IPC/WHMA-A-620 Cable Assembly Classes 1 vs 2 vs 3: Which Class Do You Need?

How Do Classes 1, 2, and 3 Differ?

IPC/WHMA-A-620 Classes 1, 2, and 3 represent different expectations for product service and reliability. Class 1 emphasizes completed function, Class 2 addresses dedicated-service products that require continued performance and extended life, and Class 3 applies to high-performance or harsh-environment products where dependable operation on demand is critical. The Classes are application and acceptance categories, not simple labels for low, medium, and premium quality.

What Is Class 1?

Class 1 is intended for general electronic products where the main requirement is that the completed assembly performs its intended function. That definition is sometimes misunderstood as permission to accept careless workmanship, but Class 1 still has defined acceptance criteria. The difference is that the product is not being classified around the same service expectations or failure consequences associated with dedicated-service or high-reliability equipment.

A Class 1 assembly may be appropriate when a connection supports a low-consequence function, the equipment is easy to service, replacement is inexpensive, and a temporary interruption creates little operational impact. Examples can include certain general electronic devices, non-critical accessory wiring, development fixtures, or low-risk internal assemblies. The application still matters, because a low-cost cable can become commercially important when it is buried deep inside equipment that is expensive to disassemble or difficult to service in the field.

This is the first reason experienced engineering teams do not assign Class from appearance. Two harnesses can use the same terminal family, conductor size, connector housing, and jacket material while carrying completely different system responsibilities. One might supply an indicator light; the other might connect a control module whose failure shuts down equipment. If those two products are treated as equivalent simply because the harnesses look similar, the classification process is missing the actual risk.

Class 1 can also be a rational choice when the project does not benefit from higher-cost controls that do not materially reduce the expected failure consequence. The goal is not to push every assembly upward through the Classes; the goal is to select an acceptance level that matches how the product is used. Good engineering avoids both under-specification and unnecessary over-specification.

What Is Class 2?

Class 2 is intended for dedicated-service electronic products where continued performance and extended life are required. It is commonly relevant to professional electronics, industrial equipment, communications hardware, instrumentation, automation systems, and other products expected to operate reliably over a meaningful service period. The core expectation is stronger than “the cable worked once during final test”; the assembly should be produced consistently enough to support dependable service under its intended operating conditions.

This moves attention toward repeatable conductor preparation, crimping, connector assembly, strain support, routing, soldering, shielding, inspection, and controlled documentation. It also makes lot-to-lot consistency increasingly important. A customer purchasing the same assembly over several production cycles expects the later lots to match the approved design and workmanship baseline, which makes BOM control, drawing revision, tooling, inspection method, and test-program revision part of the product definition rather than administrative details.

Class 2 is often a practical middle ground when service interruption is undesirable, equipment is expected to operate for extended periods, repairs are possible if needed, and the normal environment is not inherently extreme. However, “industrial equals Class 2” is too simplistic. A single machine can contain twenty harnesses with very different functions, from indicator wiring to encoder connections, communication links, power distribution, and safety circuits. Those assemblies may not deserve the same risk classification.

The decision should therefore follow the function of the assembly rather than the industry label on the machine. That approach is especially useful for large OEM programs because it allows critical and non-critical interconnects to receive appropriate controls without forcing every cable into the highest-cost requirement set. It also makes later supplier audits easier because each part number has a defensible reason for its assigned Class.

What Is Class 3?

Class 3 is intended for high-performance or harsh-environment products where continued operation or performance on demand is critical. It is commonly associated with applications in which downtime cannot be tolerated, equipment may operate in demanding conditions, or failure could create significant operational consequences. Certain aerospace, defense, medical, transportation, safety-related, and high-reliability industrial systems may contain Class 3 cable assemblies, although the industry name alone should never be treated as automatic proof that Class 3 applies.

Class 3 usually places greater pressure on manufacturing discipline because workmanship conditions that might be acceptable at a lower Class can become defects at the higher acceptance level. That can increase the importance of operator and inspector competence, tooling control, documented setup, material identity, process evidence, traceability, approved deviations, and consistent interpretation of the controlled standard. The practical difference is not simply that someone “looks harder” at the finished cable; the manufacturing system should be capable of producing the tighter acceptance result repeatably.

Class 3 still cannot rescue an unsuitable design. A perfectly executed termination does not correct an undersized conductor, an overloaded contact system, inadequate shielding, poor strain relief, an unrealistic bend radius, a jacket that cannot withstand the chemical environment, or a connector that is not rated for the required temperature or sealing condition. Those issues must be solved through engineering before workmanship acceptance becomes meaningful.

That is the right way to view Class 3: it is one layer of a high-reliability product strategy. It strengthens how the assembly is manufactured and accepted, but it does not replace electrical design, mechanical design, environmental validation, interface requirements, or system-level reliability engineering. The Class works best when those engineering requirements are already specific enough to be measured and verified.

Is Class 3 Always Better?

Class 3 is not automatically better for every cable assembly; it is better only when the application needs the reliability and acceptance level associated with Class 3. Over-specification can increase manufacturing effort, inspection attention, documentation, training requirements, rejection exposure, and change-control burden without producing a meaningful improvement in the customer’s actual risk. Under-specification is more dangerous, but unnecessary Class 3 requirements can still create avoidable cost and complexity.

A useful decision begins with the consequence of failure. If a cable failure causes a harmless restart and the assembly can be replaced in a few minutes, the business case for the highest acceptance Class may be weak. If the same failure creates a safety issue, shuts down expensive equipment, causes loss of critical data, interrupts a mission, or requires hours of disassembly in a remote installation, the reliability requirement changes substantially even when the cable itself costs only a few dollars.

The mistake is to treat the Classes as a quality ladder in which Class 1 means cheap, Class 2 means normal, and Class 3 means premium. A better mental model is a risk ladder. The Class should reflect what the assembly must do, how long it must do it, the environment it faces, and what happens when the connection does not perform as intended. SINO-CONN follows the same principle in project review: Class 3 should not be used as a default claim for all assemblies, and the agreed revision and Class should be tied to the actual project requirement.

This risk-based approach also improves commercial discussions. Instead of arguing about whether Class 3 “costs too much,” the customer and manufacturer can identify which interconnects carry higher consequences and which can remain at a more appropriate Class. That creates a specification that is easier to manufacture, easier to audit, and easier to defend when quality and procurement teams review the project later.

Which IPC Class Does Your Cable Assembly Need?

The correct IPC Class should be selected by evaluating the assembly’s function, failure consequence, service expectations, operating environment, accessibility, and customer-specific requirements. Start with what the cable does and what happens when it fails, then document the agreed Class in controlled project records. A medical, automotive, industrial, or communications product can contain different Classes because risk is determined at the assembly level, not by industry name alone.

Which Applications Fit Each Class?

Application examples are useful for orientation, but they are not a substitute for engineering judgment. Class 1 can fit products where basic function is the main requirement and interruption has limited consequences. Class 2 often fits equipment that needs dependable operation and extended service life while remaining serviceable. Class 3 becomes relevant when the function must be available on demand, downtime is unacceptable, or the environment and consequences justify a high-reliability acceptance level.

Modern equipment makes this more complicated than a simple industry chart. A robot may contain power harnesses, encoder cables, camera cables, motor feedback lines, communication links, emergency-stop wiring, fan wiring, and maintenance interfaces. A medical system may contain display wiring, sensor interconnects, power distribution, user-interface cables, and connections supporting clinically important functions. Treating all of them as the same Class because they live inside the same machine can either under-control the critical assemblies or over-control the low-risk ones.

The better approach is to classify at the controlled assembly or product level. For a large equipment program, engineering can identify a small group of high-consequence interconnects that deserve more demanding requirements while allowing lower-risk harnesses to use a suitable Class 1 or Class 2 baseline. This is not cutting corners; it is applying quality resources where they create the most value.

The decision also needs to consider lifecycle support. A harness may be technically non-critical but extremely expensive to replace after installation. If replacement requires a service technician, machine shutdown, panel removal, recalibration, or shipment back to a repair center, the total cost of failure can justify stronger manufacturing controls even though the circuit itself is simple. Reliability classification makes more sense when it accounts for those real service conditions.

How Should Failure Risk Be Judged?

Risk assessment should consider more than the probability that a termination will fail. The consequence of failure, ability to detect degradation, availability of redundancy, service access, operating demand, and expected product life all influence how much workmanship control is justified. This is similar to the thinking used in broader reliability and quality planning: two defects with the same probability can have very different priorities when one is easy to recover from and the other creates a major operational loss.

The purpose of this matrix is not to generate a numerical Class score. A cable exposed to vibration does not automatically become Class 3, and a cable in a clean indoor cabinet is not automatically Class 1. The matrix is useful because it forces the project team to discuss the conditions that actually determine reliability rather than relying on an industry label or a habit carried over from a previous design.

Serviceability can be particularly important. An inexpensive harness installed behind an easily removable cover may create little downtime. The same harness embedded inside a sealed module or routed through a large machine can create hours of labor and lost production when replaced. If field service cost is several hundred or several thousand dollars, improving manufacturing control may be economically justified even when the assembly itself has a modest unit price.

The team should also consider whether the system can fail safely. A connection that causes equipment to enter a predictable safe state presents a different risk from an intermittent signal that creates uncontrolled behavior or removes monitoring. That distinction often matters more than the nominal voltage, wire gauge, or physical size of the cable. It should be discussed explicitly rather than inferred from a general product category.

How Does Environment Change the Decision?

Environmental stress can increase the need for reliability, but the IPC Class does not define the actual environmental design. A dynamic robotic cable still needs a specified bend radius, motion type, stroke, torsion, cycle target, routing, and fixation method. A waterproof assembly still needs a defined IP target, mating condition, sealing interface, test method, and post-test electrical criteria. A high-speed interconnect still needs protocol, impedance, loss, skew, shielding, and test configuration requirements.

This separation is critical because a project can be over-specified in workmanship while remaining under-specified in performance. Calling a cable “Class 3” does not tell the manufacturer whether the jacket must survive coolant exposure, whether the harness will flex ten thousand times or ten million times, whether the connector will be mated outdoors, or whether the differential pair must maintain a particular impedance through a tight routing path. Those parameters have to be stated independently.

A useful engineering sequence is to define the application first, then define electrical and mechanical requirements, then define environmental conditions, then select the appropriate workmanship Class, and finally establish the verification plan. When those steps are reversed, teams sometimes use Class 3 as a substitute for missing engineering information. The result can be an assembly with demanding visual acceptance but unclear flex-life, sealing, current, thermal, or signal-performance requirements.

The same logic applies to harsh environments. Temperature, vibration, moisture, chemicals, UV exposure, abrasion, and electromagnetic interference each create different failure mechanisms. The correct response may involve material changes, shielding changes, strain-relief design, connector sealing, routing, or additional testing. Raising the IPC Class can strengthen workmanship expectations, but the design still has to address the specific stress mechanism. A harsh-environment label by itself is not a complete engineering requirement.

Who Should Make the Final Decision?

The customer or designated User should ultimately establish the applicable Class because that organization understands the equipment function, regulatory obligations, acceptable downtime, and consequences of failure. The cable manufacturer still has a meaningful responsibility to identify contradictions, missing data, or requirements that appear inconsistent with the application. A supplier that blindly accepts every Class callout without reviewing the drawing may be easier to quote with, but it is not necessarily easier to build a reliable product with.

For new product development, the best decision often involves product or system engineering, electrical engineering, quality, manufacturing engineering, procurement, and the cable assembly manufacturer. Engineering understands system function; quality understands acceptance, traceability, and failure evidence; manufacturing understands process capability; procurement understands cost and supply continuity; and the cable supplier understands how the requirement translates into terminals, tooling, assembly sequence, inspection, and test.

This cross-functional approach is especially useful when the product contains several harnesses with different risk levels. Instead of choosing one Class for the whole machine, the team can create a controlled matrix by assembly part number or functional group. Critical assemblies receive the appropriate acceptance level, while non-critical harnesses avoid unnecessary requirements. That makes production planning and supplier audits much more precise.

The decision should then be frozen in controlled documentation. If the Class changes later, the change should be handled like any other engineering revision, with review of affected workmanship criteria, inspection methods, training, approved samples, and possibly validation. A verbal statement in an email or meeting should not become the permanent production baseline for a long-term program. The released record should be easy for production and quality teams to retrieve during repeat orders.

How Do IPC Classes Change Workmanship Criteria?

IPC Classes change how specific workmanship conditions are accepted or rejected. The difference is criterion-specific rather than a universal percentage, so there is no responsible rule such as “Class 3 is 20% tighter than Class 2.” Crimps, soldered terminations, splices, shielding, connectorization, routing, and other features must be evaluated against the applicable clause, product Class, customer drawing, and component requirements used by the project.

How Do Crimp Criteria Change?

Crimping deserves special attention because a terminal crimp is simultaneously an electrical connection and a mechanical retention system. A good result depends on a chain of variables: correct terminal identity, compatible wire construction, strip length, conductor position, insulation position, applicator selection, tooling condition, crimp setup, inspection method, and verification. A small error at the beginning of that chain can create a termination that looks acceptable from one angle but performs poorly under pull, vibration, thermal cycling, or repeated service.

The Class affects how workmanship conditions are judged, but the Class does not create one universal crimp-height or pull-force value. Those measurable limits usually come from the terminal manufacturer’s application specification, customer drawing, validated tooling setup, or project-specific requirement. For example, two nominally 22 AWG wires may use different conductor strand constructions, insulation diameters, or materials, so the correct terminal and crimp settings must be based on the actual wire and contact system rather than the AWG number alone.

A mature crimp-control system commonly links the wire part number, terminal part number, applicator or tooling identity, setup condition, first-piece verification, in-process measurement, and applicable inspection criteria. Crimp height is commonly recorded in millimeters, mechanical retention in newtons, and low-level electrical resistance in milliohms when the project requires those measurements. The important point is not the unit; it is that every measured value has a controlled acceptance limit tied to the actual terminal and wire system.

Visual workmanship remains important, but it should not be the only evidence of a stable crimp process. Strip damage, conductor brush position, bellmouth condition, insulation support, terminal deformation, contact damage, and insertion into the connector housing all contribute to the final result. The strongest manufacturing approach controls the process that creates the crimp and then uses inspection and testing to confirm that the process stayed within its approved window.

How Do Solder and Splice Criteria Change?

Soldered terminations can pass a basic continuity test while containing weaknesses that only appear later. Excessive heat can damage insulation, solder can wick too far into a stranded conductor and create a rigid transition, insufficient support can place mechanical movement directly at the soldered region, and contamination or poor wetting can create unstable electrical performance. Higher reliability expectations therefore increase the importance of a controlled soldering process rather than relying on a final visual check alone.

The production system should define conductor preparation, terminal condition, solder material, soldering method, temperature control where applicable, acceptable wetting, insulation condition, cleanliness, mechanical support, and inspection timing. Personnel qualification also matters because hand soldering is a skill-sensitive process. Even when two operators use the same equipment and solder alloy, differences in dwell time, heat transfer, wire preparation, and handling can produce different results if the method is not controlled.

Splices require similar discipline because they are often hidden after heat shrink, sleeving, wrapping, or harness protection is installed. If the critical workmanship is no longer visible at final inspection, the process needs an inspection or verification gate before the covering operation. This is one reason good harness manufacturing uses in-process quality checkpoints rather than expecting the final inspector to discover every problem after the assembly has already been completed.

The applicable IPC Class helps determine whether the workmanship condition is acceptable, but customer specifications can add further requirements. A regulated or high-reliability program may also require specific materials, operator qualifications, process records, or validation beyond the normal A-620 acceptance criteria. Those additional controls should be stated in the project documentation instead of being assumed from the word “Class 3.”

How Are Shielding and Connectorization Judged?

Shielding is a good example of the boundary between engineering design and workmanship acceptance. Engineering first decides what the shield must do and how it should be terminated. The construction may use foil, braid, drain wire, a conductive backshell, a pigtail, a 360-degree termination, or another method. Manufacturing then has to reproduce that released design without introducing damage, excessive unshielded length, incorrect grounding, loose hardware, or other variation that changes the intended result.

Important production characteristics can include shield preparation length, braid damage, drain-wire routing, shield-to-shell bonding, insulation between shield and signal conductors, backshell assembly, strain-relief position, connector orientation, contact seating, locking, and marking. The workmanship Class governs acceptance of applicable conditions, while the drawing and electrical specification define the actual shield architecture. A harness can therefore be perfectly acceptable visually and still have poor EMC performance if the released design itself is wrong for the system.

For RF and high-speed differential assemblies, this distinction becomes even more important. Continuity testing confirms that the expected conductors are connected, but it does not prove that the cable performs properly at operating frequency. Depending on the product, additional verification may include TDR measurements, insertion loss, return loss, network certification, VNA testing, or protocol-level functional testing. These methods measure performance that a workmanship standard alone cannot guarantee.

Connectorization follows the same logic. Correct contacts must be fully seated in the intended cavities, secondary locks must be engaged where applicable, keying and orientation must be correct, seals must be positioned as designed, and strain relief must not place unintended load on the termination. The Class strengthens how these conditions are accepted, but component manufacturer’s instructions and customer drawings remain essential because connector systems have their own limits and assembly methods.

Do Customer Drawings Override IPC Criteria?

The released customer drawing, contract, purchase order, and approved specifications form part of the controlling requirement set and can supplement or define requirements beyond the general workmanship standard. The important principle is not that one document automatically “wins” in every situation; it is that the hierarchy must be clear and conflicts must be resolved before manufacturing. Production should never choose the easier interpretation simply because two controlled documents disagree.

A drawing may define the exact connector and terminal part numbers, wire specification, pinout, length tolerance, strip dimensions, shield termination, labels, torque, test voltage, electrical limits, IPC revision, Class, or special workmanship conditions. If the purchase order calls for a different standard revision or Class, the discrepancy should trigger engineering clarification. Once resolved, the approved requirement should be captured in a controlled revision, deviation, clarification record, or other project document that remains accessible for repeat orders.

This matters because long-term cable programs often outlive the people who handled the first sample. A verbal exception accepted by one engineer during prototype development can become a serious problem two years later when a different quality team audits the same part number. If the exception was never documented, the supplier may have no defensible explanation for why production differs from the drawing or standard.

SINO-CONN‘s project approach is to convert customer drawings, BOMs, pinouts, connector part numbers, samples, and technical requirements into a controlled manufacturing definition before scale-up. That same discipline should apply to IPC requirements. The standard, revision, Class, customer exceptions, and approved deviations need to remain connected to the released product baseline rather than being treated as separate quality paperwork. This keeps the manufacturing and inspection teams working from the same definition throughout the product lifecycle.