IPC Class 1 vs Class 2 vs Class 3: Which PCB and Assembly Acceptance Class Do You Need?

Detailed Explanation

"IPC Class 1/2/3" shows up constantly in fabrication drawings, assembly quality agreements, and contract manufacturer (CM) quotes — but it actually refers to two separate standards that happen to share identical class names and a similar three-tier philosophy:

• IPC-6012 — Qualification and Performance Specification for Rigid Printed Boards — defines acceptance classes for the bare, unpopulated printed board: plating thickness, annular ring size, hole-wall quality, and other fabrication-level construction tolerances.
• IPC-A-610 — Acceptability of Electronic Assemblies — defines acceptance classes for the populated, soldered assembly: solder joint fillet quality, component placement accuracy, and defect acceptability after SMT and through-hole assembly.

Both standards use the same three class names and the same underlying logic — the class reflects how severe the consequence of a defect is, not simply "how good" the board looks — but they govern different manufacturing stages and are specified separately. Confusing the two, or assuming a single "Class 2" callout covers both fabrication and assembly, is a common source of mismatched expectations between a design team and a CM.

The Three Classes

Both IPC-6012 and IPC-A-610 use the same three-tier structure, though the specific numeric tolerances and defect criteria within each class differ between the two standards:

• Class 1 — General Electronic Products. Applies where the primary requirement is that the product functions, and cosmetic imperfections or minor defects that don't affect function are acceptable. Typical for very cost-sensitive, short-lifecycle consumer products where field failure has low consequence and the product may be replaced rather than repaired.
• Class 2 — Dedicated Service Electronic Products. Applies where extended service life and reliable performance are required, though uninterrupted service isn't strictly critical. This is the default class for most commercial, industrial, and general-purpose products — the large majority of designs handled by a typical CM fall here.
• Class 3 — High Performance/High Reliability Electronic Products. Applies where continuous performance or performance-on-demand is critical, equipment downtime cannot be tolerated, and the end-use environment may be harsh or safety-critical. Typical for medical implants and life-support equipment, aerospace and defence systems, and other applications where a field failure has severe consequences.

The exact numeric tolerances each class allows — annular ring minimums and plating thickness for IPC-6012, solder fillet geometry and acceptable void percentage for IPC-A-610 — are defined in the current edition of each standard and are revised periodically; consult the specific standard edition your fabricator or CM is qualified against rather than assuming a figure from a previous revision still applies.

How the Two Standards Interact

A single product typically needs both a fabrication class (IPC-6012) and an assembly class (IPC-A-610) specified, because they cover different physical risks:

• IPC-6012's class governs whether the bare board itself — the copper, plating, and dielectric structure — is built to a tolerance appropriate for the product's expected service life and environment, before any component is ever placed on it.
• IPC-A-610's class governs whether the soldering and placement process that turns that bare board into a working assembly meets an acceptability standard appropriate for the same service requirement.

These don't have to match numerically. A board using conventional, well-proven fabrication (standard layer count and via structures, no exotic materials, generous annular rings) might reasonably specify IPC-6012 Class 2, even if the product's overall reliability requirement — driven by, say, a safety-critical sensor interface or a harsh vibration environment — justifies holding the assembly workmanship to IPC-A-610 Class 3. The right approach is to evaluate the actual risk each standard governs independently, not to pick one number and apply it uniformly out of habit.

Choosing a Class for Your Product

• Consumer electronics with a short expected lifecycle and low failure consequence (toys, low-cost accessories, promotional devices) — Class 1 is often sufficient, and can meaningfully reduce cost and lead time versus over-specifying a higher class.
• Most commercial and industrial products (general-purpose IoT devices, instrumentation, consumer products expected to last years, most B2B hardware) — Class 2 is the appropriate default, and is what most CMs assume if no class is stated.
• Medical, aerospace, defence, and other mission-critical or safety-critical products — Class 3 is typically required, either by the standard's own guidance or by a downstream regulatory or contractual requirement (many medical device quality systems and aerospace contracts mandate Class 3 explicitly).

State the chosen class explicitly on the fabrication drawing (for IPC-6012) and in the assembly drawing or quality agreement (for IPC-A-610) — see PCB fabrication output files for where fabrication-level specifications belong in the output package. Don't rely on a CM's default assumption, particularly if the product's actual requirement is Class 1 (where over-specifying Class 2 adds unnecessary cost) or Class 3 (where under-specifying Class 2 risks accepting workmanship the product's actual reliability requirement doesn't tolerate).

Design Considerations

• Specify class explicitly — don't assume a CM's default matches your product's actual requirement. Most CMs default to Class 2 in the absence of an explicit callout, which is the right choice for most designs but not a substitute for a deliberate decision on a safety- or reliability-critical product.
• Class is a specification input, not a substitute for good design-for-manufacturability practice. A well-laid-out Class 2 board with sound DFM practices can outperform a poorly designed Class 3 board in practice — class defines the acceptability bar for workmanship and construction, not the underlying design quality.
• Higher class generally means higher cost and longer lead time, both from tighter fabrication tolerances and more rigorous inspection and test requirements (more thorough AOI review, potentially 100% X-ray rather than sampling, more detailed workmanship documentation). Confirm the actual cost delta with your specific fab house and CM rather than assuming a fixed premium.
• A defect acceptable under Class 2 may require rework or rejection under Class 3 — the same physical solder joint can pass one class's inspection and fail another's, which is why the class must be agreed with the CM before production, not discovered during an inspection dispute.
• Choosing the right acceptance class for your product's actual reliability requirement — neither over-specifying (unnecessary cost) nor under-specifying (unacceptable field-failure risk) — is a decision Zeus Design's rapid prototyping and production support works through with clients before committing to a fabrication and assembly order.

Common Mistakes

• Assuming "IPC Class 2" is a single, unified specification that covers both fabrication and assembly, when in practice IPC-6012 (fabrication) and IPC-A-610 (assembly) are separate standards that need to be specified independently.
• Leaving the class unstated and assuming the CM will apply the "right" one — most CMs default to Class 2, which may not match a product's actual Class 1 cost target or Class 3 reliability requirement.
• Specifying Class 3 by default "to be safe" for a cost-sensitive product with a low failure consequence, adding cost and lead time without a corresponding reliability benefit the product needs.
• Treating class as a proxy for overall design quality rather than what it actually is — an acceptability tier for construction tolerances and workmanship defects, independent of whether the underlying schematic and layout design is any good.
• Discovering a class mismatch only during an inspection dispute — for example, a CM rejecting a joint that would have passed Class 2 inspection but was actually ordered under an unstated Class 3 assumption — rather than agreeing the class explicitly with the CM before production starts.