Electronics Guide

A test report is a claim about a measurement, and a measurement is only as good as the laboratory that made it. Accreditation is the mechanism the world uses to decide whether that claim deserves belief. It is a formal, third-party attestation that a laboratory is competent to perform specific tests or calibrations, granted after assessors have examined the laboratory's people, equipment, methods, records, and results against an international standard. For testing and calibration laboratories, that standard is ISO/IEC 17025.

Accreditation matters in electronics because almost nothing in the compliance chain works without it. A certification body will not accept a safety test report from a laboratory outside its accredited scope. The Federal Communications Commission will not accept certification test data from a laboratory that is not accredited to ISO/IEC 17025 and recognized by the Commission. A customer auditing a supplier's incoming inspection will ask who calibrated the instruments and to what traceability. When a product fails in the field and the dispute reaches a courtroom or an insurer, the accredited status of the laboratory that produced the original evidence becomes one of the first questions asked. Accreditation converts a private measurement into public evidence.

This article covers the accreditation system as a whole: what accreditation is and how it differs from certification, how the ISO/IEC 17000 family of conformity assessment standards fits together, what ISO/IEC 17025:2017 actually requires clause by clause, how metrological traceability and measurement uncertainty support every accredited result, how the scope of accreditation defines and limits what a laboratory may claim, how the assessment and surveillance cycle works, and how accreditation granted in one economy comes to be accepted in another. It closes with the practical questions an electronics manufacturer faces when buying accredited testing or building an in-house laboratory.

What Accreditation Is, and What It Is Not

Attestation, Accreditation, and Certification

ISO/IEC 17000 defines the vocabulary that the whole field uses, and the distinctions it draws are not pedantic. Conformity assessment is the demonstration that specified requirements are fulfilled. Attestation is the issue of a statement, based on a decision, that those requirements have been fulfilled. Certification is third-party attestation about a product, process, service, system, or person. Accreditation is third-party attestation about a conformity assessment body itself, conveying formal demonstration of that body's competence, impartiality, and consistent operation.

The practical consequence is a hierarchy. A product is certified. A laboratory is accredited. A management system is certified against ISO 9001 by a certification body, and that certification body is in turn accredited. Saying that a laboratory is "ISO/IEC 17025 certified" is a category error, and an informed customer treats it as a warning sign: it usually means the laboratory has purchased a consultant's declaration or a certificate from an organization that is not an accreditation body, rather than undergoing peer assessment by a recognized one.

A related distinction separates accreditation from compliance. A laboratory may implement ISO/IEC 17025 fully and never seek accreditation. Nothing prevents it, and for internal development work the discipline alone may deliver most of the benefit. What such a laboratory cannot do is issue reports bearing an accreditation symbol or claim accredited status, and its data will generally not be accepted where accreditation is a stated requirement. Accreditation buys external recognition, not internal quality; the internal quality is a prerequisite for obtaining it.

Why the Model Exists

The alternative to accreditation is bilateral assessment: every purchaser of test data audits every laboratory that produces it. That model still exists in pockets of aerospace, automotive, and defense work, and its cost explains why the accreditation model displaced it almost everywhere else. Accreditation is an economic device for pooling the assessment burden. One competent assessment body examines the laboratory once on a defined cycle, publishes the result, and every purchaser relies on that single assessment instead of performing its own.

For that pooling to work, the assessment must mean the same thing everywhere, which is why accreditation bodies are themselves standardized under ISO/IEC 17011 and peer-evaluated by their international cooperation. The chain of trust runs upward from the measurement to the laboratory, from the laboratory to the accreditation body, and from the accreditation body to a multilateral arrangement whose signatories have evaluated one another. Break any link and the report at the bottom loses its portability.

What Accreditation Does Not Guarantee

Accreditation attests to competence in performing defined activities. It does not certify that any particular result is correct, that the laboratory is fast or affordable, that its engineers give useful design advice, or that a product tested there will pass. It also does not extend beyond the accredited scope, which is the single most frequently misunderstood point in the entire system and the subject of a dedicated section below. A laboratory holding an impeccable accreditation for radiated emissions has demonstrated nothing about its ability to run a dielectric strength test, and a report it issues on the latter carries no accredited status whatsoever.

The Conformity Assessment Standards Family

The CASCO Toolbox

The standards that govern conformity assessment are developed by the ISO Committee on Conformity Assessment, known as CASCO, jointly with the IEC. Collectively they are called the CASCO toolbox, and they share a deliberate common architecture: general requirements covering impartiality and confidentiality, structural requirements covering organization and governance, resource requirements covering people and equipment, process requirements covering the activity itself, and management system requirements. Once the pattern is familiar in one standard, the others read quickly.

The toolbox is organized by the activity performed rather than by industry. The question a laboratory or body must answer first is not what it tests but what kind of conformity assessment activity it carries out, because that determines which standard applies.

The Accreditation Chain

At the top of the chain sits ISO/IEC 17011:2017, which specifies requirements for accreditation bodies accrediting conformity assessment bodies. It governs the accreditation bodies themselves: their impartiality, their competence requirements for assessors, the conduct of assessments, the accreditation decision process, the duration of the accreditation cycle, and the handling of appeals and complaints. Accreditation bodies are evaluated against ISO/IEC 17011 by their peers, which is the mechanism that makes international recognition possible.

Below it sit the standards for the bodies being accredited, one for each activity:

ISO/IEC 17025 applies to testing and calibration laboratories, including laboratories that perform sampling associated with subsequent testing or calibration. It is the standard behind nearly every EMC chamber, safety laboratory, environmental test house, and calibration laboratory that serves the electronics industry.

ISO/IEC 17020 applies to bodies performing inspection, an activity defined as examination of a product, process, service, or installation and determination of its conformity, often involving professional judgment. Factory inspection under a product certification scheme, electrical installation inspection, and pressure equipment inspection all fall here. The long-standing 2012 edition was replaced by ISO/IEC 17020:2026, published on 27 March 2026, which strengthens the treatment of risk and impartiality, addresses digital inspection management, and simplifies the historic Type A, Type B, and Type C independence classification into a distinction between Type A and non-Type A bodies. Accreditation bodies set their own transition timetables against it, most running about three years from publication: the Singapore Accreditation Council allows until 27 March 2029, while UKAS makes assessment to the new edition mandatory from 1 January 2028. Both editions will therefore be encountered in scope documents for several years.

ISO/IEC 17021-1:2015 applies to bodies providing audit and certification of management systems. A registrar issuing ISO 9001, ISO 14001, or ISO/IEC 27001 certificates works to this standard, supported by sector-specific parts and by ISO/IEC TS 17021 documents that define auditor competence for particular disciplines.

ISO/IEC 17065:2012 applies to bodies certifying products, processes, and services. This is the standard behind product certification marks. When a certification body issues a listing, grants the right to apply a mark, and conducts follow-up factory surveillance, it does so under an ISO/IEC 17065 accreditation, and it relies on ISO/IEC 17025-accredited testing to generate the evidence. The relationship between the two standards is the structural reason that laboratory accreditation and product certification are always discussed together.

Two further standards complete the picture for personnel and validation activities. ISO/IEC 17024 covers bodies operating certification of persons, the route by which welders, nondestructive testing technicians, and functional safety engineers obtain recognized personal credentials. ISO/IEC 17029:2019 covers validation and verification bodies, the framework used for greenhouse gas assertions and other declared claims.

Supporting Standards Laboratories Depend On

ISO/IEC 17043:2023 specifies requirements for the competence of providers of proficiency testing schemes. Because ISO/IEC 17025 obliges laboratories to monitor the validity of their results and because accreditation bodies require participation in proficiency testing where it is available and appropriate, the accredited status of the scheme provider matters directly to the laboratory that participates.

ISO 17034:2016 specifies general requirements for the competence of reference material producers. Certified reference materials supply the assigned values against which many measurements are anchored, and a certified reference material from an accredited producer carries a documented uncertainty and traceability statement that an uncertified material does not.

Several guides underpin the measurement mathematics, all of them produced by the Joint Committee for Guides in Metrology and issued in parallel as JCGM documents. ISO/IEC Guide 98-3:2008, commonly called the GUM, defines the framework for evaluating and expressing measurement uncertainty, and ISO/IEC Guide 98-1:2024 was added as an introduction to it. ISO/IEC Guide 98-4 extends the framework to the role of uncertainty in conformity assessment, which is where decision rules and guard bands come from. ISO/IEC Guide 99, the International Vocabulary of Metrology or VIM, fixes the terminology, including the definition of metrological traceability that ISO/IEC 17025 invokes.

Inside ISO/IEC 17025:2017

Structure of the Standard

ISO/IEC 17025:2017 is the third edition, published in November 2017. It replaced ISO/IEC 17025:2005 after a transition period that ILAC originally set to end on 30 November 2020 and then extended to 1 June 2021 in response to the assessment disruption caused by the COVID-19 pandemic. Since that date, accreditation to the 2005 edition no longer exists, and any scope document or report still referencing it is out of date.

The standard is short by the standards of its own subject matter, and its requirements sit in clauses 4 through 8. Clause 1 states the scope, clause 2 lists normative references, and clause 3 gives terms and definitions. The five requirement clauses divide as general requirements, structural requirements, resource requirements, process requirements, and management system requirements. Everything a laboratory must do falls into one of those five buckets, and assessment findings are written against their subclause numbers, so fluency in the numbering is a working skill rather than an academic one.

Clause 4: Impartiality and Confidentiality

Clause 4 contains only two requirements, and both cut to the heart of what accreditation is worth. Impartiality requires that laboratory activities be undertaken impartially and be structured and managed to safeguard impartiality. Laboratory management must commit to impartiality, and the laboratory must identify risks to its impartiality on an ongoing basis, including risks arising from its relationships, and eliminate or minimize them. The word "ongoing" carries weight: a one-time impartiality analysis filed at accreditation and never revisited is a standard finding.

The risks are real and specific in electronics. A laboratory owned by the company whose products it tests faces obvious pressure. So does an independent laboratory whose largest customer accounts for most of its revenue, or one that sells design consulting to the same client whose product it later certifies as compliant. The standard does not forbid these arrangements. It requires that the laboratory identify them and demonstrate the safeguards that prevent commercial, financial, or other pressure from compromising results.

Confidentiality requires the laboratory to be responsible, through legally enforceable commitments, for the management of all information obtained or created during laboratory activities. This matters commercially, because a compliance test campaign exposes schematics, bills of materials, firmware behavior, and unreleased product plans. The standard also requires that when the laboratory is compelled by law or authorized by contract to release confidential information, the customer be notified unless prohibited by law.

Clause 5: Structural Requirements

Clause 5 fixes the laboratory as an identifiable legal entity or a defined part of one, legally responsible for its laboratory activities. It requires identification of management with overall responsibility, definition of the range of activities for which the laboratory claims conformity, documentation of procedures to the extent necessary for consistency, and identification of personnel who have the authority and resources to carry out their duties. It also requires that the laboratory have personnel who, irrespective of other responsibilities, have the authority and resources needed to implement, maintain, and improve the management system and to identify departures from it.

The clause is brief, but it settles a question that arises constantly in corporate laboratories: where the laboratory ends and the rest of the organization begins. An in-house test facility inside a manufacturer must define its boundary, its reporting line, and the activities inside its declared scope of conformity, because accreditation attaches to that defined entity and not to the parent company.

Clause 6: Resource Requirements

Clause 6 covers everything the laboratory needs in order to perform work: personnel, facilities and environmental conditions, equipment, metrological traceability, and externally provided products and services.

Personnel requirements center on competence rather than credentials. The laboratory must document competence requirements for each function, ensure personnel have the competence to perform the activities for which they are responsible, and authorize personnel to perform specific laboratory activities. Records of competence, training, and authorization must be maintained. In practice this means a named authorization matrix that maps each person to each method they may run, sign, or review, kept current as staff turn over.

Facilities and environmental conditions must be suitable and must not adversely affect the validity of results. Where the environment influences results, the laboratory must monitor, control, and record the relevant conditions. Electronics laboratories carry unusually demanding versions of this requirement: a calibration laboratory holds temperature and humidity within narrow bands and records them continuously, a semi-anechoic chamber depends on validated site attenuation and controlled ambient radio-frequency levels, and an electrostatic discharge test area needs controlled humidity because the discharge waveform itself depends on it.

Equipment requirements cover access, handling, transport, storage, use, and planned maintenance. Equipment must be verified as conforming to specified requirements before being placed into service, calibrated where measurement accuracy or uncertainty affects the validity of results, uniquely identified, and labeled or otherwise marked with its calibration status. Records must include identification, manufacturer and serial number, verification evidence, calibration dates and results, due dates, and details of any damage, malfunction, modification, or repair. Where intermediate checks are needed to maintain confidence, the laboratory must define and perform them, and it must have a documented procedure for handling equipment found to be defective, including examination of the effect on previously issued results.

Metrological traceability is treated separately below because of its weight. Externally provided products and services closes the clause: the laboratory must ensure that externally provided reagents, consumables, calibration services, and subcontracted testing are suitable, must define requirements for them, must evaluate and monitor providers, and must retain records of that evaluation. Subcontracting a portion of a test campaign is permitted, but the customer must be informed and the arrangement documented.

Clause 7: Process Requirements

Clause 7 is the operational core of the standard and the longest clause. It follows the life of a job from inquiry to report.

Review of requests, tenders, and contracts requires the laboratory to confirm before starting work that requirements are defined and understood, that it has the capability and resources to meet them, that appropriate methods are selected, and that any decision rule to be applied is defined and agreed with the customer. Differences resolved between the laboratory and the customer must be recorded, and the customer must be informed of any deviation from the contract.

Selection, verification, and validation of methods requires the use of appropriate methods, preferably those published in international, regional, or national standards. The laboratory must verify that it can properly perform a standard method before introducing it, by demonstrating the required performance. Laboratory-developed or modified methods must be validated, with the validation extent matched to the need, and records must include the validation procedure, the requirements, the performance characteristics obtained, and a statement of validity for the intended use. When the laboratory needs to deviate from a method, the deviation must be documented, technically justified, authorized, and accepted by the customer.

Sampling applies where the laboratory takes samples of substances, materials, or products for subsequent testing. Handling of test and calibration items requires procedures for transportation, receipt, handling, protection, storage, retention, and disposal, along with a system for unambiguous identification and recording of any abnormalities or departures from specified conditions on receipt.

Technical records must contain the results, the report, and sufficient information to facilitate identification of factors affecting the result and its uncertainty, and to enable repetition of the activity under conditions as close as possible to the original. They must include the date and the identity of the personnel responsible, and amendments must be traceable to previous versions with original data retained. Ensuring the validity of results requires a procedure for monitoring validity, using tools such as quality control samples, checks with reference standards, intermediate checks, replicate testing, retesting of retained items, correlation of results for different characteristics, review of reported results, interlaboratory comparison, and proficiency testing. The data must be analyzed and, where results fall outside predefined criteria, action must be taken to prevent incorrect results from being reported.

Evaluation of measurement uncertainty is required for testing and calibration laboratories alike, with testing laboratories permitted to base the evaluation on an understanding of theoretical principles or practical experience where the test method precludes rigorous evaluation. Reporting of results requires that results be reviewed and authorized before release and be provided accurately, clearly, unambiguously, and objectively. The standard enumerates the minimum contents of a test report and a calibration certificate, including identification of the laboratory and the customer, identification of the method, the date of receipt and of performance, the results with units, and identification of the person authorizing the report. Where uncertainty is relevant to the validity or application of results, or where the customer instructs, or where uncertainty affects conformity to a limit, the uncertainty must be reported.

Statements of conformity carry a specific obligation. When the laboratory states conformity to a specification or standard, it must document the decision rule employed, taking into account the level of risk associated with that rule, and must apply and report it. This requirement, introduced in the 2017 edition, changed daily practice in calibration laboratories more than any other single change.

The clause closes with complaints, nonconforming work, and control of data and information management. Nonconforming work requires evaluation of significance, a decision on acceptability, action on affected results including recall where necessary, and definition of responsibility for authorizing resumption of work. Laboratory information management systems must be validated for functionality before introduction, protected against unauthorized access and tampering, and maintained, and changes must be authorized and validated before implementation.

Clause 8: Management System Options A and B

Clause 8 gives the laboratory a choice. Option A requires the laboratory to implement a management system addressing, as a minimum, the specific requirements listed in clause 8 itself: management system documentation, control of management system documents, control of records, actions to address risks and opportunities, improvement, corrective action, internal audits, and management reviews. Option B is available to a laboratory that has established and maintains a management system in accordance with ISO 9001 and is capable of supporting and demonstrating the consistent fulfillment of clauses 4 through 7; such a laboratory is deemed to fulfill at least the intent of the clause 8 requirements.

Option B is not a shortcut. The ISO 9001 system must actually cover the laboratory's activities and must demonstrably support the technical clauses. Most standalone laboratories choose Option A because it is self-contained; laboratories embedded in a manufacturer that already holds ISO 9001 certification often choose Option B to avoid maintaining two parallel systems. Either way, the assessment covers the same ground.

One characteristic of the 2017 edition deserves note: it replaced the prescriptive "preventive action" of the 2005 edition with actions to address risks and opportunities, aligning the standard with the risk-based thinking adopted across the ISO management system standards. The laboratory must plan actions proportionate to the potential impact on the validity of results, but the standard deliberately does not require a formal risk management method or documented risk register. Assessors look for evidence that risk thinking informs decisions, not for a particular artifact.

What Changed from the 2005 Edition

Laboratories and purchasers still encounter documents shaped by the older edition, so the differences are worth knowing. The 2017 edition adopted a process-based structure aligned with the other CASCO standards, replacing the 2005 division into management requirements and technical requirements. It introduced the explicit decision rule requirement for statements of conformity. It replaced preventive action with risk-based thinking. It broadened and modernized the treatment of information technology, recognizing that laboratory data now lives in information management systems rather than paper notebooks. It shifted the vocabulary of personnel requirements from qualification toward demonstrated competence, and it dropped the mandatory quality manual and mandatory quality manager role in favor of requirements expressed as outcomes.

Metrological Traceability and Measurement Uncertainty

The Traceability Chain

Metrological traceability, in the VIM definition that ISO/IEC 17025 adopts, is the 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. Every word in that definition is load-bearing. The chain must be unbroken, each link must be documented, and each link must contribute a stated uncertainty. A calibration sticker with no uncertainty statement does not establish traceability.

ISO/IEC 17025 requires laboratories to establish and maintain traceability of their measurement results to the International System of Units, through calibration by a competent laboratory or through certified values of certified reference materials, or, where neither is possible, through other means that provide confidence in the result, such as consensus standards or agreed methods. Competence at the calibrating link is normally demonstrated by that laboratory's own ISO/IEC 17025 accreditation, by the calibration and measurement capabilities of a national metrology institute published in the BIPM key comparison database under the CIPM Mutual Recognition Arrangement, or by an equivalent recognized route. ILAC-P10, the policy on metrological traceability of measurement results, sets out the acceptable routes in detail, and accreditation bodies enforce it directly.

In an electronics laboratory the chain has a familiar shape. A national metrology institute maintains primary realizations of the volt through the Josephson effect and of resistance through the quantum Hall effect. An accredited calibration laboratory maintains reference standards calibrated against those realizations, and calibrates the laboratory's working standards, which in turn calibrate or verify the bench instruments: the multimeters, oscilloscopes, spectrum analyzers, power meters, signal generators, LCR meters, and hipot testers that produce the numbers appearing in reports. Uncertainty accumulates at each step, which is why the reference standard at the top of a laboratory's own chain must be substantially better than the measurement it ultimately supports.

Evaluating Uncertainty

Measurement uncertainty quantifies the dispersion of values that could reasonably be attributed to the measurand. The GUM framework classifies the evaluation of individual contributions into Type A, obtained by statistical analysis of a series of observations, and Type B, obtained by other means such as calibration certificates, manufacturer specifications, published data, or judgment based on experience. The classification refers to the method of evaluation, not to the nature of the contribution, and a common error is to treat Type B contributions as somehow less rigorous.

The contributions are combined into a combined standard uncertainty, normally by root-sum-square of the standard uncertainties weighted by their sensitivity coefficients, then multiplied by a coverage factor to give an expanded uncertainty. Accredited calibration certificates conventionally report expanded uncertainty with a coverage factor of k = 2, corresponding to a coverage probability of approximately 95 percent for a normal distribution, and the certificate must state the coverage factor used. ILAC-P14, the policy on uncertainty in calibration, governs how accredited calibration laboratories express these values and how their best measurement capability appears on scopes of accreditation.

Testing laboratories face a softer requirement than calibration laboratories, because many electronics test methods do not permit rigorous metrologically valid uncertainty evaluation. Radiated emissions measurement is the standard example: CISPR 16-4-2 provides the uncertainty framework for the discipline, defining measurement instrumentation uncertainty for each measurand, and compliance standards define how it is treated when comparing a result against a limit. The laboratory must still identify the contributions and make a reasonable estimate, and assessors will examine whether the estimate matches the equipment and procedure actually used.

Decision Rules and Guard Banding

A decision rule describes how measurement uncertainty is accounted for when stating conformity with a specified requirement. Since the 2017 edition made it mandatory to document, agree, and report the rule whenever a statement of conformity is issued, decision rules have become one of the most-discussed topics in accredited testing. ILAC-G8:09/2019, the guidelines on decision rules and statements of conformity, is the reference document; it describes the available approaches, adds a decision flowchart for selecting among them, and gives worked examples.

Simple acceptance, sometimes called shared risk, compares the measured value directly to the tolerance limit and ignores uncertainty in the decision. It is the least conservative rule and is only appropriate where the uncertainty is small relative to the tolerance and where the customer has explicitly accepted the associated risk. Guarded acceptance shifts the acceptance limit inward from the tolerance limit by a guard band, typically some multiple of the expanded uncertainty, so that a pass statement carries a controlled maximum probability of false acceptance. Guarded rejection shifts the limit the other way, reducing the probability of false rejection at the cost of accepting more marginal items. Where the measured value lands inside the guard band, the laboratory may report the result with no conformity statement, or with a conditional statement, according to the agreed rule.

The commercial consequence is direct. A tighter guard band means more units fail. If a manufacturer specifies simple acceptance and a competitor's laboratory applies a 95 percent guarded acceptance rule to the same product against the same limit, the two laboratories can reach opposite conclusions from identical measurements without either being wrong. This is precisely why the standard requires the rule to be agreed in advance and reported, and why contract review is the moment to settle it.

The Scope of Accreditation

What a Scope Document Contains

Accreditation is granted for defined activities, and the scope of accreditation is the authoritative statement of those activities. Accreditation bodies publish scopes in searchable directories, and the published document, not a laboratory's marketing material, is the thing to read. A well-formed scope for an electronics testing laboratory lists, for each entry, the field of testing, the specific test method or standard with its edition, the range or parameters covered, and any limitations. A calibration laboratory's scope adds the measurement range and the calibration and measurement capability, expressed as the smallest expanded uncertainty the laboratory is accredited to claim for that measurement.

Reading a scope well means reading the details. An entry for CISPR 32 tells you the laboratory is accredited for multimedia equipment emissions; it does not tell you the laboratory can perform the immunity tests of CISPR 35. An entry citing a superseded edition of a standard means the laboratory has not yet extended its accreditation to the current edition, which will matter if the certification body or regulator has moved on. A frequency range that stops at 6 GHz means the laboratory cannot support a device whose highest internally generated frequency requires measurement to 40 GHz. A calibration and measurement capability of 5 parts per million cannot support a measurement that needs 1 part per million.

Fixed and Flexible Scopes

Most scopes are fixed: each method is listed explicitly, and adding a method requires an extension to scope, generally involving assessment of the new capability before the laboratory may issue accredited reports under it. Accreditation bodies also grant flexible scopes to laboratories that have demonstrated the competence to develop, modify, or adopt methods within a defined technical field without prior assessment of each one. A flexible scope suits laboratories working in fast-moving fields or performing frequent method modifications, and it carries a corresponding obligation: the laboratory must maintain an internal, current list of the methods it operates under the flexible scope and demonstrate at each assessment that its method validation and authorization processes control the freedom it has been granted.

Extensions to scope are routine and can often be handled alongside a scheduled surveillance visit, but they are not instantaneous. A laboratory adding a new standard to its accreditation typically needs validated procedures, trained and authorized personnel, suitable and calibrated equipment, an uncertainty evaluation, and evidence of satisfactory results, and then an assessor must review or witness the work. Planning a product launch around a laboratory capability that is still awaiting an accreditation extension is a common scheduling mistake.

Scope Traps in Electronics Testing

Several scope problems recur often enough in electronics compliance work to deserve naming. The first is the assumption that accreditation for a product-family safety standard covers all of its clauses; some laboratories are accredited for a subset, excluding, for example, the glow-wire or needle-flame tests that require specific apparatus. The second is the split between emissions and immunity, which are separate accreditations resting on entirely different equipment. The third is the difference between accreditation for a test method and designation or recognition under a regulatory scheme. In the United States, a laboratory producing data for an FCC equipment certification must be accredited to ISO/IEC 17025 by an accreditation body the Commission recognizes, and must itself be listed by the Commission; 47 CFR 2.948 sets the measurement facility requirement, 47 CFR 2.949 governs recognition of the accreditation bodies themselves, and the FCC KDB 974614 publications set out the accompanying roles and responsibilities. A laboratory can hold a perfectly valid ISO/IEC 17025 accreditation for radio testing and still not be an FCC-recognized laboratory. The fourth is the difference between accreditation and OSHA's Nationally Recognized Testing Laboratory program, which rests on recognition under 29 CFR 1910.7 rather than on accreditation; the two systems overlap in practice but are not interchangeable.

The remedy in every case is the same: obtain the scope document, find the exact standard, edition, clause set, and parameter range the campaign requires, and confirm the laboratory is also designated or recognized under whatever scheme will consume the data. Doing this before booking chamber time costs an hour. Doing it after a report is rejected costs a test campaign.

Obtaining and Maintaining Accreditation

Choosing an Accreditation Body

In much of the world the choice is made for the laboratory. Regulation (EC) No 765/2008 requires each European Union member state to appoint a single national accreditation body operating as a public authority on a non-profit basis, which is why laboratories in Germany go to DAkkS, in France to COFRAC, and in the United Kingdom to UKAS. Australia has NATA, India has NABL, China has CNAS, and Canada works through the Standards Council of Canada and its recognized bodies.

The United States is the notable exception, with several accreditation bodies competing: A2LA, ANAB, NVLAP within NIST, IAS, and PJLA among them. NVLAP operates program-specific handbooks in the NIST Handbook 150 series, including handbooks addressing EMC and telecommunications testing and the FCC's requirements. Where a choice exists, the criteria are practical: whether the body's own recognitions cover the schemes the laboratory's customers care about, whether it has technical assessors with genuine experience in the laboratory's discipline, the fee structure, and the availability of assessors within a workable schedule. An assessor who has never run a semi-anechoic chamber will not conduct a useful assessment of one.

Application, Document Review, and Assessment

The process begins with an application describing the laboratory's organization, personnel, facilities, equipment, and the scope sought. The accreditation body reviews the application, confirms it can supply competent assessors for the scope, and quotes the assessment. A document review follows, in which assessors examine the management system documentation, method procedures, uncertainty budgets, and records against the standard. Deficiencies found at this stage are cheaper to fix than those found on site.

The on-site assessment combines a system audit with technical evaluation. A lead assessor examines the management system: document and record control, internal audits, management review, corrective action, complaints, nonconforming work, impartiality risk analysis, and personnel authorization. Technical assessors, who are practitioners in the field, examine methods, equipment, calibration and traceability records, uncertainty budgets, and, critically, witness the laboratory performing tests. Witnessing is where paper systems meet practice, and it is where problems most often surface: an operator who deviates from the written procedure, a fixture that does not match the method drawing, a chamber setup that fails the standard's positioning requirements, or an instrument whose calibration does not cover the range in use.

Findings are documented and classified by severity. The laboratory must respond with root cause analysis, correction, and corrective action, supported by objective evidence. The accreditation body verifies the responses, sometimes by document review and sometimes by a return visit for significant findings. The accreditation decision is made by a person or panel independent of the assessment team, which is an ISO/IEC 17011 requirement designed to separate assessment from decision.

Surveillance and Reassessment

Accreditation runs on a cycle. ISO/IEC 17011 caps the accreditation cycle at five years, and accreditation bodies commonly operate a four-year or five-year cycle with surveillance visits inside it, the first typically within a year of the initial decision and further visits roughly annually thereafter. Surveillance is narrower than a full assessment, sampling parts of the system and witnessing a subset of methods, while reassessment at the end of the cycle covers the entire scope again.

Between visits the laboratory carries continuing obligations. It must notify the accreditation body of significant changes, including changes of key personnel, location, ownership, or major equipment, and of anything affecting its capability or the validity of its accreditation. It must participate in proficiency testing according to a documented plan. It must run internal audits and management reviews on its own schedule. Accreditation may be suspended, reduced in scope, or withdrawn where the laboratory fails to maintain the requirements, and suspension takes effect on named parts of the scope, not necessarily on the whole.

Time and Cost

A laboratory starting from an established, disciplined operation and seeking a modest scope will typically spend somewhere between six months and a year from decision to accreditation certificate, with the elapsed time dominated by building method validation and uncertainty evidence, generating enough records to audit, and waiting for assessor availability. A laboratory starting from nothing should expect longer. The direct fees paid to the accreditation body are usually a minor part of the total: the substantial costs are staff time to build and run the management system, calibration of the reference standards and instruments that support the scope, proficiency testing participation, and the ongoing labor of internal audit, management review, and assessment preparation. Laboratories that treat accreditation as a documentation exercise underestimate all of these.

Proficiency Testing and Interlaboratory Comparison

Why Participation Is Required

Internal quality control confirms that a laboratory is consistent with itself. It cannot detect a systematic error shared by every measurement the laboratory makes, such as a misapplied correction factor, a reference standard drifting undetected, or a method interpretation that differs from everyone else's. Proficiency testing, in which an accredited provider distributes items or artifacts and compares participants' results against an assigned value, is the only routine mechanism that finds such errors. ISO/IEC 17025 requires monitoring the validity of results and names interlaboratory comparison and proficiency testing among the available tools, and accreditation bodies require documented participation where suitable schemes exist for the accredited scope.

Programs Available in Electronics

Availability is uneven across the discipline. Electrical calibration is well served, with round-robin artifacts circulating for direct-current voltage, resistance, capacitance, alternating-current measurements, oscilloscope parameters, radio-frequency power, and attenuation. EMC has established comparison programs in which a stable comparison source or a characterized device is circulated among chambers and the measured emission profiles compared, which detects site anomalies that routine site validation can miss. Safety testing is harder to serve with a circulated artifact, since much of the test program is destructive or depends on construction assessment, so laboratories often rely on interlaboratory comparison of specific measurable quantities, on witnessed testing, and on internal replicate work.

Where no suitable proficiency testing scheme exists, the laboratory must define an alternative and defend it to the assessor. Acceptable alternatives include bilateral comparison with another accredited laboratory, measurement of a certified reference material or characterized artifact, replicate testing by different personnel and equipment, and retesting of retained items. What is not acceptable is doing nothing and citing unavailability without having searched for a scheme.

Responding to Poor Results

Proficiency testing results are conventionally scored, most often as a z-score comparing the participant's deviation from the assigned value against a target standard deviation. Scores within two are usually treated as satisfactory, scores between two and three as questionable, and scores beyond three as unsatisfactory. An unsatisfactory result is not a disqualification, but it triggers obligations: investigation of root cause, evaluation of the effect on previously reported results including the possible recall of reports, corrective action, and demonstration of effectiveness, often through participation in the next available round. Accreditation bodies review how a laboratory handled poor results as closely as they review the results themselves, because the response reveals whether the quality system actually functions.

International Recognition

The Mutual Recognition Arrangement

Accreditation would be a local currency if accreditation bodies did not recognize one another. The mechanism is a multilateral arrangement whose signatories evaluate each other against ISO/IEC 17011 through structured peer evaluation and agree to recognize the equivalence of each other's accreditations. The practical effect for a manufacturer is that a test report from an accredited laboratory in one signatory economy is accepted in another without reassessing the laboratory. That single sentence carries an enormous amount of trade.

The arrangement has recently been restructured. The International Laboratory Accreditation Cooperation, which administered mutual recognition for testing, calibration, inspection, and proficiency testing, and the International Accreditation Forum, which administered multilateral recognition for certification bodies, consolidated into a single organization, Global Accreditation Cooperation Incorporated which commenced full operations on 1 January 2026 together with its own Multilateral Recognition Arrangement, the Multilateral Recognition Arrangement. That single arrangement succeeds both the ILAC Mutual Recognition Arrangement and the IAF Multilateral Recognition Arrangement. Recognition granted under the predecessor arrangements continues without interruption, and accredited results issued before and during the transition remain valid. In practice, laboratory scopes, accreditation certificates, policy documents, and customer specifications will carry the legacy ILAC and IAF names for years, and the guidance documents cited throughout this article keep their original designations, ILAC-P10, ILAC-P14, and ILAC-G8 among them, as do the IAF mandatory documents. Both sets of names should be read as referring to the same chain of trust.

Regional Cooperations

Regional bodies sit between national accreditation bodies and the global arrangement, conducting peer evaluation within their region and feeding into global recognition. European co-operation for Accreditation covers Europe, the Asia Pacific Accreditation Cooperation covers the Asia-Pacific region, the Inter American Accreditation Cooperation covers the Americas, and the African Accreditation Cooperation and Arab Accreditation Cooperation cover their respective regions. These bodies also publish regional guidance that laboratories in their territory must follow alongside the international documents.

Where Recognition Stops

Mutual recognition of accreditation is not the same as regulatory acceptance, and conflating the two produces expensive surprises. Recognition means an accreditation body's competence judgments are accepted as equivalent. It does not oblige any government to accept a foreign test report for a regulated approval. Regulatory acceptance depends on separate instruments: government-to-government mutual recognition agreements, which typically designate specific conformity assessment bodies for specific regulations; scheme-level arrangements such as the IECEE CB Scheme, which operates its own recognition of testing laboratories and certification bodies; and unilateral national rules such as the FCC's recognition of accreditation bodies and listing of laboratories, or a national requirement that testing be performed in-country.

The working rule is to ask two questions about every laboratory, not one. Is it accredited for exactly the method required? And is it accepted by the specific body that will consume the report, whether that is a certification body, a notified body, a regulator, or a customer's quality organization? A yes to the first and a no to the second still leaves the report unusable.

Accreditation in Electronics Practice

EMC and Radio Laboratories

EMC and radio laboratories carry the heaviest facility burden in the field, because their accredited scope depends on the validated performance of the test site itself and not only on instruments. Semi-anechoic chambers, fully anechoic rooms, open area test sites, and reverberation chambers are validated against the requirements of the CISPR 16-1-4 site validation criteria, and that validation is part of what an assessor examines. Antenna calibrations, cable and attenuator characterizations, receiver and analyzer calibrations, and the calibration of line impedance stabilization networks all feed the traceability chain, and each contributes to the uncertainty budget that CISPR 16-4-2 frames. A laboratory's scope in this discipline is therefore unusually detailed, listing standards, frequency ranges, and often the specific chamber used.

Safety Testing Laboratories

Safety testing laboratories mix measurement with judgment. Dielectric strength, leakage current, temperature rise, and ground continuity produce numbers, but construction review against creepage and clearance tables, evaluation of insulation systems, and assessment of component suitability rest on the assessor's confidence in the engineer's competence. This is why personnel competence records carry so much weight in safety laboratory assessments, and why witnessing tends to include a review of how the laboratory reaches construction conclusions, not only how it operates instruments. Many safety laboratories operate inside or alongside a certification body accredited to ISO/IEC 17065, and the accreditation of the two functions is separate even where the staff and building are shared.

Manufacturer In-House Laboratories

A manufacturer's internal laboratory can pursue three distinct levels of standing. It may operate to good practice without accreditation, generating data for design decisions and pre-compliance screening only. It may seek ISO/IEC 17025 accreditation in its own right, which requires it to satisfy the same impartiality and competence requirements as an independent laboratory, with particular attention to the safeguards that insulate the laboratory from pressure by product development. Or it may seek recognition under a certification body's scheme for manufacturer testing, in which the certification body qualifies the laboratory, defines the procedures and reporting, and supervises the work, sometimes through witnessed testing at the manufacturer's site.

The economics favor accreditation only where volume justifies it. The cost of maintaining an accredited scope, including calibration, proficiency testing, internal audit, and assessment, is fixed and continuing, while the cost of buying accredited testing is proportional to campaigns run. Manufacturers with steady, high-volume compliance work in a narrow set of methods, particularly EMC pre-compliance that graduates to accredited final testing, are the usual candidates. Everyone else is better served by disciplined internal pre-compliance combined with purchased accredited testing.

Calibration Support for the Test Floor

Every accredited testing laboratory depends on calibration, and the quality of that support is a frequent weak point. The requirement is not that every instrument be calibrated by an accredited laboratory, but that calibration establishes traceability with stated uncertainty and that the laboratory can justify its choices. Where the calibration affects the validity of a reported result, accredited calibration is the defensible route, and the calibration laboratory's scope must cover the parameter and range and provide a capability adequate for the measurement. Reviewing a calibration certificate on receipt, rather than filing it, catches the common failures: a range that does not cover the working range, an as-found condition out of tolerance that requires evaluation of previously issued results, a missing uncertainty statement, or a statement of conformity issued under an undisclosed decision rule.

Common Assessment Findings

Records That Cannot Reconstruct the Work

The most common category of finding concerns technical records. The standard requires enough information to identify factors affecting the result and its uncertainty and to permit repetition of the activity under conditions as close as possible to the original. Records that give a result but not the equipment used, the setup configuration, the environmental conditions, the software version, or the identity of the operator fail that test. In EMC work, records that omit the cable routing, the support equipment, the operating mode of the device under test, or the chamber position make a repeat measurement impossible and are a reliable source of findings.

Uncertainty Budgets Detached from Practice

Uncertainty budgets are frequently written once, at accreditation, and then never revised. Assessors compare the budget to what the laboratory actually does, and findings follow when the budget cites an instrument that has been replaced, a repeatability figure derived from a study never repeated, a coverage factor inconsistent with the reported values, or contributions that do not include a source obviously present in the setup. A budget is a living document tied to a configuration, and a change of instrument or method is a trigger to revisit it.

Method Deviation Without Authorization

Laboratories deviate from published methods for practical reasons: an item too large for the specified fixture, a step that the equipment cannot perform as written, an ambiguity in the standard resolved by local convention. The standard permits deviation, but only when it is documented, technically justified, authorized, and accepted by the customer. Undocumented deviation, discovered during witnessing when the operator does something the procedure does not describe, is among the more serious findings because it goes directly to the validity of reported results.

Impartiality Treated as a Formality

Impartiality findings tend to arise where the laboratory has written a policy statement and stopped. Assessors look for a current analysis of actual relationships and pressures, and for evidence that identified risks led to concrete safeguards: reporting lines that do not run through the customer-facing sales function, review and authorization by personnel with no stake in the outcome, and a mechanism through which staff can raise pressure without consequence. In an in-house laboratory, the question of what happens when a product fails a week before launch is not hypothetical, and the assessment will pursue it.

Weak Corrective Action

Corrective action requires the laboratory to react, evaluate the need to eliminate the cause, implement action, review effectiveness, and update risks and the management system as needed. Responses that correct the individual instance and stop, without asking why the system permitted it, produce repeat findings at the next assessment, and repeat findings are treated far more seriously than first occurrences. The most useful discipline is to ask, for every finding, whether the same cause could have produced errors elsewhere and whether any issued report is affected.

Buying and Using Accredited Testing

Before Booking

Verify the accreditation on the accreditation body's public directory rather than relying on a certificate supplied by the laboratory, since directories show current status and suspensions. Confirm that the scope lists the exact standard, edition, and clauses required, and that any parameter ranges cover the product. Confirm that the laboratory is accepted by the body that will consume the report. Agree the decision rule before work begins, and record the agreement. Establish what the laboratory needs from you: a bill of materials with safety-critical components identified, schematics, a construction data form, operating modes and support equipment for EMC work, and enough samples, since safety testing consumes them.

Reading the Report

An accredited report should be checkable at a glance. Confirm that the accreditation symbol or reference appears and that the report identifies the accreditation body and the accreditation number. Confirm that the methods cited match the accredited scope; where a report mixes accredited and non-accredited work, the non-accredited portions must be clearly identified, and a report that quietly includes out-of-scope testing under an accreditation symbol is a serious problem for both parties. Check that uncertainty is reported where it affects conformity, that any statement of conformity names the decision rule, that the sample identification matches the units actually shipped, and that the dates of receipt and testing are consistent with the configuration you intended to test. Check that any opinions and interpretations are marked as such, since the standard requires them to be clearly identified and to be issued only by authorized personnel.

Accreditation Symbols and Their Misuse

Accreditation bodies license the use of their symbols and the combined symbol referencing the international arrangement, and the rules are specific: symbols may appear only on reports covering accredited activities, may not appear on products, and may not be used in a way that implies the accreditation body endorses a product. Manufacturers occasionally reproduce a laboratory's accreditation symbol in their own marketing to suggest product approval. That misuse exposes the laboratory to enforcement by its accreditation body and the manufacturer to a claim of misrepresentation, and it is worth understanding before a marketing department discovers the symbol.

Key Points

Accreditation is third-party attestation of a conformity assessment body's competence, and it sits above certification rather than beside it. Laboratories are accredited; products are certified; the two words are not interchangeable.

ISO/IEC 17025:2017 governs testing and calibration laboratories, with requirements in clauses 4 through 8 covering impartiality and confidentiality, structure, resources, processes, and the management system, and with Option A or Option B available for the last of these. Its most consequential change from the 2005 edition was the requirement to document, agree, and report a decision rule whenever a statement of conformity is issued.

The accreditation chain runs from ISO/IEC 17011 for accreditation bodies down to ISO/IEC 17025 for laboratories, ISO/IEC 17020 for inspection bodies, ISO/IEC 17021-1 for management system certification bodies, and ISO/IEC 17065 for product certification bodies, supported by ISO/IEC 17043 for proficiency testing providers and ISO 17034 for reference material producers.

Accreditation is always scope-specific. The published scope document, with its standards, editions, ranges, and calibration and measurement capabilities, is the only reliable statement of what a laboratory may claim, and checking it before booking work prevents the most common and most expensive failure in compliance scheduling.

Metrological traceability and evaluated measurement uncertainty are the technical substance of an accredited result. Traceability requires an unbroken, documented chain of calibrations each contributing uncertainty; uncertainty must be evaluated and, where it affects conformity, reported.

International portability of accredited results rests on multilateral peer evaluation, now administered by Global Accreditation Cooperation Incorporated under the Multilateral Recognition Arrangement following the consolidation of ILAC and IAF on 1 January 2026. That portability is not the same as regulatory acceptance, which depends on separate government and scheme-level arrangements.

Related Topics