Development of an Automated System for MIL-STD-461 EMI Testing

This paper discusses the development of an automated MIL-STD-461 EMI measurement system and some of the lessons learned in the process. The system hardware and software are described, and the tradeoffs associated with the various configuration and design choices are discussed. Problems occurring in the change from manual to automated EMI measurements are discussed, along with the solutions employed with this particular system. Development of a system to automate MILSTD-461 EMI testing at the Pacific Missile Test Center (PMTC) was begun in 1978, and has now reached the point where a discussion of the system should be of value to others involved or interested in the development of similar systems. The design and development of this system has taken place over a seven year period, and the changes in available hardware have been very dramatic. This continuing development of available technology, along with changes in test requirements and in the devices to be tested indicate that the measurement system design will never be complete, but that it will be regularly updated. The emissions portion of the system is operational and has been in use for several years, and the main emphasis now is on completing the development of the susceptibility section. This paper presents a description of the hardware design of the system, the software design, the calibration approach, operational useage, and some discussion of the tradeoffs involved in the various choices available in the system design. HARDWARE DESCRIPTION One of the first questions that the designer of an emissions measurement system must address is whether to use a set of EMI receivers or a spectrum analyzer. The advantages and drawbacks of both have been covered in several papers [1,2,3] and are by now familiar to most EMC engineers. The principal problems with spectrum analyzers are the lack of sensitivity and the likelihood of gain compression and distortion, sometimes characterized as a lack of dynamic range. These problems are avoided in EMI receivers by the use of fundamental mixing, preamplifiers, and tuned RP preselection as integral parts of the receiver design. At the time design was begun on the PMTC system the available technology in analyzers and receivers dictated the decision to use EMI receivers. An additional consideration was the availability of a set of receivers with a programmable interface allowing relatively straightforward automation. Since that time there have been major changes in spectrum analyzer design, and the choice is not as clear cut as it was. There is at least one analyzer available using fundamental mixing to improve its sensitivity, and there are preamplifiers available for most analyzers for this purpose. In order to avoid gain compression and distortion, tuneable preselectors and banks of filters can be employed. The drawbacks of this approach are that different sets of filters may be required depending on the characteristics of the device being tested and the specification being tested to. By the time preamplifiers and the appropriate filters and coaxial switching networks are installed the system complexity has increased significantly, and some of the advantages of using an analyzer in the first place, such as wide frequency spans in one sweep and fast scan rates, have been degraded. On the positive side, modern spectrum analyzers have many characteristics that make them admirably suited as general purpose EMI measurement instruments, particularly for development measurements in the design stage and for field use. The availability of synthesizer frequency accuracy, internal amplitude calibration, a wide variety of video and resolution bandwidths, digital trace processing functions, and ease of automation provide tremendous versatility in general purpose EMI measurements. At this time it appears that for performing automated MIL-STD-461 EMI tests the use of EMI receivers is still the best choice, but for general purpose use one of the new, compact spectrum analyzers would be optimum. The receiver system chosen at PMTC was the Singer/Ailtech NM series of four field intensity meters covering the frequency range of 20 Hz to 40 GHz. A complete system block diagram is shown in Figure 1. Most of these receivers were already in use at PMTC for manual EMI testing, and with the availability of the Singer CP-7 programmable controller they became an obvious choice on which to base an automated system. The first system consisted essentially of the receivers, the controller, and an analog X-Y plotter. The CP-7 was used for standalone control of the receivers and to acquire data from them. All antenna and sensor changes were done manually, as was correction of the data. The next step in automating the system was to include some sort of computer as an overall system controller. 482 U.S. Government work not protected by U.S. copyright 4 83 H P 1 50 T E R M IN A L V E R S A TE C 1 10 0A LI N E P R IN TE R H P 98 66 A P R IN TE R H P 27 48 B P A P E R T A P E R E A D E R H P 79 70 B M A G T A P E U N IT H P 79 00 A D IS C D R IV E C A L IB R A T IO N D A TA P M T C D /A C O N V E R TE R