Clearance Rules: Electrical and Non-Electrical Spacing

Every pair of copper features on a board has a spacing between them, and that spacing is either dictated by electricity or by manufacturing. Designers who treat all clearances as one category tend to over-constrain the layout in some places and under-constrain it in others, and the errors appear as either manufacturing complaints or as reliability failures. Separating the two categories makes the rules much easier to apply.

Why Spacing Is Not One Number

An electrical clearance exists because a voltage difference between two conductors can break down the insulation between them. The relevant quantity is the voltage, the pollution degree of the environment, the altitude and the coating, if any. A non-electrical clearance exists because a machine or a person has to work on the board, because a component needs room, or because the laminate needs material around a hole to survive mechanical load.

The two requirements often differ by much more than a designer expects. A high-voltage isolation gap may be several millimetres, while the clearance around a mounting hole is a fraction of that. Treating them as a single global rule produces a board that is both larger than necessary and still incorrect in the places that matter.

Electrical Clearance

The general form of the rule is that the safety spacing must increase with the voltage. Standards express this as a pair of values for each voltage band: one for the distance through air, which governs the physical gap between conductors, and one for the distance over the surface, which is the creepage. The creepage requirement is usually the larger of the two because contamination on the surface forms a conductive path that a clean air gap does not.

Conformal coating changes the calculation, because a coating that adheres properly interrupts the surface path. That is why coated assemblies are permitted smaller creepage distances than uncoated ones at the same voltage. The benefit is only available where the coating covers the feature completely, which means the reliability of the coating process becomes part of the electrical safety argument rather than a cosmetic detail; the mechanisms by which coatings fail are described in this article on conformal coating and board protection.

Inside a low-voltage digital design the electrical clearance requirement is usually modest, and the limiting factor becomes something else: the spacing needed to control crosstalk, or the spacing required by the fabrication process. Both of those are covered below and in the discussion of the 3W rule.

Non-Electrical Clearance

The non-electrical clearances are the ones that decide whether the board can be built and assembled. Around a mounting hole, copper and components must be kept back by enough distance that a screw head or a standoff does not short a net or crush a part. Around a tooling hole, the clearance is set by the locating pin and the need to engage and release it without binding.

Component clearance follows from the assembly process. A placement nozzle needs space to approach a component, a rework tool needs access to its joints, and a connector needs room for its mating action and for any latch or screw. Where a component sits inside a keepout intended for a connector body, the board may be electrically perfect and mechanically impossible.

Board edge clearance belongs to this category as well. Copper must be kept back from the profile so that the routing operation does not expose it, and components must be kept further back so that handling and depanelling do not strike them. The requirements differ from fabricator to fabricator and are set out alongside the outline rules in this discussion of board outline and mounting design.

Via and Pad Clearance

The spacing between a via and a neighbouring pad sits in a middle category. Electrically, the gap only has to resist the voltage between the two nets and avoid excessive capacitance, both of which are usually easy at low voltage. Geometrically, it has to survive the tolerance stack of drilling, plating and layer registration, because the pad may be trimmed and the hole may not land exactly where it was drawn.

That is why the practical rule for via-to-pad clearance is larger than the electrical requirement, and why it depends on the layer count and the construction. The arithmetic behind it, and the way the movement of the panel during lamination enters the calculation, is described in this article on via to trace clearance.

Building a Clearance Scheme

Once the two categories are separated, a clearance scheme can be assembled from them. Electrical clearances become a set of net-class rules driven by voltage. Non-electrical clearances become a set of regions and edge rules driven by the mechanical drawing and the assembly process. Fabrication clearances become a global minimum derived from the fabricator capability.

Each of those is enforceable in a layout tool, and each should be defined before routing begins. Defining them afterwards produces a design that satisfies the tool but not the intent, because a rule added late will be satisfied by whatever geometry already exists rather than by the geometry the design required.

Review

Before release, confirm that every high-voltage net pair has the spacing the applicable standard requires for the environment and the coating, that every mounting and tooling feature has its keepout honoured on all layers, that the edge clearance matches the fabricator requirement, and that the global minimum clearance reflects the chosen process rather than a default value. Those four checks cover the categories above and are the ones most often found to be inconsistent on a finished design.

Spacing Inside a Low-Voltage Design

Most digital boards never approach the voltages at which the safety spacing tables become the dominant constraint, and the spacing in those designs is set by something else. The three factors that matter are the process minimum, the crosstalk requirement and the assembly requirement. The process minimum is a hard floor that applies everywhere. The crosstalk requirement applies to the nets that are fast or sensitive, and it is normally larger than the process minimum. The assembly requirement applies wherever a machine or a tool must reach the board.

Where those three conflict, the resolution is normally to move a component rather than to violate a rule. A part that cannot be reached by the placement nozzle, or whose joints cannot be reworked, will generate more cost in production than the space it occupies is worth, and the same is true of a trace that has been squeezed below the process minimum to make a route fit.

FAQ

What is the difference between clearance and creepage? Clearance is the shortest distance through air between two conductors, while creepage is the shortest distance along the surface of the insulating material. Creepage requirements are usually larger because contamination on the surface can form a conductive path that air does not provide.

How much clearance does a mounting hole need? Enough that the screw head or standoff does not contact copper or components, and enough that the laminate around the hole can carry the mechanical load. The figure depends on the fastener and the number of layers, and it is a mechanical requirement rather than an electrical one.

Can conformal coating reduce the required spacing? Yes, provided the coating adheres completely over the feature and is applied by a controlled process. A coating that is incomplete or that lifts at the edges does not interrupt the surface path, so the reduction depends on process reliability rather than on the intention.