Placement Order and Pad Positioning in PCB Layout

Placement is the half of layout that decides how difficult the other half will be. Two engineers can route the same schematic into boards with the same component count, and one will finish in a week while the other fights for space on every layer. The difference is almost always the order in which the components were placed, and the discipline that placement order imposes.

Start With the Core Device

The core device is the component that everything else serves: the microcontroller, the processor, the converter or the analog front end. It is placed first, in the position that the mechanical constraints and the signal flow diagram indicate, normally toward the centre of the board or where the majority of its connections are shortest.

Placing the core first fixes the geometry for everything else. Its supply pins define where the decoupling has to go, its clock pins define where the crystal or oscillator must sit, and its interfaces define where the connectors need to be. A design that places the connectors first and then tries to fit the core between them usually ends up with signal paths that are longer than necessary and with a power distribution network that has to route around the device instead of under it.

Power Follows Immediately

The power module comes next, positioned according to the supply requirements of the core device. A regulator should sit close enough to its load that the distribution impedance is low, and its input and output capacitors belong adjacent to its pins rather than at a convenient distance. The power traces between the regulator and the load should be short and wide.

Placing the power section before the interfaces is what keeps the supply network from being routed around obstacles later. Where several rails are needed, each with its own regulator, the placement of those regulators is a compromise between the load they serve and the space available, and that compromise is easiest to reach while the rest of the board is empty.

Clock, Analog and Interface Sections

The clock source follows the power section, because the oscillator is the most sensitive component on the board and its layout is defined by the device it serves. The crystal and its load capacitors belong immediately adjacent to the pins, with no other routing beneath them, and the ground around the oscillator should be local and well stitched rather than shared with the switching supplies.

Analog sections are placed next, at the end of the board furthest from the digital activity, with their own reference arrangement. Interface components come after that: the connectors, their protection and filtering, and the transceivers that drive them. Finally the passive components fill in around the placed parts, positioned mainly to support the devices they serve rather than to fill the remaining space. The decoupling capacitors belong to this stage as well, and their position is determined by the pins they serve rather than by the space that happens to be free.

Pad Positioning and Its Effect on Routing

Once a component is placed, the geometry of its pads determines how the copper leaves it. The pad position relative to the direction of the connection is the deciding factor, and the rule is simple: a trace should leave a pad from its centre and continue in the direction the pad faces, rather than leaving at an angle or from a corner.

That rule matters most for connectors, where the mechanical stress of mating and unmating acts on the escape. A trace that leaves a connector pad at a tangent creates a stress concentration at the point where the copper meets the pad, and repeated insertion can tear it. The same reasoning governs the escape geometry for other components, and the techniques are described in this article on escape routing and pad fanout.

Pad positioning also determines whether two pads on the same net should be joined directly. Where two adjacent pads share a net, connecting them with a short trace between the pads removes the solder mask dam and creates a bridge that is easy to short during rework. The correct approach is to escape each pad normally and join the traces away from the pads.

Placement Symmetry and Assembly

Placement has an assembly consequence as well as a routing one. The two pads of a chip component must be thermally balanced, since an imbalance produces shift or tombstoning during reflow. Components of similar height should be grouped so that the stencil can print evenly and so that the placement nozzle can approach without colliding. Large parts should be distributed rather than clustered, both to balance the mechanical load and to avoid creating a region of the board that heats more slowly than the rest.

The pad geometry that supports all of this, including the dimensional rules for size, hole diameter and spacing, is set out in this discussion of PCB pad design standards.

A Placement Sequence Worth Following

The order that works is mechanical parts first, since they are fixed; core device second; power third; clock fourth; analog and interface sections fifth; and passives last. Within each stage, the goal is to fix the geometry that the next stage depends on, so that the designer is never routing around a decision that should have been made earlier. A board placed in that order is usually easier to route, easier to assemble and cheaper to build than the same schematic placed in an arbitrary sequence.

Passive Placement and Decoupling

Passive placement deserves its own note because it is usually treated as filler work. The decoupling capacitor belongs at the pin it serves, with its ground connection returning to the plane through a via adjacent to the pad, and that requirement fixes its position more tightly than any preference about tidy component arrangement. A capacitor placed where the space happened to be free rather than where the pin is has already lost most of its value.

The remaining passives should be arranged so that they do not obstruct the routing of the nets they support. A pull-up resistor belongs near the pin it drives, a series termination belongs near the driver, and a filter belongs where the signal enters the section rather than in the middle of it. Each of those rules is about shortening a loop, and the loop length is what determines whether the component does its job. Where the placement leaves a choice about how the traces run, the principles described in this article on ground routing and power trace planning apply to those connections as well.

FAQ

Why place the core device before the connectors? Because the core determines where its own support components must sit, and because its interfaces dictate which connectors are needed and where they can reach. Placing connectors first fixes the outline around them and forces the core into whatever space remains, which usually lengthens the critical paths.

Should power components be placed before or after signal components? Before. The supply network has to reach the devices it feeds with low impedance, and that requirement becomes harder to satisfy once the rest of the board is occupied. Placing the regulators and their capacitors first leaves the routing free to serve them.

Why should a trace leave a pad from its centre? To avoid creating a stress concentration where the copper meets the pad. This is most important at connectors, where mating forces act on the escape, but the same principle applies to any pad that carries mechanical load or repeated movement.