I’m working on a new system that does everything for everyone. I’ll take one of those, and one of those, and two of these. Do you have it from some other supplier in blue? Our sales/marketing department wants to offer it in blue. OK, wrap em up!
Wouldn’t it be great if designing new military systems with commercial-off-the-shelf (COTS) hardware were this easy? We could speed up our product time to market and all but eliminate development costs. It’s possible that our profit margin will be skimpy, and our customers might not think we were the most innovative company since our products are limited to existing components.
But one thing is certain: Our reliability would be good because of our use of proven hardware, and we could save testing costs because there would be no need to perform any expensive, time-consuming tests like EMC. After all, combining multiple EMC-compliant subsystems results in an EMC-compliant system. Right? The answer is maybe.
When it comes to electromagnetic compatibility, EMC + EMC = EMC does not necessarily hold. Consequently, there is a need to understand the fundamentals of systems EMC design and analysis which are quite different from those used for equipment.
Systems Design Process
The systems design process is loosely divided into five elements:
• Need Analysis—Understanding the problem or operational need that is to be addressed by the system, such as what it is supposed to do.
• Definition—Defining the requirements for the individual building blocks of the system.
• Design—Determining what kinds of equipment are needed to perform the operational tasks and if they can be purchased or do they have to be custom designed and built.
• Development and Test—Combining the equipment into a functional system and performing the tests needed to validate operation.
• Operation—Using the system for its intended function. Since systems evolve, operation can be considered a real-time evaluation of the system.
Inter- and intra-systems EMC must be considered from the very beginning. EMC permeates through and must be considered during each of the elements of the systems design.
Since most systems have some form of analog interface to detect and measure low-level analog signals and some form of higher-level digital or RF signal processing, it is logical that even though each equipment item is EMC qualified, the system may have a susceptibility problem.
If the system is composed of unintentionally radiating RF devices that are being installed into a benign RF environment, EMC hardening simply may consist of a good ground and some careful cable routing and box shielding. It’s different if the system is composed of multiple RF receivers and transmitters installed in a small, hostile RF environment along with similar types of equipment.
EMC hardening then may require frequency assignment, signal amplitude controls, and careful placement and orientation of each individual component of the system along with operational scheduling based on the priority of the system. And there still may be problems.
The fact that each equipment or subsystem item may already meet some high-level EMC specification such as MIL-STD-461F helps it to function in a hostile RF environment. But that does not guarantee that the system won’t have an EMC problem.
The DoD, which has been fighting this battle since the beginning of shipboard and airborne radio in the early 1900s, has developed systems-level EMC specifications to aid in controlling the problem. Since that time, many systems-level specs have come and gone.
The current systems specification is MIL-STD-464A. This spec and its predecessor MIL-E-6051 rely heavily on individual equipment and subsystems meeting MIL-STD-461. That’s a step in the right direction as long as the manufacturers continue to upgrade their equipment to meet the current version of MIL-STD-461.
Supplier and user groups have done a reasonable job of keeping MIL-STD-461 aligned with the actual RF environment, and it’s a good thing. MIL-STD-461 spans 41 years, and during that time, all the new electronics equipment has created a completely new RF environment.
The primary electromagnetic environmental concerns are radiated and conducted emissions and susceptibility. Figure 1 illustrates the EMC issues faced by a system’s component. For a one-of-a-kind system at a fixed installation, you might perform a site RF environment measurement. It still will be necessary to periodically evaluate the environment because of future new equipment.
There are equipment classes such as low-level analog or RF receivers that are extremely susceptible to small RF ambient changes. With a simple mobile platform, the ambient RF environment changes from city to city, runway to runway, or port to port, and it may be much more cost-effective to simply call out a spec based on a generalized RF environment such as MIL-STD-464.
In a complex platform where multiple equipment and subsystem items are installed, the platform-generated environment may greatly exceed that of the surrounding nonplatform ambient emitters. An example is the 27,000-V/m E-field called out in MIL-STD-464A for shipboard radar systems.
The extent of the effect that the RF environment has on the system is determined by four interrelated factors abbreviated by the acronym FAST: frequency, amplitude, spatial separation, and timing. FAST also represents a quick culling analysis technique based on these four factors.
Since any element of the system will be susceptible at some frequency and amplitude, the first concern is the overlap of culprit and victim frequencies, especially with intentional RF devices such as transmitters and receivers. If a culprit transmitter is operating at the same frequency as a victim receiver, then interference is likely. After all, the receiver may be able to respond to 0.5 µV. However, there’s more to it than that. It could be one of the many transmitter/receiver intermodulation frequency pairs mixing in a
Even with frequency alignments, if the amplitude of the culprit interference (I) picked up by the victim is so low that the victim’s signal-to-noise ratio (S/N + I) is not degraded, then there is not likely to be a systems interference problem. As the culprit amplitude increases to the point where the interference pickup is approximately 10% or more of the noise, then interference begins to show up. As the level continues to increase, eventually the culprit level will interfere because of circuit overload, and frequency alignment won’t be required for interference to occur.
The spatial relationship of the culprit and victim RF coupling process also affects the amplitude of the culprit interference that reaches the victim. Increasing physical separation distances between antennas, cables, and boxes; changing the lengths and relative orientations of cables and harnesses; or choosing directional vs. omnidirectional antennas will reduce the amplitude of the culprit interference.
That said, physical separation on a mobile platform is limited by the size of the platform and the number of collocated emitters and receptors. Every time new equipment is added, the system has to be evaluated with respect to this new item. The smaller the system, the more stringent the requirements need to be.
Timing is an often-overlooked parameter. Sometimes it is nearly impossible to solve the interference problem except by turning off either the culprit or the victim. This typically is done by an operating schedule so that we use one without the other. Timing can be applied in other ways as well.
The FAST Test Procedure
• Identify all emitters and receptors and determine the scenarios for simultaneous operation. The platform operation may dictate what equipment is running at the same time. Systems test should represent actual operational modes. Some devices can be both an emitter and a receptor but not usually at the same time. Record the information on Figure 2.
• Perform a visual check on all emitters and receptors for bad bonds, corrosion, painted RF gaskets, cable problems, broken antennas, missing screws/covers, and any other nonstandard conditions.
• Turn on the receptors one at a time and set to the most sensitive normal operating mode. If the device is a receiver, tune in a weak signal from a source that is NOT one of the test emitters.
• As the receptors are turned on, observe ALL of them for any interference. On very rare occasions, the observed interference that occurs during the receptor tune procedure may be receptor-to-receptor. If the interference does not respond to the turn on/off sequencing of the emitters, look at the receptors. Record the receptor-to-receptor pair on Figure 2.
• Turn on the emitters one at a time. If it is a transmitter, use 100% worst-case normal modulation.
• Observe all receptors for interference. Record the emitter-to-receptor pair on Figure 2.
• When all the emitters are on, tune each receptor through all the bands, modes, and ranges while observing for interference.
• Record any problems, identify emitter-to-receptor pairs by turning off emitters one at a time and attempt to determine corrective action. It’s common to find that the outputs of two or more emitters are mixing in a nonlinear junction, and the resultant is the culprit. Consequently, receptor interference will stop whenever either emitter is turned off. This requires all emitters to be checked each time and identified as joint culprits.
Systems EMC design and test can be simplified by identifying potential sources of EMI prior to the start by using the FAST analysis technique. FAST is a simplified and organized approach to processing and comparing pertinent information about the equipment items. All platform equipment is categorized as emitters or receptors based on their function and recorded for further processing.
Figure 2 is a sample matrix for listing the equipment. The amplitude or sensitivity vs. frequency is required for each item. This includes both tunable and nontunable devices. Antenna gains, tuning ranges, and bandwidths are required. If the item’s operation is time dependent, then time sequencing will have to be specified as well.
Much of this information typically can be found in the manufacturer’s literature, which saves making measurements. For large platforms such as a ship or large aircraft, relative position information may be required for the analysis to determine path loss and blockage.
To check for the interaction at the fundamental frequency, each emitter-tuned operating frequency range is compared to each receptor-tuned operating frequency range. Any overlap represents a worst-case problem at the transmitter/receiver fundamental for that emitter-receptor pair.
A fundamental frequency alignment occurs when the receiver and transmitter are tuned to the same frequency, such as the difference in frequency is less than (BWT + BWR)/2. If the receiver BWR is greater than the transmitter BWT, all transmit energy that reaches the receiver will be available as an in-band EMI signal.
The amplitude of this EMI is compared with the intentional received signal amplitude and the out-of-band susceptibility characteristics of the receiver to determine if a susceptibility condition exists. The modulation characteristics for both the transmitter and receiver must be considered when determining potential susceptibilities.
After evaluating the equipment at the fundamental frequencies, the range is expanded to include frequencies from 0.1 to 10 times the operational frequencies to evaluate the harmonics and spurs.
When all transmitters and receivers have been compared with each other at the fundamental, harmonics, and spurs, they are compared with the composite worst-case environment to determine compatibility with the RF environment. Most military systems specifications require a 6-dB margin or up to 20 dB for ordnance.
Once the FAST analysis is completed, it will be possible to define the level and type of EMC hardening required. Since systems often are configured from existing hardware, hardening generally is limited to techniques that can be applied without making any hardware changes. The techniques used will depend on the RF signal coupling.
Part 2 of this article focuses on EMI control techniques needed when analyzing the design of EMC systems, specifically grounding, bonding, shielding, filtering, and cable/wiring design. It will appear in EE’s November issue.
About the Author
Ron Brewer currently is a senior EMC/RF engineering analyst with Analex at the NASA Kennedy Space Center. The NARTE-certified EMC/ESD engineer has worked full-time in the EMC field for more than 30 years. Mr. Brewer was named Distinguished Lecturer by the IEEE EMC Society and has taught more than 385 EMC technical short-courses in 29 countries and published numerous papers on EMC/ESD and shielding design. He completed undergraduate and graduate work in engineering science and physics at the University of Michigan. e-mail: [email protected]