Conformity Magazine - November 2007 - (Page 40)

Reverberation Chamber Testing Anechoic testing is unlike most real-life radiated EM environments, so reverberation chamber methods have been developed to give more confidence (see [14] [15] [16]). Unlike anechoic chambers, their results can be correlated mathematically with the reflectivity of the operational EM environment. Reverberation chambers and their RF power amplifiers cost a great deal less than anechoic chambers, and thorough testing can take less time than in anechoic chambers because there is no need to test with many angles, or with vertical and horizontal antenna polarizations. A “reverberation chamber” test method currently used for some safety-critical systems rotates the chamber’s “stirrer” or “paddlewheel” over a full revolution using between 20 and 120 angular steps. At each step of the paddlewheel, radio fields are generated in the chamber, comparable in frequency range and magnitude with the foreseeable worst-case EM environment(s). The frequency range is covered in small steps (e.g. 0.1%). At each frequency step, the field is modulated with the appropriate “CW, chirp plus OFF/ON pulse”, or other modulations, at a rate that is slow enough to be sure to detect any errors or malfunctions in the functions being monitored. Where equipment is too large, or frequencies too low, or when testing on-site with no transmitting license, conducted coupling test techniques may be able to replace radiated methods. But conducted testing is not a true alternative to radiated testing, so it may be more realistic to use striplines, TEM cells, Helmholtz coils, or other test methods. Test Levels and Uncertainty In the context of this article, SILs are measures of confidence that, when the EM or physical threats occur, they will not result in harm. So SILS are related to the design techniques used, and to the confidence achieved by the validation and verifications techniques employed. To achieve a given level of confidence in EM and/or physical immunity testing, the EM and physical threat specifications will need to be higher than the environmental threats by a “test margin” that takes care of the various uncertainties. There are uncertainties in the: • Specifications of lifetime EM and physical threats; • The stresses actually applied during immunity tests; • Performance of individual units (e.g. due to component tolerances, variations in assembly and installation, etc.). For example, MIL-STD-464 adds a 6dB test margin for safety-critical and mission-critical equipment, and a 16.5dB margin for ordnance. Figure 3 shows that, when an immunity test is performed exactly at the specified threat level (for example, a test at 10V/m to simulate an RF field in the operational environment of up to 10V/m), there is only a 50% chance that the test was actually carried out at or above the desired level. There are standard methods for adding together various types of uncertainty, taking their type of statistical distribution into account (see, for example, [17]). Assuming a Normal (Gaussian) distribution (for example) in Figure 3, increasing the test level by one standard deviation improves the confidence that the test level reached/exceeded the specification to 68%, increasing by three standard deviations improves it to 99.7%, and four standard deviations achieves 99.99%. The level of confidence achieved by the testing should generally be at least of the same order as the required SIL. For example, SIL 3 represents a probability of dangerous failure of a safety function “on demand” or “in a year” of between 0.01% and 0.1%, which is comparable with the 99.99% confidence given by testing with a level that is four standard deviations above the specified level. To avoid testing at very high levels, with its attendant risks of over-design and unnecessarily high costs, it is important to use test methods and quality control that achieve low standard deviations. Where the final safety system employs “zones” protected by EM mitigation measures (e.g., filtering, shielding, transient suppression, etc.), some or all of the EMC tests applied to the equipment in the protected zones may not need Figure 2: Example of a reverberation chamber 40 Conformity noVEmBEr 2007

Table of Contents Feed for the Digital Edition of Conformity Magazine - November 2007

Conformity - November 2007 Contents Editor’s Note Commission Sets Date, Terms for 700 MHz Auction FCC Reaffirms Roaming Obligations Commission Moves to Improve USF Management FCC Proposes Fines for Violations of E911 Handset Requirements Commission Levies $1 Million Plus Fines for Unsolicited Fax Advertisements GR-3150 - Lithium Battery Certification Levels Based on Criteria for General Product, Safety, and Performance, Part 2 ESD Open Forum Machinery Field Evaluation for the United States Validation, Verification, and Immunity Testing Techniques for EMC for Functional Safety Thermal/EMC Simulation Addresses New Mechanical Design Challenges Product Safety Equipment Buyer’s Guide FCC Releases Report on Telephone Numbering Resource Utilization Canada Still Has PCS Spectrum to Assign Updated Standards List for the EU’s Medical Devices Directive Updated Standards List for EU Directive on Active Implantable Medical Devices Call for Papers for the 2008 EOS/ESD Symposium Announced Mobile Phone Use in Hospitals May Be Dangerous CPSC Actions in the News IEC Standards Update UL Standards Update Telcordia Standards Update From Our “You Can’t Make This Stuff Up” Department Product Reviews Looking Back: Items from Past Issues of Conformity Upcoming Events Advertisers

Conformity Magazine - November 2007

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.