by Lorenzo Cividino and Paul Kingsepp, SL POWER ELECTRONICS CORP
An awareness of standards for conducted and radiated emissions helps in fielding power supplies that don’t cause havoc in attached electronics.
Innovations such as the Internet of Things (IoT), 5G cellular standards, WiGig, 4K video, and other developments are causing quite a flurry of market activity. In fact, they are driving the creation of myriad new devices and making existing products more functional. But these innovations are also creating formidable challenges for test equipment designers. One of them involves integrating a power supply into a design.
Specialized power supplies are now designed with operational parameters most important to test equipment performance. These important parameters include conducted and radiated electromagnetic interference (EMI), electromagnetic compatibility (EMC), differential/common mode noise, thermal performance, regulatory requirements, product life and other more application-specific factors. A power supply that excels at these parameters can be integrated into test equipment with less effort.
This is particularly true for electromagnetic noise. Equipment designers can spend significant amounts of time alleviating power supply noise effects by adding filters and/or taking numerous measurements to ensure readings are accurate and not influenced by noise.
It is useful to consider the standards that can apply to noise levels in test equipment. EMI limits, for example, are detailed in various industry standards:
EN55011 is a European standard that applies to industrial, scientific and medical (ISM) equipment. (CISPR 11, by the International Committee on Radio Frequency Interference, is an international standard that is essentially equivalent to EN55011.) The standard divides these products into two groups, Group 1 and Group 2. EN55011 Group 1 limits are identical to EN 55022 limits. Group 2 pertains to products that use RF as an output. Therefore, switch-mode power supplies are not considered in Group 2.
EN55022 (CISPR 22) applies to information technology equipment (ITE). EN55015 (CISPR 15) applies to lighting equipment. Both EN55011 and EN55022 contain two classes: Class A and Class B. EN55015 has only one set of limits: Class A equipment is that used in a domestic environment which may cause radio interference with other equipment nearby. Class B equipment is designed to be used in a domestic environment and won’t cause radio interference with nearby equipment.
For conducted interference, EN 55022 and EN 55011 measure the emissions from equipment in the 150 kHz to 30 MHz range, while EN 55015 measures over the extended range of 9 kHz to 30 MHz using a conducted measurement technique on the ac mains input cable. The conducted emission limits in these standards are intended to protect equipment connected to the same ac mains supply.
For radiated interference, EN 55022 and EN 55011 measure equipment emissions in the frequency range from 30 MHz to 1 GHz while EN 55015 measures the emission from equipment in the frequency range 30 MHz to 300 MHz using a receiving antenna and recording the over-the-air signals. The radiated emission limits applied by these standards aim to protect equipment near the device being tested.
Power supply manufacturers test their products to these standards, usually powering a resistive load. Of course, once power supplies are integrated into end equipment, the load profiles may change. End equipment can change the EMI profile, perhaps forcing the supply to stray from the EMI limits. It then becomes the job of equipment designers to bring EMI in the end equipment below the applicable limits. This is one reason equipment designers often use power supplies that meet EMI standards with a wide margin of safety: Such supplies are more likely to avoid EMI concerns when built into end equipment.
Other operational parameters that test-equipment designers are concerned with relate to EMC. In general, EMC standards cover many different parameters that deal with outside influences that can affect the operation of both power supplies and the end equipment they are used in. EMC standards for power supplies are listed under IEC61000 (in Europe, EN61000). Equipment that goes through testing to this standard can have one of four levels of acceptance criteria:
• Criteria A — Normal performance within limits specified by the manufacturer, requestor or purchaser.
• Criteria B — Temporary loss of function or degradation of performance which ceases after the disturbance is removed and from which the equipment-under-test recovers its normal performance without operator intervention.
• Criteria C — Temporary loss of function or degradation of performance, and the operator must intervene to correct it.
• Criteria D — An unrecoverable loss of function or degradation of performance caused by damage to hardware or software, or loss of data.
Not all of the above standards may apply for all test equipment, and the performance levels and criteria of those standards will also vary, depending on the environments in which the end equipment is being designed to operate.
Makers of power supplies should specify not just the IEC/EN61000 standard the supply follows, but also the level and operational criteria with which the supply complies. Stating that the power supply complies to a given standard without specifying the test level at which it complies and the performance criteria during the test, does not give the system designer enough details to evaluate the supply. Power supplies that meet tougher standards can give equipment designers confidence that their system-level EMC tests won’t cause problems without adding filters or other protections.
An equally important source of noise is CMN. CMN on the output of a power supply is often overlooked and rarely specified. This oversight may be because the designer doesn’t know how it can impact product performance, but also because it may not be an issue. Power supply outputs are often referenced to earth or chassis ground in end applications. In this case, CMN will likely not be an issue. However, it can be instructive to understand the common-mode currents that flow while making their way to ground.
The EN 55103-1 EMC standard (for audio, video, audio-visual and entertainment lighting control apparatus), EN55022, and associated standards (CISPR 22) give limits and measurement methods for common-mode disturbances on telecommunications and network ports. However, it is common practice to reference the output of the power supply to ground in this type of equipment. Refer to these standards for the measurement method and limit.
For applications where these standards do not apply, loads can see adverse effects from common-mode voltages. If the output of the power supply connects to chassis/earth ground somewhere on the load board (electronics circuitry), there can be common mode currents flowing through the board. Also, if the printed circuit board layout is subpar, current can flow through sensitive parts of the circuitry and cause malfunctions. In the case of test equipment, the current may cause erroneous measurement readings unless the common-mode noise (CMN) is filtered or potential errors filtered out by data sampling.
CMN can be specified in Volts rms or peak-to-peak, or in units of current (mA). Designers concerned with CMN tend to analyze it based upon its effects on their design.
Power supply datasheets typically do not specify CMN. It may be impractical to eliminate CMN, but a power supply that can specify a CMN value, both high frequency noise and low frequency noise, will give system designers some idea of the components needed to integrate the power supply into their equipment.
All in all, many power supplies specify performance to a few parameters, leaving designers to tackle others during development. A power supply that can performs well for most, if not all of these important parameters, will ensure an easier integration effort for design engineers. Review those datasheets carefully, as what you don’t see may cause extra work later.
SL Power Electronics Corp.
Leave a Reply
You must be logged in to post a comment.