ANSI C63.4 and CISPR 22-Harmony

ANSI C63.4 and CISPR 22-Harmony at Last? Donald N. Heirman Lucent Technologies, Bell Laboratories Innovations Holmdel, New Jersey 07738 USA Abstract: This paper compares the most prevalent emission measurement procedures used internationally-ansi C63.4 and CISPR 22. In the past year, an amendment to CISPR 22 has been approved and published which contains much of the test setup and principles contained in ANSI C63.4. However, there are differences which my affect the test results. These differences are specifically addressed in this paper with the view of their significance in the measurement result, if any. The reader should be able to use this paper to prove that one test is possible to show conformance to the respective limits, provided the same option is selected from both standards. Introduction It has taken over 10 years of effort to develop radiated and conducted emission test techniques for digital devices in the United States. The FCC had various publications which contained their emission measurement techniques such as their now replaced Measurement Procedure Number 4 (MP-4) [l]. Test site requirements (in particular requirements for an open area test site (OATS)) and limits were in another part of their Rules. All, except the limits location which remains in the FCC Part 15 Rules, changed first with the FCC s adoption of ANSI C63.4 in 1991 [2]. This ANSI document was updated a year later to its present version [3]. Both the 1991 and later the1992 documents replaced MP-4 measurement methods and at the same time included test site requirements all in one place. Since 1985, a parallel effort has been underway internationally to develop an international version of MP-4 at the start. That work was conducted in the International Electrotechnical Commission (IEC) Special International Committee on Radio Interference (CISPR). In fact if the reader reviews the 1985 edition of CISPR 22, it contains similar language to that of MP-4. Since then through the efforts of the US delegation to CISPR, wording in C63.4 was proposed as additions to CISPR 22. This lead to the publication of the second edition of CISPR 22 in December of 1993 which picked up one of the most important part of C63.4-the addition of test site validation identical to that in C63.4 for so-called alternates to open area test sites (OATS@[4]. In parallel to the effort to update CISPR 22, the US delegation also proposed and it was accepted to have the site validation for OATS in CISPR 16, Part 1 [S]. But the necessary convergence of the CISPR work and that of C63.4 was still incomplete in that the second edition of CISPR 22 still did not have test setups which is the hallmark of the C63.4 documents (both in the 1991 and 1992 versions). That last detail came about in August of 1996 when the second amendment of CISPR 22 was approved and published [6]. The amendment now has test setups which are virtually identical to those in C63.4. But the word virtually sends shakes through any test community which is concerned with repeatability and answering the question: can one test be done to meet both the CISPR 22 requirements as well as the FCC s C63.4 requirements? This paper will describe the similarities and in particular look at significant differences between the two documents that, in the author s opinion, may affect measurements results, Comparison between CISPR 22, Amendment 2, and C63.4 Definitions The definition of Equipment Under Test was modified slightly in Clause 3.2 in Amendment 2 of CISPR 22 by indicating that the EUT is used for evaluation purposes rather than being evaluated. In C63.4, the EUT is defined further to indicate that it is representative of a product to be marketed. The C63.4 definition brings in the representative nature of the EUT as it is marketed, clearly showing the emphasis on the use of the typical use of the EUT by the consumer. The CISPR 22 definition does not. EUT con&m&on There is a departure in the amendment to CISPR 22, Second Edition. In Amendment 2, the concept of preliminary or initial testing is introduced as is also contained in C63.4. The bulk of the disturbance maximization process is included in these preliminary or initial tests. Within the context of preliminary testing, Amendment 2, clause 9.1.1, requires that investigations shall be made at a number of significant frequencies to give confidence that the probable frequency of maximum disturbance (this is used instead of interference) has been found with respect to the limit. This shall be done while the EUT is being rnn through its typical modes of operation and cable positions in a test setup which is representative of typical system configurations. In C63.4, 0-7803-4140-6/97/$10.00 112

there is a more explicit requirement to manipulate cables (See Clauses 7.2.3 (conducted emissions) and 8.3.1.1 (radiated emissions)). Further in C63.4, clauses 6.2.1.3 and 6.2.3.3, cable manipulation is explicitly mentioned, but warns that the cables shall never be placed on top of or under the EUT or its components (unless such placement is inherent in the EUT design). Thus, Amendment 2 requires less maximization based on cable manipulation than C63.4 and hence C63.4 may be considered more rigorous in this respect. In both Amendment 2 and in C63.4 there are many test setups are used for these initial tests. If there are multiple interface ports of the same type, Amendment 2 indicates that the number of additional cables populating these ports shall be limited to the condition where adding another cable does not decrease the margin by, for example, 2 de!. Note that in C63.4 where there are multiple identical ports, there is a requirement in clause 6.1.3 to cease adding cabling when the change is less than 2 de? aped the emissions are below the limit. Though the wording is somewhat different, the intent is the same. Hence, both CISPR 22, Amendment 2, and C63.4 are equivalent in employing preliminary testing to determine the configuration, mode of operation and cable position that yields the maximum emission at a particular frequency which has the least margin with respect to the limit. To summarize, the extent of cable manipulation is significantly reduced in Amendment 2 to that which is representative of typical system configurations. In C63.4, there is more cable manipulation guidance for the maximization process used in preliminary testing. It is interesting to note that in C63.4, clause 8.3.1.1, it states that for floor-standing equipment, cabling should be installed as it would be by the user and no further manipulation is made. In this case, no cable manipulation is allowed. No such guidance is given in Amendment 2. The basic guidance, however, is identical to that in Amendment 2, i.e. use typical or representative configurations. The next change that Amendment 2 made to the second edition of CISPR 22 was to introduce how to handle multiple modules such as plug-in cards and boards. As usual the multiple modules should be a mix of those which can be added and that the number to use is representative. Amendment 2 then indicates that the number of modules used should be limited that when another module is attached, the margin with respect to the limit is not reduced by for example 2 db. C63.4 indicates that the addition of modules should stop when the change in emissions are less than 2 dl3. Hence, in both standards the rule for adding more modules is virtually the same as that for adding additional cabling to identical ports. Amendment 2 then goes into considerable detail for two examples of a minimum representative configuration: a personal computer or computer peripheral and a point of sale terminal. The configuration for personal computers or their peripherals are identical to that in C63.4, Clause 11.2. The point of sale terminal configuration is not contained in C63.4 and hence for such EUTs, Amendment 2 should be used when conducting C63.4 tests since that configuration is not in conflict with the general guidance in C63.4. Amendment 2 also adds a clause which allows equipment which might be a subsystem of a distributed system such as a data processing terminal, workstation, and private branch telecommunications exchange to be tested independently of a host or system. In addition, distributed networks such as local area networks may be simulated on the test site by lengths of cables and actual peripherals or remote network communications simulators located so that these simulators do not affect the emissions. The ability to measure the subsystem and to simulate distributed networks is also allowed in Clause 6.1.2.1 of C63-4. Emission Maximization As in C63.4, Amendment 2 has a two step process where initial measurements are made first to identify the highest disturbance relative to the limit while the EUT is operated in its typical modes of operation and cables located in its representative positions for a typical system configuration. To show samples of these representative setups, Amendment 2 introduces 10 setup figures. These figures are quite close to those in C63.4 and are described in the Comparison of Test Setups section of the paper to follow. Using the figures will help determine the frequency where the highest emissions with respect to the limit are found and the associated cable position, configuration, and modes of operation of the EUT at this frequency is fixed and identified. C63.4 splits up this preliminary testing in separate clauses for conducted emission testing (See Clause 7.2.3) and for radiated emission testing (See Clause 8.3.1.1). However, the requirements are the same as for Amendment 2. The second step in the process is the final measurement which is made using the results of the preliminary measurements, but now the entire frequency spectrum covered by the limits are scanned without further changes to the cable position, con@uration, and modes of operation of the EUT found to yield the highest emissions with respect to the limit during preliminary testing. These final tests are covered in Amendment 2, clauses IO (conducted) and 11 (radiated), and C63.4 clauses 7.2.4 (conducted) and 8.3.1.2 (radiated). The similarities in these tests are covered in the following section. Of particular interest is that C63.4 also presents radiated emission measurement procedures which 113

cover the frequency range 1 to 40 GHz. Amendment 2 does not provide any information for measurements above 1 GHz. However, CISPR Subcommittee G (ITE) has a voting document to amend CISPR 22 to contain limits and a method of measurement up to and including 18 GHz [7]. C63.4 only contains methods of measurement. Applicable limits are called out by regulatory agencies which in the USA is the Federal Communications Commission in their Part 15 Rules and Regulations. EUT Operation In amendment 2, clause 9.2, the operation of the EUT is described. These are the same conditions as shown in clauses 11.1 and6.1 of C63.4. F%ual Display Units Amendment 2, clause 9.2.1 contains the operating rules that shall be used when visual display monitors are part of the EUT. This is identical to clause 11.1.3 in C63.4. Facsimile Devices reference groundplane a horizontal surface. Thus the EUT can be placed on a non-conductive table 40 cm high with respect to the horizontal conductive groundplane ( normally the floor of a shielded room or open area test site). C63.4, clause 5.2.2 prescribes that only a vertical conducting surface shall be used for table-top products. C63.4 does, however. allow the option to not use the vertical conducing surface at the OATS with the proviso that in case of disputes, a measurement using the vertical conducting surface takes precedence. Finally, for floor-standing EUTs, both standards do not require the use of a vertical conducting reference groundplane. These differences may have an impact on the emissions measured. Clearly, the coupling that is present with the bottom of a table-top device next to a horizontal reference conducting groundplane is different than the coupling of the back of the device to a vertical reference conducting groundplane. Hence to meet the requirements of both standards for one test, it is prudent to select the vertical reference conducting groundplane and the EUT and peripherals located on the 80 cm high table-top. This arrangement is allowed in both standards. Amendment 2, clause 9.2.2 presents the mode of operation to use in measuring emissions f?om a facsimile machine. It also notes that for such devices, the referenced test pattern may have to be sent many times to obtain the full disturbance potential of the device. There is no such product specific instructions in C63.4. Telephone Sets Clause 9.2.3 in Amendment 2 cites the mode of operation that shall be used when testing telephone sets which are capable of transmitting voice information using digital signals (and hence classified as an ITE). The received signal uses a referenced standard speech data pattern. C63.4 does not specify this product specific test condition. This concludes the comparisons for the general measurement conditions for both standards. Mains (AC) Ports Conducted Disturbance Measurements Amendment 2, clause 10.2 introduces the second LISN (AMN) which is used to power all non-eut mains cords (including those non-eut cords which are connected to a multiple outlet power strip which in turn is connected to the second LISN (AMN). This matches clause 7.2.1 of C63.4. Groundplane Clause 10.3 of Amendment 2 includes the option of measuring conducted emissions for table-top EUTs with the Equipment Setup for Conducted Measurements Clause 10.4 of Amendment 2 introduces several test setup figures that are used during the conducted emission tests. These figures are very similar to those in C63.4 and are further described later in this paper in the section on Comparison of Test Setups. The figures in Amendment 2 show the option of referring the measurement to either a vertical or horizontal reference conducting groundplane with the EUT 80 or 40 cm above the floor, respectively. Recording of Conducted Measurements Amendment 2, clause 10.5 indicates that only those emissions that are within 20 db of the limit are to be recorded. In that band, only the amplitudes of the frequencies of the 6 highest emissions with respect to the limit are recorded for each mains port (of the EUT). The current carrying conductor within each mains cord for which these emissions were recorded must be identified. This requirement is the same as C63.4, clause 10.1.8.1. That C63.4 clause goes further to state that if emission amplitudes are more than 20 db below the limit, then the level of the instrumentation noise at representative frequencies shall be recorded. Interestingly in Amendment 2, clause 10.5 also indicates that conducted emissions shall also be recorded at telecommunications ports. This is in anticipation of a further amendment or edition of CISPR 22 which will contain conducted emission requirements and test setups for 114

measuring emissions at telecommunications ports between 150 khz and 30 MHz-the same frequency range as that of the mains measurements Equipment Setup for Radiated Disturbance Measurements Clause 11.4, Amendment 2, shows in figures how the EUT shall be contigured and operated for making radiated disturbance measurements for three situations: tabletop, floor-standing and combined floor-standing and tabletop equipment. These figures are quite similar to those in C63.4. The particular comparisons will be discussed in the section on Comparison of Test Setups. The figures showing the use of overhead cable racks in Amendment 2 is more complete than that of C63.4, however. This clause also indicates that if an EUT is designed for both tabletop and floor-standing operation, the EUT shall be tested only in the tabletop configuration, unless the typical installation is floor-standing. Clause 6.2.3 of C63.4 has an identical statement but adds that the regulatory authorities require a specific choice for the two configurations. In addition, wall or ceiling mounted equipment shall be tested as tabletop equipment in both standards, Recording of Radiated Disturbance Measurements Clause 11.5, Amendment 2, states that the six highest disturbances with respect to the limit and above 20 db below the limit shall be recorded including the polarization. This is identical to clause 10.1.8.2 in C63.4 which also gives guidance on how to record disturbances when there are less than 6 frequencies which have amplitudes either not above the 20 db below the limits level or are in the noise. This is identical to the instructions for recording conducted emission measurements. Other Comparisons There are many other comparisons that can be made between the 1991 and 1992 editions of C63.4 and the 1985 and 1993 versions of CISPR 22 181. These comparison will not be repeated in this paper. However, there are a few comparisons of particular interest, The Second Edition of CISPR 22 contains as an Annex tables to use in comparing measured site attenuation with theoretical values for alternate (to an OATS) test sites such as absorber lined rooms and all-weather enclosed open area test sites. These tables are identical to Tables 1 and 2 of C63.4. Note that for the time being in C63.4, an all-weather covered OATS is not considered to fall under the alternate test site normalized site attenuation requirements (See clause 5.4.2 of C63.4). However, C63 Subcommittee 1, chaired by the author, is studying the need to remove this exemption. In CISPR, the exemption does not exist. Comparison of Test Setups Perhaps an even easier way to compare C63.4 and CISPR 22, Amendment 2, is to concentrate now on the differences and similarities in the test setup figures. The figures parallel each other so that tabletop and floor-standing setups can be compared directly. There are figures added to CISPR 22 which are not in C63.4 and vice-versa. We start by stating the basic differences and similarities which are in each of the three basic test setups discussed below. Tabletop Setups Conducted Measurements 1. 2. 3. 4. 5. 6. 7. Amendment 2, Figure 4. does not have two peripherals on either side of the EUT as is the case with C63.4 Figure 9 (a). Instead, it just shows one peripheral. Amendment 2, Figure 4, shows an associated equipment connected to at least one of the I/O cables attached to the EUT. That I/O cable lies directly on the conductive groundplane as does the Artificial Mains Network @MN). Note that the AMN is called a Line Impedance Stabilization Network (LISN) in C63.4. C63.4 does not show any location for associated equipment not on the tabletop but the LISN is also located directly on the conductive groundplane. Keyboards for PCs represented in Amendment 2, Figure 4, are close to the EUT CPU; in C63.4, Figure 9 (a), the keyboard is positioned at the front edge of the 1.5 by 1 meter rectangular tabletop. Amendment 2, Figure 4, shows more clearly the position of the AMN, the LISN in C63.4 does not although the text and figure footnote instructions on placements in both standards are clear. Amendment 2, Figures 5 through 7, shows the mounting of the AMN on the conductive groundplane below the EUT or to the reference vertical conducting wall which is located 40 cm from the EUT; C63.4, Figures 9 (a), 9 ( c) and 11 only allows the LISN on the conductive groundplane. When the AMN is located on the vertical wall, the associated equipment can be moved away from the conductive groundplane as shown in Figure 5 in Amendment 2. Both standards allow the use of a multiple outlet strip to power all other equipment which is part of the EUT system under test. Amendment 2, Figure 7, has a setup for conducted emission measurements at an open area test site (OATS) where the vertical conducting surface is not used; C63.4 does not show this setup, but has wording in clause 5.2.2 that allows the optional use of the vertical conducting surface when conducted emissions are made at an open area test site. However, in case of disputes, the use of the vertical conducting surface takes precedence. 115

Radiated Measurements For radiated emission testing using both standards. the LISN (AMN) is allowed to be used as long as it is installed below the groundplane and the mains (AC) junction box or receptacle is bonded to and flush with the groundplane. The powerline (mains) cord(s) for the EUT is not bundled in this test as it was when connected to the LISN (AMN) for the conducted emission tests. In fact the setups show that all powerline (mains) cords are connected to their power receptacles located immediately under the non-conducting tabletop. This arrangement allows powering the EUT and associated equipment from receptacles located in or near the center of a turntable. Other differences include: 1. Amendment 2, Figure 10, does not show the location of associated equipment (AE) which may be on the turntable or remotely via a cable that goes through the stationary part of the turntable at its center. For the conducted emission measurements using Figures 4, 5, and 7 the location of the AE is clearly shown. 2. Amendment 2 and C63.4 both do not show the use of extension cords for these radiated measurements. Floor Standing Setups Conducted Measurements 1. Amendment 2, Figure 8, shows the location of associated equipment nearby to the EUT and located on the reference groundplane (though not shown explicitly, the AE is insulated from the groundplane); C63.4, Figure 9 (b) does not show any AE. 2. Amendment 2, Figure 8, does not show in the test setup figure the serpentine cable arrangement for I/O cables connecting the two adjacent EUT floor standing cabinets yet in the notes associated with the test setup figure does allow this arrangement only if the excess I/O cable cannot be folded back and forth forming a bundle 30 to 40 cm long; C63.4, Figure 9 (b), explicitly shows an I/O cable bundled in a serpentine fashion, 3. Amendment 2 and C63.4 both show the same arrangement for the primary LISN (AMN) and the secondary LISN (AMN), the latter used for connecting the mains cords for non-eut equipment in the test setup. 4. Both Amendment 2, Figures 13 and 14, and C63.4, Figure 10, show test setups with EUTs which are installed using overhead cable trays and supports. Amendment 2 shows two cabinets as EUTs with cables connecting them via an overhead cable tray; C63.4 shows only 1 EUT cabinet with the interconnecting cables to remote (not on the turntable) associated equipment. For both tests, the LISN (AMN) is located on the turntable next to the EUT (80 cm away) with Amendment 2 indicating that the AMN can be located below and bonded to the conductive groundplane. Radiated Measurements 1. Amendment 2, Figure 11, shows that the only way to handle the excess I/O cable between two adjacent EUT floorstanding cabinets is by the usual bundling (folding back and forth the cable) though in a footnote serpentine bundling is allowed C63.4, Figure 9 (d) allows both types of bundling as well. 2. Amendment 2, Figure 11, allows bundling of the excess powerline (mains) cable between the EUT and the receptacle on the conducting ground plane (insulated above the ground plane by up to 12 mm of insulating material). Amendment 2 also allows the mains cable to be shortened to an appropriate length. C63.4, Figure 9 (d), allows the same conditions for the mains cords. 3. Both standards require the powerline (mains) receptacle to be flush with the groundplane and if LISNs are used, they must be located under the groundplane. Combination Tabletop and Floor-standing EUT systems Amendment 2 has more detail in test setups for the situation where the EUT is comprised of equipment which rests on a tabletop and on the floor (insulated from the conducting groundplane). C63-4, Figure 13, is a generic setup for both conducted and radiated emissions. The details are described below. Conducted Measurements 1. 2. 3. Amendment 2, Figure 9, shows two power mains cords associated with the EUT system. Hence, the location of the two LISNs (AMNs) are clearly shown. The LISNs (AMNs) are 80 cm away from both parts of the EIJT, i.e. the tabletop portion and the floor-standing portion. C63.4, Figure 13, does not show LISNs but the text in the general equipment setup for conducted emissions in clause 5.2.3 sets similar requirements for the LISN location. Amendment 2, Figure 9, does not show any I/O cables leading to associated equipment remote from the test site. C63.4. Figure 13, shows how cables exit the test site, but does not show isometric or plan view the relative positions of the tabletop equipment and the floorstanding parts of the EUT system and does not show any I/O cabling between the two parts of the EUT system. Both Amendment 2 and C634. allow for the use of additional LISNs (AMNs) to power associated equipment or equipment which is not being measured for its conducted emissions. 116

Radiated Measurements Again, Amendment 2 has more details for this setup. 1. Amendment 2, Figure 12, and C63.4, Figure 13, both show that the LISN (AMN) is removed from above the groundplane for radiated emission measurements. If the EUT is connected to an LISN (AMN), the LISN (AMN) must be mounted under the groundplane for both standards. 2. Amendment 2, Figure 12, shows an I/O cable attached to the floor-standing EUT to simulate connection to remote associated equipment. This connection was not shown in the conducted emission test setup in Figure 9-this is a curious omission, 3. Both Amendment 2 and C63.4 indicate that the floor standing part of the combination EUT comprised of a tabletop and the floor-standing portions should be placed on an insulated surface up to 12 mm in height (C63.4 allows between 3 and 12 mm). In both standards, the tabletop and floor-standing portions of the EUT should be separated with typical spacing. Amendment 2. Figure 12, shows that the I/O cable between the tabletop and the floor-standing units can be draped to the floor but if the cable cannot reach the floor, the cable is draped to the height of the connector of the floor standing unit or 40 cm above the groundplane whichever is lower. C63.4 does not give this guidance. Equipment using overhead cables Conducted and Radiated Emissions Both Amendment 2 and C63.4 show EUT setups where the EUT is a floor-standing rack of equipment and I/O cabling is normally routed from the top of the EUT to an overhead cable manager or rack. Figure 13 in Amendment 2 shows such setups where there are two floor standing racks interconnected via an overhead cable rack, where C63.4 shows a much simpler single rack system where the I/O cable goes to presumably a remote location under the groundplane. In any case, there is more detail in Figure 13, In addition, Amendment 2 splits the setup into two figures: one is a plan view and the other is a front view. The same constraints hold i.e. for radiated emission measurements, the LISN (AMN) must be located below the groundplane with the power receptacle flush with the groundplane. The vertical riser (that part of the cable management system from the groundplane to the overhead rack) used to support the overhead cable rack or manager must be the one that is used when the EUT is installed and that only one cable riser is used for the path from the groundplane to the top horizontal portion of the cable rack (for each floorstanding unit of the EUT system). Summary This paper has shown comparisons of measurement procedures contained in ANSI C63.4 and CISPR 22, Amendment 2. The test setups are so close to each other that it is now possible to show product compliance using either measurement procedure with the possible exception of conducted emissions on tabletop EUTs. In this case, using a 40 cm high tabletop allowed in Amendment 2 instead of the 80 cm tabletop height preference in C63.4 may lead to different results. The procedures in CISPR 22 and its amendments were heavily influenced by contributions of the United States National Committee of the International Electrotechnical Commission and its Special International Committee on Radio Interference (CISPR). The US members (the author was one of the principle contributors) of CISPR Subcommittee G (Information Technology Equipment) were instrumental in getting C63.4 inputs into CISPR 22 in order to harmonize both documents. As will be the case, each document as it is changed will be slightly out of step with each other since both documents do not undergo changes in synch and in time. It is the challenge of the working groups within C63 and CISPR/G to keep their respective documents so synchronized (or maybe integrate into one document). References [l] FCCYOET MP-4 (1983), FCC Measurement Procedure MP-4, FCC Measurement of Radio Noise Emissions from Computing Devices. [2] American National Standard for Methods of Measurement of Radio-Noise from Low-Voltage Electrical and Electronic Equipment in the Range of 9 khz to 40 GHz, ANSI C63.4-1991. [3] American National Standard for Methods of Measurement of Radio-Noise Emissions from Low-Voltage Electrical and Electronic Equipment in the Range of 9 khz to 40 GHz, ANSI C63.4-1992 [4] CISPR 22 (1993)-Second Edition, Limits and methods of measurement of radio disturbance characteristics of information technology equipment [5] CISPR 16-l (19931, Part 1: Radio disturbance and immunity measuring apparatus [6] CISPR 22 (1996)-Amendment 2, Limits and methods of measurement of radio disturbance characteristics of information technology equipment [7] CIS/G/l18/CD, Proposal for emission limits above 1 GHz, Comments on this draft are due by 30 April 1997. [8] D. N. Heirman, Progress in ITE Emission Measurement Methods, 1 lti International Zurich EMC Symposium and Technical Exhibition, 7-9 March 1995. 117