high-traffic fiber-optic networks is fueling the global economy. Bandwidth-hungry Internet, web-hosting, and e-commerce applications continue to force network growth. The demand for optical components far outpaces supply, and yields remain low. The mantra of further, faster, and more reliable is chanted throughout the industry.
A growing demand for reliability assurance testing accompanies this push to bring new technologies to market with confidence. Reliability of these new technologies is key as businesses and consumers expect uninterrupted service, data integrity, and fault-free operation.
Some of the optical-networking products currently being tested include long-haul dense wavelength division multiplexing (DWDM) gear, synchronous optical network (SONET) equipment, optical switches, lasers, transceivers, filters, and splitters.
Reliability Assurance
The critical nature of many fiber-optic branching components and the rapid evolution of designs and manufacturing practices facilitate the need for a reliability assurance program. Environmental reliability life tests help determine if the basic device design, fabrication materials, and processes are sound and can be expected to provide adequate long-term reliability. Testing to Telcordia (formerly Bellcore) standards is performed to characterize or qualify the reliability of fiber-optic materials and components.
Telcordia generic requirements play a major role in shaping the way the optical-networking community tests its products. Documents such as GR-468-CORE, GR-326-CORE, GR-418-CORE, GR-1209-CORE, and GR-1221-CORE assist manufacturers of optical components in demonstrating reliability conformance. Compliance with these reliability assurance criteria supplies the confidence that service providers and end users need to ensure the products they are buying will have a useful life expectancy in excess of 20 years.
Reliability test procedures common to many of the Telcordia specs include high- and low-temperature storage tests, temperature cycling, damp heat, cyclic moisture resistance, and thermal shock. Many of the test conditions within the Telcordia requirements are derived from well-accepted industrial and military specifications.
Reference is made throughout the generic requirements to EIA/TIA-455, MIL-STD-883, MIL-STD-202, and IEC 68. Although many of the Telcordia reliability test procedures have been adopted from industry-standard procedures, details regarding test conditions, durations, and sample sizes suit specific telecommunications-related applications.
Endurance Testing
• Accelerated Aging Tests commonly are performed at temperatures within the range of +70°C to +85°C for 2,000 h to 5,000 h. Actual specifications are based on whether products will be used in a central office (CO) or an uncontrolled remote terminal.
Typically, the remote-terminal conditions are more extreme than those encountered in a CO installation. As an example, 25 laser-module test samples designed for use in an uncontrolled, remote-terminal environment must be subjected to +85°C for 2,000 h with zero failures allowed.
For added confidence, device and component manufacturers often test for periods longer than those required. A 2,000-h test commonly will be stretched to 5,000 h to gain more information about the expected life and performance of the product.
• High- and Low-Temperature Storage Testing is performed at maximum storage temperature conditions. This testing characteristically runs for 2,000 h to 5,000 h at a high temperature of +85°C and a low temperature of -40°C.
A sample size of 11 passive optical components such as wavelength division multiplexers (WDMs) will be tested at minimum and maximum storage temperatures, typically +85°C and -40°C respectively, for a period of 2,000 h. In situ functional testing of the components is performed at specified increments during the environmental storage test.
• Temperature Cycling Tests are run routinely between the maximum and minimum storage temperatures. High temperatures range from +85°C to +70°C, and the low temperature is consistently -40°C. Temperature change rates for these cycling tests can be as high as 10°C/min.
The number of thermal cycles varies from 100 for CO environments to 500 for remote-terminal environments. To meet information-gathering objectives, the number of thermal cycles increases from 100 to 500 and from 500 to 1,000, respectively.
Generally in the case of a photodiode module used for installations in a CO, 11 samples are exposed to 100 temperature cycles between -40°C and +70°C. The minimum temperature ramp rate shall be 10°C/min.
• The Cyclic Moisture Resistance Test replicates the stresses of condensation, freezing, and thawing that can occur in field-use conditions (Figure 11, see the April 2001 issue of Evaluation Engineering). Relative humidity (RH) conditions are controlled from 85% RH to 95% RH at +75°C and then uncontrolled as the temperatures are cycled to lower set-point conditions of +25°C and -40°C.
Dwell times at the extremes vary between 3 h and 16 h. Depending on the end-use environment and the type of component, the number of cycles can vary between five complete cycles with five subzero cycles and 20 complete cycles with 10 subzero cycles.
Optical filters deployed in remote-terminal environments, for example, must undergo five complete accelerated stress test cycles of the temperature-humidity profile illustrated in Figure 1. During this test, the RH is uncontrolled at +25°C and -40°C.
Additional mechanical and physical testing are performed in the areas of mechanical shock, sine vibration, thermal shock, and other strength-of-materials analyses to confirm the integrity of the component or device.
Production Test
Many companies in the fiber-optics industry use highly labor-intensive manufacturing processes. This can introduce workmanship variables that influence product reliability. Production screening and burn-in frequently weed out defective products destined for premature failure.
System-level burn-in can aid in identifying weak devices (infant mortality) and might highlight marginal circuit designs. Elevated temperatures also can exaggerate marginal speed or parametric situations to the point where they cause failures.
Stress screening of fiber-optic components typically involves rapid thermal cycling. Temperatures are slightly derated from those used in reliability assurance testing. For example, screening temperatures will range from 70°C to -10°C while reliability assurance test conditions can go from 85°C to -40°C.
Testing Challenges
Due to the long-term nature of these test requirements, it is essential that environmental test chambers have adequate insulation to minimize thermal losses, proper thermal and moisture seals at the access points, random-in/random-out sample loading and unloading, reliable equipment and software, and a dependable backup power source.
Full access to one or both sides of the chamber accommodates product-to-test equipment connection for in situ functional product testing and diagnostics. This configuration also permits samples to be introduced and removed from the workspace in a random fashion without upsetting long-term environmentally controlled conditions (Figure 2).
Fixturing and associated loading and unloading of optical components must take into account the fragile nature of the fiber-optic cabling. Characteristics of good fixturing include well-planned cable management, minimization of setup time and additional handling steps, accommodations for required product sample sizes, lightweight yet rigid construction, and ample airflow circulation. Fixturing must withstand repeated use but not represent a large thermal mass that impedes chamber performance.
Measurements of key optical performance characteristics such as optical attenuation, modal dispersion, insertion and return loss, and back reflection must not be adversely affected as a result of improper fiber-optic cable management techniques.
Most optical-networking reliability assurance stations include automated, functional test equipment consisting of a laser or light source, a polarizer, a controller, routing switches, and a power meter. This equipment monitors and measures the performance of the optical components that are being endurance tested. Integration and automation of the complete test facility are vital to the reliability and repeatability of the test results.
Moisture must not migrate into the chamber during cold test conditions or out of the chamber during long-term high-humidity tests. During long-term assurance reliability testing, conditions in the workspace should remain noncondensing.
For moisture-sensitive products, a source of dry/compressed air or gaseous nitrogen must be provided to maintain a positive pressure inside the environmental test chamber and to limit condensation. The integrity of seals around access ports, fiber management trays, and fiber-optic cables is vital to successful long-term testing.
Evolving Needs
By performing these endurance tests, the optical-networking industry can prove the reliability of components used to build the fiber network. As technologies and materials change, refinements will be made to these test procedures, and new methods will evolve. Equipment used to perform these tests continually will be enhanced to meet the needs of this industry segment.
Reference
- GR-1221-CORE, Generic Reliability Assurance Requirements for Passive Optical Components, Bellcore, Issue 2, January 1999.
About the Author
Kevin Ewing is a market manager at Thermotron Industries. He has 17 years of experience in the application of environmental test equipment and holds a B.S. in mechanical engineering from Michigan Technological University. Thermotron Industries, 291 Kollen Park Dr., Holland, MI 49423, 616-393-4580.
Return to EE Home Page
Published by EE-Evaluation Engineering
All contents © 2001 Nelson Publishing Inc.
No reprint, distribution, or reuse in any medium is permitted
without the express written consent of the publisher.
April 2001