How to Tell How Many Layers a PCB Has
Knowing how many layers a board contains matters more often than it sounds. It determines whether a design can be copied, whether a repair is feasible, and whether a supplier’s quotation is credible. Fortunately the layer count can usually be established from the outside, using the vias, the visible edges and a light source.
What the Layers Do
A four-layer board routes signals on the first and fourth layers and uses the two inner layers for ground and power. Six-layer and eight-layer boards add auxiliary signal layers between the planes. Each layer has a defined role, and the arrangement of those roles is what makes a stack-up work: a signal layer needs a reference plane next to it, and the planes need to be adjacent to each other to form the distributed capacitance that makes the supply behave.
This is why the layer count is normally an even number. A stack-up is built symmetrically around a core so that the board does not warp during lamination, and symmetry is much easier to achieve with an even number of layers.
The Via Test
The quickest external method uses the vias. On a four-layer board the routing is on the outer layers, so the vias pass through the whole board and appear on both faces. On a six or eight-layer board, blind vias are used to reach the inner layers, and a via that appears on the top face may have no counterpart on the bottom.
The procedure is to find a via on one side and look for it on the opposite side at the same position. If every via has a matching hole on the other face, the board is four layers. If some vias appear on one side only, the board has blind vias and therefore contains at least six layers.
The Light Test
A related check uses a light source. Hold the board up to a bright light and look at the via holes from one side. If light passes through the holes, the holes are clear through the board, which again points to a four-layer construction. If the holes do not transmit light, the vias being examined are blind and the board has more layers.
Neither test is completely reliable. A four-layer board with plugged or filled vias will not transmit light, and a board with a mixture of through and blind vias can mislead an examiner who samples only one. Both tests work best as a first indication, confirmed by the edge view or by the documentation.
Reading the Edge
The edge of a board shows the laminate and the copper layers, and a cleanly cut edge with magnification reveals the individual layers. A glass-fibre laminate with a smooth, even edge is a healthy sample; a rough, frayed edge indicates a material or cutting problem and also makes the layers harder to count. Polishing or lightly abrading the edge improves the visibility of the copper, although this destroys the sample for further use.
A cross-section prepared by a laboratory is the authoritative method. The board is encapsulated, ground back and polished so that the plated through holes and the copper layers are visible in a microscope image. The same cross-section also reveals barrel copper thickness, via fill and plating quality, which is why it is the standard method for process verification rather than only for counting layers.
Why a Count Can Be Misleading
Counting copper layers is not the same as counting routing capacity. Two adjacent signal layers without a plane between them couple to each other and cannot both be used freely, so a six-layer stack-up that wastes a layer pair on two adjacent routing layers may perform no better than a well-planned four-layer board. The number is a starting point for asking what each layer is doing.
Blind and buried vias also change what the same count delivers. A stack-up that uses blind vias can escape from a fine-pitch device in less area than one restricted to through holes, which is one reason two boards with the same layer count can have very different routing densities. Our layer count selection notes explain the trade-offs, and the layer assignment guidance explains how the layers should be arranged.
What to Do With the Answer
If the purpose is a repair, the count tells you whether a blind via is likely to be involved and therefore whether the fault can be reached from the surface. If the purpose is a quotation, the count tells you what the fabrication process will involve, which is why a supplier who quotes without asking is quoting a guess. If the purpose is to reproduce a board, the count is the first of several parameters, and the stack-up, the copper weights and the via types have to be established as well.
Confirming the count from two independent methods, and recording which methods were used, prevents the common situation in which a board is documented as four layers on the basis of a single observation and turns out to be six when the drawings are compiled. The design release checklist covers the documentation that should accompany the board, and the reverse engineering notes describe the wider process of recovering a design from a physical sample.
Layer Count and the Board You Are Looking At
Consumer electronics have moved in both directions over the same period. Integration removed the need for many of the packages that once surrounded a processor, so some designs that would previously have needed more layers now fit on fewer. At the same time, ball-grid-array packages and fine-pitch components pushed escape routing demand up, so others have grown. The count on a given board is therefore a statement about the routing task, not a statement about how modern or how capable the product is.
That context is useful when the count is being used as evidence. A board with more layers than expected usually indicates a routing constraint that could not be solved another way, such as a fine-pitch device without enough room to fan out on two layers. A board with fewer layers than expected usually indicates a well-planned stack-up and a component set with modest pin counts. Neither observation says anything about quality on its own.
FAQ
Can a multilayer board be identified without damaging it? Usually yes, by combining the via test with the light test and by examining the board edge under magnification. X-ray inspection gives a clearer view of the internal structure and is non-destructive, which makes it the practical choice where the board is valuable. The only method that shows both the layer count and the plating quality in full detail is a prepared cross-section, which is destructive.
Why do some four-layer boards not let light through the vias? Because the vias have been plugged, tented or filled as part of the fabrication process. Via filling is increasingly common on boards with fine-pitch components, where a filled and plated via allows a pad to be placed directly over it. A plugged via blocks the light path without changing the layer count, which is why the light test should be treated as an indication rather than a verdict.
What does gopcb need if a stack-up is being reproduced? The layer count, the copper weights of each layer, the dielectric thicknesses and the via types. Layer count alone does not determine an impedance, because the impedance depends on the trace geometry and the dielectric separation. Providing the full stack-up table from the original drawing, or asking us to derive one from a cross-section, gives a reproduction that behaves as the original did.
