Blind Slot and Open Slot Routing in PCBs: Depth, Tolerance and Sequence

A slot in a printed circuit board is usually there for a mechanical reason: to seat a connector, to separate two circuits, to let air move or to give a daughter card somewhere to sit. Whether that slot is cut all the way through the panel or only part way into it decides how the board is made, what it costs and how much tolerance the designer can expect. The two cases are a blind slot and an open slot, and they are not interchangeable.

What a Blind Slot Is

A blind slot removes material to a controlled depth without breaking through the far side of the board. Material, and usually copper, remains underneath. That remaining web is what the feature is for: it can provide a shielded cavity for a radio front end, a recess that lets a component sit flush, or a locating feature that holds a connector body without weakening the panel.

Producing it needs depth control that is measured rather than assumed. The routing depth is set from the surface, the slot tolerance is one-sided, and the depth is confirmed on a first article whose remaining web thickness is measured, and the tolerance has to allow for laminate thickness variation across the panel as well as for machine accuracy. A blind slot is therefore a controlled depth milling operation, and the depth is a specification in its own right.

What an Open Slot Is

An open slot passes through the full board thickness. It is produced by routing, punching or, in flexible material, by die cutting, and it needs no depth control because the tool simply exits the far side. Edge finish and positional tolerance are the only quality attributes that matter, and both are within the range of a normal routing process.

Because the process is simple, the cost is low and the yield is high. Card edge fingers, connector keying features, ventilation openings and cable pass-throughs are all open slots, and where the application allows one, there is rarely a good argument for making the operation more difficult than it needs to be.

Where the Two Diverge

Depth control is the first difference and it drives everything else. A blind slot has to be milled with a depth-controlled cutter or, at fine scale, ablated with a laser whose energy is calibrated to stop within the intended layer. Either way the tool approaches a target depth from one side only, so the tolerance is one-sided and the process needs verification the open slot does not.

The second difference is registration to the inner layers. On a multilayer board the depth of a blind slot is often defined by an inner copper layer that is intended to act as an etch stop or as the floor of the cavity. That makes the slot position a layer-to-layer alignment problem as well as a depth problem, and both errors add. The general rules for routing features in the board edge are covered in slot and edge routing rules.

Cost and Tolerance Compared

An open slot priced as a prototype will typically cost a fraction of a blind slot, because it needs one pass of a standard cutter and no depth verification. A blind slot adds setup, first article measurement and a higher scrap risk, and the penalty grows with the number of slots on the panel and with the tightness of the depth tolerance.

Tolerance follows the same pattern. An open slot holds a positional tolerance in the region of a tenth of a millimetre, which is adequate for a connector cut-out or a keying feature. A blind slot on a controlled depth milling process holds roughly half that, and a laser ablated slot in a thin HDI stack can hold tighter still, at proportionally higher cost.

Applications That Favour Each

Blind slots appear where the cavity is functional rather than mechanical. Millimeter wave modules use a recessed cavity to keep a radiating structure at a defined height above a ground plane. Embedded component designs need a recess deep enough to bury a die or a passive without adding height to the finished assembly. Rigid flex constructions use partial depth cuts to control where the flexible window begins.

Open slots appear wherever the board only has to fit something else. Card edge connectors, standoff clearances, cable exits, airflow openings and isolation barriers are all through features. So is the boundary of a flexible window in a rigid flex board, where the material is removed completely rather than thinned.

Design Rules Worth Following

Keep the remaining web under a blind slot thick enough to survive handling and assembly, and remember that the web is a structural element subject to the same bending behaviour as the rest of the board. Where the slot sits under a heavy component, the web thickness and the copper beneath it are doing real work, and the local stiffness matters as much as the electrical function. General guidance on edge features appears in board outline and mounting design.

Smooth internal corners on both slot types, keep copper a defined distance from the cut, and remember that a plated slot, including one that needs edge plating along its wall, requires a plating process that can reach into it. Where the slot connects to a via structure in a multilayer board, the arrangement has to be planned with the stackup, as described in via and stack selection.

Manufacturing Sequence

Open slots are normally cut at the end of the process, after plating and surface finish, so the coating is not damaged by the cut. Blind slots are harder to place in the sequence: cutting them late risks damaging a controlled depth surface, while cutting them early exposes the cavity to every subsequent chemical step. Most fabricators mill them after lamination and before final finish, then protect the floor of the cavity during plating.

Where a blind slot has to be plated, the sequence becomes substantially longer, because the plating has to reach a surface that is not on the outside of the panel. Sequential lamination is often the answer: the cavity is formed in a sub-assembly, plated, and then bonded into the final stack. That is a fabrication plan rather than a routing operation, and it needs to be agreed before the design is released.

FAQ

Why is a blind slot more expensive than an open slot? Because the depth has to be controlled and verified rather than simply cut through. That adds setup, inspection and scrap risk, and the penalty rises with the number of slots and the tightness of the depth tolerance.

Can a blind slot be plated? Yes, but usually only through sequential lamination, where the cavity is formed and plated in a sub-assembly before final bonding. Plating a cavity after final lamination is difficult and rarely offered.

What is the minimum practical depth for a blind slot? Roughly 0.1 mm in a laser processed thin stack and a few tenths of a millimetre with mechanical milling. The limit is set by the remaining web thickness and by the ability to hold depth across the panel.

Should the slot be cut before or after surface finish? Open slots are cut after finish so the coating is not disturbed. Blind slots are normally milled after lamination and before finish, with the cavity floor protected through the remaining steps.