PCB Thickness Chart: Standard Sizes and How to Choose
What Determines the Finished Thickness
The thickness quoted on a drawing is the finished board, measured from the top of one solder mask surface through the stack to the bottom, and it is the sum of several separate contributions. The dielectric cores and the prepreg that bonds them set most of it, the copper on each layer adds a fraction, the solder mask adds a few tens of micrometres on each face, and the surface finish adds a little more. Because the press cycle determines how much resin flows out of the prepreg, the finished thickness is a process result rather than a simple arithmetic sum, which is why fabricators specify a tolerance rather than an exact number. The working units vary by audience: millimetres in Europe and Asia, inches in older American drawings, and thousandths of an inch, or mils, in most impedance and stackup calculations. One mil is 0.001 inch, so 1.6 mm is 63 mil, and a 0.063 inch callout and a 1.6 mm callout mean the same thing.
The Standard Thickness Chart
The values below are the sizes that laminates are stocked in, and a stackup that lands on one of them is faster and cheaper to build than one that does not.
0.2 mm (0.008 in) for ultra thin wearables and sensor inlays. 0.4 mm (0.016 in) for smart cards and modules. 0.6 mm (0.024 in) for mobile phones and camera modules. 0.8 mm (0.031 in) for portable electronics and HDI boards. 1.0 mm (0.039 in) for general consumer devices. 1.2 mm (0.047 in) for automotive electronics. 1.6 mm (0.063 in) for standard industrial and consumer boards, by far the most common. 2.0 mm (0.079 in) for power equipment. 2.4 mm (0.094 in) for communications equipment. 3.2 mm (0.126 in) for heavy industrial and backplane applications, with thicker builds available on request.
The Standards Behind the Numbers
The rigid board qualification and performance specification sets the requirements for the finished board and leaves the exact thickness to the drawing, while the laminate standards define the nominal core and prepreg thicknesses that are available. In practice the chart above has become a de facto international standard because it reflects what laminates exist. A drawing that asks for 1.55 mm will be built from a 1.6 mm stack and delivered inside the tolerance band, so it is simpler to start from the standard and adjust the stackup for impedance and mechanical requirements than to specify an arbitrary finished thickness.
Layer Count and Typical Thickness
Layer count does not fix the thickness; the two are chosen together. A four layer board is commonly built at 1.6 mm with two cores and two copper layers, a six layer board at 1.6 mm with thinner dielectric to keep the same overall height, and an eight layer board is often specified at 2.0 mm to keep each dielectric layer practical. HDI boards go the other way, since a 0.8 mm or even 0.6 mm thickness is normal when microvias and thin dielectrics are in use. A dense multilayer board that has to fit a 1.6 mm envelope needs thin dielectrics, which in turn changes the trace widths required for a controlled impedance, so the stackup and the layer count should be planned in one pass rather than sequentially.
How Thickness Affects Performance
Impedance. The dielectric thickness between a signal layer and its reference is one of the four variables behind characteristic impedance, so choosing the overall thickness also fixes the range of trace widths that can hit 50 or 100 ohms. A thin board forces narrow traces, and a thick board forces wide ones.
Mechanical stiffness. Stiffness rises steeply with thickness, so a 2.4 mm board resists bending during connector insertion far better than a 1.2 mm board of the same area, and a thin board needs support from the enclosure or from a stiffener.
Connector and press fit compatibility. Press fit pins, card edge fingers, backplane connectors and board to board mezzanines are all designed for a thickness range. Specifying outside it means either a connector that does not retain properly or an assembly that cannot be inserted.
Thermal and mechanical cycling. A thicker board has a larger thermal mass and a higher Z axis expansion, which changes how through hole joints and large BGAs are stressed during reflow and in service.
Weight and space. In handheld and wearable products the board is often the largest single component, and 0.6 mm instead of 1.6 mm saves both volume and mass across a product line.
Typical Thickness by Industry
Wearables and medical sensors sit between 0.2 mm and 0.6 mm. Mobile phones and camera modules use 0.6 mm to 0.8 mm, often HDI. General consumer electronics and computer peripherals use 1.0 mm to 1.6 mm. Automotive control units are usually 1.2 mm to 1.6 mm, with thicker boards where connector forces demand it. Power supplies and inverters use 2.0 mm and heavier copper. Telecommunications and networking equipment use 2.4 mm, and industrial backplanes with press fit connectors run from 3.2 mm upward. These are conventions rather than rules, but they reflect decades of accumulated experience about what survives assembly and service in each application.
Why 1.6 mm Became the Global Standard
Three practical reasons converged on 1.6 mm. It is stiff enough that a typical board holds its shape through reflow and handling without a carrier, so it resists warpage and drop damage. It matches the pin length of the through hole connectors and card edge guides that were standardised decades ago and never changed. And it is the thickness at which the commonly used cores and prepregs are cheapest and most available, which feeds back into the price. The result is a self reinforcing standard: designers specify it because it is available and proven, and laminators stock it because designers specify it.
How Thickness Affects Cost
Thickness itself is not expensive; the deviation from standard is. Boards built at one of the stocked thicknesses cost less than custom stacks, particularly in prototype quantities where setup dominates. Very thin boards become difficult to handle and to keep flat, and very thick boards need longer drilling and more careful plating, so both ends of the range carry a premium. As an indication of where the money goes, a two layer FR-4 board at 1.6 mm might run from about 5 to 15 dollars per piece in modest quantity, a four layer board 20 to 80 dollars, a six layer 60 to 150 dollars, an eight layer board 120 to 350 dollars, and an HDI build at 0.8 mm from 150 to more than 800 dollars depending on layer count and line width. Thickness tolerance, copper weight and material choice then move those numbers again.
How to Choose
Start from the connector, the enclosure and the mechanical requirement, because those are the constraints that cannot be negotiated. Then check that a standard laminate thickness can satisfy the stackup and the impedance targets, and adjust the layer count or the dielectric thickness rather than inventing a new board height. Confirm the tolerance with the fabricator, especially for thin boards and for controlled impedance designs where the dielectric thickness is part of the electrical specification. Finally, review the choice against warpage, weight and cost before releasing the design, since changing thickness after the stackup is frozen means redoing the impedance calculation. Detailed stackup guidance sits in PCB design and layout, the thicknesses a supplier can actually build are listed in PCB capabilities, and the process that holds the finished dimension is described in PCB manufacturing. A prototype PCB assembly run is the cheapest way to confirm that the thickness, the footprints and the connectors agree before volume production in 2026.
FAQ
What is the most common PCB thickness? 1.6 mm, or 0.063 inch, is the default for the majority of industrial and consumer boards because it balances stiffness, connector compatibility and cost.
What is the thinnest PCB that can be manufactured? Around 0.2 mm is achievable for simple thin boards, and thinner flexible circuits are possible, but handling and flatness become the limiting factors rather than the laminate.
Does a thicker board improve thermal performance? Only slightly through added copper, and it adds thermal mass. Heat is normally managed with copper area and thermal vias rather than with board thickness.
Can I specify a non standard thickness? You can, but the price and lead time suffer because the stack has to be built from available cores and prepregs. Choosing a stocked thickness is almost always better.
How does thickness tolerance work? The finished board is specified with a tolerance, commonly plus or minus ten percent for standard builds, and the fabricator holds it by controlling the press cycle and the copper thickness.
Conclusion
PCB thickness is one of the first decisions in a design and one of the hardest to change later, because it fixes the stackup, the impedance geometry, the connector interface and the mechanical behaviour all at once. Start from the standard chart, keep the stack on a stocked laminate thickness, confirm the tolerance with the fabricator, and check the choice against the connector and the enclosure before the stackup is frozen. Handled that way, the thickness stops being an arbitrary number on a drawing and becomes part of the design specification.
