Pad Design for Wave Soldering and Reflow

A pad is not just a place to put solder. It is the shape that decides how much solder the joint receives, where that solder goes when it melts, and whether the part stays in place while the alloy is liquid. Two pads that look identical on the screen can behave differently in a wave and in a reflow oven.

This article covers the pad and hole geometry that the two soldering processes demand, from the shared pad that pulls two parts together to the thief pad that empties a bridge, and the solder mask and copper decisions that go with them. Most of the geometry is inherited from the process rather than chosen, so knowing which process will build the board is the first step in getting the pads right.

Shared Pads Between Two Surface Mount Parts

Placing two chip components on one large common pad looks economical, because it saves a connection and a piece of board area. In practice the molten solder on the pad pulls both parts toward the middle of the pad, and the parts are free to move until the alloy freezes, so the final position depends on which side melted first.

The reliable arrangement is two separate pads joined by a narrow trace instead of one wide pad. Where the connection has to carry a larger current, several narrow traces in parallel do the same job without presenting a single large molten area to the parts. A part that shifts during reflow is very often the result of exactly this geometry.

Vias In or Near Surface Mount Pads

A via inside a pad, or close enough that the paste reaches it, is a leak. When the solder melts it follows the hole rather than the lead, which leaves the joint starved, and if the via is open it can carry the solder to the opposite side of the board and bridge whatever is there.

Where a via has to be used to carry heat or current away from a pad, it must be filled and capped, or placed far enough away that the paste cannot reach it and a short trace carries the connection instead. The via filling process exists for exactly this case, and the requirement belongs on the fabrication drawing rather than in a note on a layout layer.

Matching Hole Shape to Lead Shape

Where a component is inserted through the board, the hole and the pad should follow the shape of the lead. A square lead in a round hole touches the plating on four edges at most and leaves a gap that the solder has to fill; a square hole in a square pad with a square lead gives a joint that wets all the way round and looks the way the inspection criteria expect.

Symmetry matters as much as shape. The hole has to sit at the centre of the pad, because an offset hole leaves one side of the pad thin and the solder fillet becomes uneven, which is both a cosmetic defect and a weaker joint. Pad and hole geometry for through-hole parts is covered in more detail in the standard pad geometry rules.

Pads for Wave Soldering: Thief Pads and Solder Mask

A row of fine-pitch pins crossing a wave is the classic bridging case. The solder that is dragged forward by the last pin has nowhere to go, so it joins the next pin. The remedy is a thief pad at the end of the row: a pad separated from the last pin by a short distance, which the surplus solder prefers because it is cooler and has no component to bridge to.

Solder mask between adjacent pads helps in the same direction, because it removes the continuous copper that a bridge can grow along, but it is not a substitute for the thief pad. Where the mask is printed over a pad it also has to wet and cure properly, and the rheology of the mask ink decides whether the opening stays clean and the dam stays square.

Solder Escape and Unwanted Hole Filling

Components that are soldered after the wave have to survive it, and any plated hole that will take its component later must not be filled with solder in the meantime. The way to prevent it is a relief: an extension of the pad in the direction opposite to the wave travel, wide enough for the solder to run back out rather than set around the hole.

The width of the relief is matched to the hole, typically about half a millimetre to one millimetre, and its direction is the one thing about it that is easy to get wrong. The relief points away from the wave, not along it, and the board orientation on the line is therefore part of the pad design.

Copper Area and Self-Centring in Reflow

In reflow soldering, the surface tension of the molten alloy can be an asset. If both pads of a chip component are heated at the same rate and have similar copper area, the part centres itself on the lands as the solder melts, and small placement errors disappear. If one pad has a plane attached to it and the other has a trace, one melts much earlier and the part is dragged toward it instead.

Enlarging the copper connected to the pads evens out the timing and strengthens the pull, which is why thermal reliefs and pad geometry are considered together rather than separately. The same imbalance is a common cause of tombstoning, and it is also the reason that the choice between lead-free and leaded solder changes the process window.

Standardising Lead Pitches for Through-Hole Parts

Every different lead pitch on a board is a different forming tool, a different setting on the insertion machine and a different setup during assembly. Reducing the number of pitches, and using the same pitch wherever the circuit allows, removes work from the line without changing anything about the design.

The saving is easy to underestimate. A handful of axial parts placed at a pitch that is used nowhere else on the board will cost more in setup time than the board area they occupy. A short list of pitches, agreed once with the assembly house, pays for itself on every subsequent build. Where the choice is free, the standard list is the better starting point, and the placement work it supports is described in the notes on placing components in a defined order.

FAQ

Should every wave-soldered row end with a thief pad? Not every row. It is used where the pitch is fine enough that surplus solder cannot drain away on its own, and at the trailing end of the direction of travel.

Can a via be placed under a pad if it is tented? Tented vias leak less but still give the solder a path once the mask is damaged during assembly. A filled and plated via is the reliable answer where the via has to be inside the pad.

Do these rules change with a different alloy? They change in degree. Lead-free solder is less fluid and wets differently, so drainage from a relief or a thief pad is slower and the process window narrows, but the geometry that helps remains the same. Where a board is built with one alloy in prototyping and another in volume, the pad geometry is the one thing that carries over unchanged.