Importance of Traceability in EMI Measurements

The recognition of the importance of measurement traceability significantly increased over the last 20 years, especially as part of the test and calibration laboratory accreditation programs that were established worldwide. The generally accepted quality system standard ISO/IEC 17025-2005 includes a set of requirements addressing the subject of traceability of measurement results. These requirements do also apply to EMC test laboratories. This article will introduce the concept of traceability, discuss the role an EMC test laboratory must assume to ensure traceability of test results and will introduce a future amendment to CISPR 16-1-1 which describes the requirements for calibration of EMI receivers and spectrum analyzers.

The definition of traceability that is globally accepted in the metrology community is included in the International Vocabulary of Metrology – Basic and general concepts and associated terms: “…property of a measurement result whereby the result can be related to a reference through a documented unbroken chain of calibrations, each contributing to the measurement uncertainty.”

Traceability means that the result of a measurement, no matter where it was made, can be related to a national or international measurement standard, and that this relationship is documented. In addition, the measuring instrument must be calibrated by a measurement standard that is itself traceable. Traceability is thus defined as the property of the result of a measurement or the value of a standard whereby it can be related to stated references, usually national or international, through an unbroken chain of comparisons all having stated uncertainties. It is essential to note that traceability is the property of the result of a measurement, not of an instrument or calibration report or a laboratory. It is not achieved by following a particular procedure or using special equipment.

The concept of traceability is important because it allows the comparison of the accuracy of measurements worldwide according to a standardized procedure for estimating measurement uncertainty.

Within a chain of traceability, the units of measurement with the highest accuracy are realized by international measurement standards. The value of the international standard is usually determined by comparison of national standards of the highest quality (or in the case of the kilogram by the mass of the International Prototype). National measurement standards, maintained in a national metrology institute or NMI (for example, NPL in the UK, NIST in the USA) must be compared with these international standards. The result of such comparisons, together with the precision and uncertainty of the national standard will be stated and will be available on, for example, the internet (see the BIPM key comparison database at www.bipm.org/kcdb/). Then the national measurement standard serves as a reference for calibration of standards of lower precision. Reference standards are kept in a national metrology institute or in an accredited calibration laboratory for calibrations not requiring the highest accuracy. Again, the result and the uncertainty will be stated.

At each stage in such a chain of traceability, one loses a certain degree of precision. Thus the highest level standards are the international standards, known with the greatest level of precision, and the lower level standards will have been determined to a lower level of precision. This lower level of precision will be one which is acceptable or appropriate for the use of that particular standard.

For an EMC test laboratory to achieve traceability it is essential to use measuring equipment that is calibrated in a traceable manner and also meets the specifications called out in CISPR 16-1-1 to ensure that the expected measurement instrumentation uncertainty for conducted and radiated disturbance measurements or disturbance power measurements can be achieved. Since the EMC test laboratory is responsible for the selection and use of adequate measuring equipment, as well as the purchase of appropriate (meaning accredited or otherwise deemed suitable) calibration services to ensure traceability of test results, a clear understanding of the calibration requirements is essential. The determination of the necessary specifics of a calibration service in the purchasing process and the review of the obtained calibration service upon receipt of the equipment back from the calibration laboratory before it is placed back into service at the test laboratory are major tasks the test laboratory must complete in order to ensure the proper calibration of test equipment. The importance of test equipment calibration and traceability aspects was also acknowledged by CISPR subcommittee A which is in preparation of normative Annex to CISPR 16-1-1, defining calibration requirements for measuring receivers.

Role and Responsibilities of the EMC Test Laboratory

An accredited EMC test laboratory is required to specify the details of a calibration service to be purchased (technical and/or administrative aspects) to the calibration laboratory to ensure that a suitable calibration service is provided and the equipment is calibrated for the actual application. This information can be included on a purchase order, be provided as a separate document as an attachment to a purchase order, be included in a general contract with a calibration laboratory or can be communicated in any other way. The following aspects must be considered when purchasing a calibration service:

• If a calibration standard is available for the calibration of a specific piece of test equipment like for a measuring receiver (i.e., CISPR 16-1-1) or for antennas (e.g., ANSI C63.5 or the future CISPR 16-1-6) the specification of the applicable standard must be included in the calibration request. In case the applicable standard does include multiple calibration methods (e.g., ANSI C63.5 or the future CISPR 16-1-6) the method to be used is to be included in the request as well.
• If no standard is available to calibrate a piece of test equipment like for spectrum analyzers or signal generators the EMC test laboratory should request the use of the equipment manufacturer’s calibration process to ensure that compliance of the equipment under calibration with its specifications can be determined without ambiguity. It is essential to know for an EMC test laboratory that equipment still meets its specifications upon arrival at the calibration laboratory.
• Technical details like the required frequency range or amplitude range, if necessary, are to be specified if equipment is used in a limited fashion. For example, a spectrum analyzer is only used in a frequency range narrower than the capability of the instrument (e.g., the instrument covers the frequency range up to 26 GHz but the laboratory performs emission measurements under its scope of accreditation to 6 GHz only).
• The requirement for an accredited calibration envelopes all calibration parameters of the equipment to be calibrated under the scope of accreditation of the calibration laboratory. This is essential to ensure proper traceability of EMC measurement results.
• The test laboratory should also request the inclusion of the accreditation body’s symbol on the calibration certificate for easy identification that an accredited calibration was performed.

When a measuring receiver is to be calibrated for the sole purpose of performing emission measurements, the EMC test laboratory has two choices: Either verification per CISPR 16-1-1 can be requested or a full calibration in accordance with the manufacturer’s calibration procedure can be ordered. A calibration laboratory will perform the verification of the instrument by performing the measurements specified in CISPR 16-1-1. Parameters to be verified are summarized in Table 1 below, per identified sections in CISPR 16-1-1. If these measurements are performed under the calibration laboratory’s scope of accreditation the EMC test laboratory will have a measuring receiver available for measuring emissions in a traceable manner. It is to be noted though that such a verification in accordance with CISPR 16-1-1 does not envelope all calibration parameters of a measuring receiver. For example, frequency accuracy, frequency stability, or displayed average noise level are not part of the CISPR 16-1-1 verification process. Therefore, if this instrument is also to be used for other purposes like measurements on intentional radiators (e.g., licensed or unlicensed transmitters) this verification will be insufficient and a complete calibration of the instrument in accordance with the manufacturer’s calibration process is required. Compliance with the specifications of an instrument can only be determined if the manufacturer’s calibration process is applied during the calibration process.

EMC test laboratories are also responsible for the selection of adequate calibration laboratories. Many accreditation bodies have established policies that define requirements related to the traceability of measurement results which very often call out the requirement for use of accredited calibration laboratories. It is to be noted that this requirement is not included in ISO 17025-2005 but established by the accreditation bodies. Use of accredited calibration service providers is the easiest way to ensure traceability for a test laboratory. Today, almost all equipment used by an EMC test laboratory can be calibrated by an accredited calibration laboratory, assuming a suitable scope of accreditation. When selecting a calibration service provider the review of the scope of accreditation of a prospect calibration laboratory is an important step in the evaluation process. EMC laboratories must maintain records of such evaluations per ISO 17025-2005, clause 4.6.4.

Upon return of calibrated equipment from the calibration laboratory the EMC test laboratory must perform an incoming inspection of the received equipment before it is put back into service. This step is essential to avoid the use of equipment for testing work which may be improperly calibrated or may have ambiguous or unclear documentation. Only after a thorough review of the equipment should it be made available for measurements in the EMC test laboratory to avoid possible non-conforming work scenarios that could cause additional investigative work or even retests.

The incoming inspection of equipment received back from a calibration laboratory should address the following items, as applicable:

• Identification: The serial number, unique identification number (if used) and the calibration date/due date (if requested by the EMC test laboratory) on the certificate must match the information on the calibration sticker affixed to the equipment.
• Accuracy: The values provided on the certificate/report must be adequate for the intended use of the equipment.
• Traceability: The information which establishes traceability to national/international standards is to be verified. The presence of a symbol of the accreditation body, or reference to the accreditation status of the calibration laboratory is to be determined. Note: Traceability is not established merely by making a statement to that effect.
• Measurement uncertainty: The certificate must include an appropriate statement of measurement uncertainty and where applicable, the before and after data of the calibration in case an adjustment was required.
• Special instructions: If any special instructions were given to the calibration service provider for the calibration of test equipment, it must be verified that they were carried out.
• Documentation of In/Out of Tolerance information: It is to be verified that information is included on the certificate which states the condition of the test equipment (i.e., In Tolerance or Out of Tolerance) when received at the calibration laboratory and before shipment back to the EMC test laboratory.
• Tamper-resistant seals: If the calibration laboratory applied tamper-resistant seals it is to be verified that these seals are not broken. If this is the case the calibration is deemed void.
• Completeness: It is to be verified that a complete calibration of the test instrument was performed under the calibration service provider’s scope of accreditation. The calibration documents are to be reviewed to determine if any calibration activities were performed outside the scope of accreditation (sometimes indicated by a foot note or a remark on the certificate).

When equipment was found to be out of tolerance, as stated on the calibration certificate, the test laboratory will have to use its non-conforming work process to determine how this out of tolerance situation may have impacted previous test results. Where necessary, technical evaluations (e.g., verification tests or an instrument self-test) are to be performed by the EMC test laboratory to establish that the equipment is functioning as expected.

Calibration Requirements for EMI Receivers per CISPR 16-1-1

The importance of equipment calibration and traceability of test results is recognized by CISPR. Since the calibration of measuring receivers (which are defined in CISPR 16-1-1 as an EMI receiver or spectrum analyzer without preselection) caused confusion in the international EMC community, CISPR subcommittee A is in preparation of a normative annex to CISPR 16-1-1 to outline the calibration requirements for measuring receivers. The following subjects will be addressed:

Calibration and verification

In CISPR 16-1-1 metrological calibration is defined as a set of operations that establishes, by reference to standards, the relationship that exists, under specified conditions, between an indication of an instrument under calibration and a result of a measurement using the corresponding traceable reference standard. Applied to the measuring receiver this means that a calibration procedure consisting of various steps is used to determine the actual values of calibration parameters like input VSWR or CW amplitude accuracy through measurements under specified environmental conditions, using measuring equipment that was calibrated by an accredited (or otherwise deemed appropriate) calibration laboratory to ensure traceability of the process. The results of these calibration measurements are used to determine if the instrument under calibration meets the specifications published by the manufacturer.

It is to be noted that the calibration process itself does not necessarily involve the instrument under calibration to be adjusted. However, adjustments may be required if the calibration process determines that the instrument does not meet the manufacturer’s specifications. The goal of the instrument calibration process is the determination of compliance of the measuring receiver under calibration with its published specifications in a traceable manner.

Furthermore, Verification should not be confused with intermediate checks (also sometimes called confidence checks or pre-checks); the latter consists of a set of operations aimed at providing evidence of the proper functioning of a test instrument. An intermediate check of a measuring receiver can differ considerably from the calibration process because the purpose of these two activities is entirely different.

Calibration and verification specifics

The calibration of a measuring receiver requires a specific process that defines the various measurements to determine if the receiver meets its specifications. In general, this calibration process has also been used by the receiver manufacturer to establish the receiver specifications. Therefore, only the manufacturer’s calibration process or verification process in accordance with CISPR 16-1-1 is to be applied by a calibration laboratory (or test laboratory performing its own calibrations) to determine whether the receiver meets its specifications at the time of calibration or the requirements called out in CISPR 16-1-1.

If a process different from the manufacturer’s calibration process or verification process in accordance with CISPR 16-1-1 is used, the applied process must be verifiably validated to demonstrate technical feasibility and it must be stated in the issued calibration certificate that the process used deviates from the calibration process defined by the manufacturer.

The calibration process for measuring receivers is very important since it defines the following essential parameters that must be used for proper calibration:

1. the specific set-up of the receiver under calibration for each measurement in the calibration process (e.g. in the case of an EMI receiver or spectrum analyzer the tuning frequency, attenuator setting, resolution bandwidth setting, and other parameters, for each measurement to be performed);
2. the required test set-up for the measurement of a specific parameter (e.g. the use of power splitters for ratio measurements and any other required measuring equipment);
3. the required accuracy of measuring equipment used to perform the measurements of the  calibration process (e.g. required amplitude accuracy and frequency accuracy);
4. the actual number of measurements to be performed and their sequence. For many types of measuring receivers this sequence is mandatory and cannot be changed because the measurements of some parameters require the measurements of previous calibration parameters to be completed. In addition, it is possible that the interpretation of a test result for a calibration parameter is dependent on the test result of a previous measurement in the calibration sequence;
5. the required environmental conditions (e.g. required ambient temperature and relative humidity), if deemed necessary by the manufacturer.

Only if the manufacturer’s calibration process is used can the results of the calibration measurements be compared to the published specifications. Consequently, the calibration laboratory or the test laboratory performing its own calibrations (also called internal calibrations) must use the manufacturer’s calibration process for a specific measuring receiver. As stated before, an alternative process must be validated to determine its technical feasibility as a calibration process its use must be documented in the calibration certificate to indicate that it deviates from the calibration process defined by the manufacturer.

Measuring receiver specifics

CISPR 16-1-1 specifies measuring receiver requirements using a black box approach. This means that the instrument must show a specific response when a defined signal is applied to its input.

Therefore, the demonstration of compliance of measuring receivers with specifications defined in CISPR 16-1-1 does not require the application of the manufacturer’s calibration process, and the procedures and measuring equipment defined in CISPR 16-1-1 are to be used. For example, the determination of intermodulation effects per 4.6 is to be performed using the test setup and input signals specified in the standard.

In case compliance of a measuring receiver is determined with the CISPR 16-1-1 specifications, the following minimum set of parameters shown in Table 1 are to be included in the verification process.

The parameters summarized in Table 1 are only applicable to the frequency ranges covered by the instrument under verification and its implemented detector functions. Specifics described in the referenced subclauses apply in their entirety as well as the stated  tolerances.

Table 1: Verification parameter summary

It is to be noted that the requirements called out in CISPR 16-1-1 constitute a subset of all the specifications the receiver manufacturer publishes. In addition, some requirements in CISPR 16-1-1 may be stated in a way that differs from the manufacturer’s specifications (e.g. CW frequency accuracy in CISPR 16-1-1 versus a combination of absolute amplitude accuracy at a reference frequency and frequency response).

If evidence of compliance with the requirements presented in CISPR 16-1-1 cannot be directly provided through the manufacturer’s calibration process, due to differences in form of the stated specifications, the verification of these requirements must be requested by the test laboratory in addition to the actual receiver calibration based on the manufacturer’s calibration process.

Partial calibration of measuring receivers

Often times the complete functionality of a measuring receiver is not utilized when performing emission measurements. For economic reasons test laboratories therefore may decide to purchase a calibration service only for those functions that are actually used to perform measurements. Care must be taken when specifying such a partial or limited calibration service because the calibration of the identified functions may require calibration of other functions as a prerequisite. Such dependencies must be determined by the test laboratory or the calibration laboratory through a review of the manufacturer’s calibration process. If the test laboratory does not have access to the manufacturer’s calibration procedure, this review must be requested from the calibration laboratory as part of the calibration service purchase.

Determination of compliance of a measuring receiver with applicable specifications

Compliance of a measuring receiver with the specifications of the manufacturer or with the tolerances specified in CISPR standards requires that measurement results reported in calibration certificates are below an upper limit, or above a lower limit, or between an upper and lower limit. The uncertainty of the calibration or verification measurement has a direct impact on the pass/fail determination. Therefore, the measurement uncertainty must be taken into account when determining compliance of a measuring receiver with its stated specifications. The application of measurement uncertainty to a measurement result can lead to one of the four cases described as follows and depicted in Figure 1:

1. the measurement result is within the specified limit range by a margin larger than the expanded uncertainty value applicable to the calibration measurement;
2. the measurement result is within the specified limit range by a margin less than the  expanded uncertainty value applicable to the calibration measurement;
3. the measurement result is outside of the specified limit range by a margin less than the  expanded
4. uncertainty value applicable to the calibration measurement; or
5. the measurement result is outside of the specified limit range by a margin larger than the  expanded uncertainty value applicable to the calibration measurement, and the  specification is not met.

Per CISPR 16-1-1 the four cases in Figure 1 should be interpreted as follows:

a) the specification is met;
b) and c) the result is inconclusive, a definitive compliance statement is not possible;
d) specification is not met.

Summary

Traceability and calibration requirements are also essential for EMC test laboratories. The interface between the test laboratory and external calibration laboratories can be complex, depending on the complexity of the equipment to be calibrated. Therefore, the EMC test laboratory is required to define the calibration requirements and communicate those to the calibration laboratory. Through the selection of proper calibration laboratories traceability of EMC measurement results is established. Since the calibration requirements of measuring receivers is complex, CISPR subcommittee A is in the process of preparing an annex to CISPR 16-1-1 that summarizes the calibration requirements for such instruments. This will allow the EMC test laboratories to easily identify the required calibration requirements in order to perform traceable emissions measurements. 

EMC test laboratories must also ensure that the provided calibration service is the one that was initially ordered. The step of an incoming inspection is performed upon receipt of the instrument back from the calibration laboratory and before the instrument is made available for measurements in the test laboratory. A thorough inspection will help avoid that improperly calibrated equipment or otherwise questionable calibration documentation causes non-conforming work situations later on which in turn can require considerable effort to determine the impact of such a situation on test results or can result in retesting of test samples.

Werner Schaefer is owner and Principal Engineer of Schaefer Associates. He has 29 years of EMC experience, including EMI test system and software design, EMI test method development and EMI standards development. He is the chairman of CISPR/A/WG1 and an active member of CISPR/A/WG2 and CISPR/B/WG1. He is an active member of the IEEE EMC Society.

He was actively involved in the development of the new standard ANSI C63.10 and the latest revision of ANSI C63.4, mainly focusing on test equipment specifications, use of spectrum analyzers and site validation procedures.

Werner Schaefer is also a RAB certified quality systems lead auditor, and an iNARTE certified EMC engineer. He published over 50 papers on EMC, RF/uwave and quality assurance topics, conducted numerous trainings and workshops on these topics and co-authored a book on RF/uwave measurements in Germany.