Wave Soldering: Design Rules and Process Limits

Wave soldering is a mature process that still produces a characteristic set of defects: bridging between adjacent pins, icicles and webs, insufficient fill in a plated through hole, solder balls and a dull, grainy surface that suggests the alloy or the flux has changed. Almost all of them can be traced to one of five variables: flux application, preheat, wave geometry, solder pot condition and the thermal design of the board itself.

Flux Application

Flux has to reach the underside of the board evenly and in the right quantity. Too little leaves oxide that the wave cannot displace, producing skips and incomplete fill; too much deposits residue that interferes with test and, in some chemistries, creates a cosmetic problem that looks worse than it is. Spray fluxing is the most common method, and its coverage depends on the spray pattern, the air pressure, the board speed and the condition of the nozzles. Spray fluxing control covers the setup details that most affect consistency.

Foam fluxing is still used where a thicker deposit is wanted, and it demands its own discipline: the flux density, the foam height and the depth of immersion all change the deposit, and the flux itself ages as water is carried in by the boards. Whichever method is used, the applied weight of flux is the measurable parameter, and it should be checked on a defined schedule rather than inferred from the appearance of the wave. Flux application control is where most skipped-joint investigations should start.

Preheat

Preheat does two jobs: it activates the flux and it reduces the thermal shock that the board would otherwise experience at the wave. The topside temperature is the practical measurement, and it has a target range that depends on the flux chemistry and the board thickness. Too little preheat leaves the flux inactive and the board cold, producing poor wetting and heavy thermal stress. Too much dries the flux before the wave arrives, which is a common cause of skips that appear on only part of the board and are then blamed on the wave.

The distribution of preheat matters as much as its average. A board with large copper planes takes longer to reach temperature than one with sparse circuitry, and a top-side measurement taken at one point may not represent the whole surface. Where a board is known to be thermally uneven, measuring at several points is worth the extra effort, because the defects that follow an uneven preheat are difficult to distinguish from flux problems.

Wave Geometry and Contact

The wave itself is shaped by the nozzle, the pump speed and the solder level, and the contact between the board and the wave is set by the conveyor angle and the conveyor speed. A wave that is too high floods the board and produces bridging; one that is too low leaves an incomplete contact and produces skips. The contact length is the parameter that ties the height and the speed together, and it should be recorded for each product rather than adjusted by feel.

Dross is the constant companion of an open solder pot. It forms on the surface, and if it reaches the wave it appears in the joint as a rough inclusion or a void. The standard controls are a nitrogen blanket to slow its formation, a regular skimming schedule and a chip wave or a dross separator where the equipment provides one. Dross that is left in place long enough also changes the alloy composition of the pot, which produces a defect that no amount of nozzle adjustment will fix.

Solder Pot Condition

The pot is a chemical system as well as a thermal one. Tin, copper and other elements dissolve from the boards and from the equipment into the alloy, and the composition drifts away from the specification. Copper in particular rises with the volume of plated through holes processed, and the change raises the liquidus temperature, which reduces the wetting window. Periodic solder pot alloy analysis is the only way to know whether the pot is still within specification.

Temperature control belongs to the same discipline. The setpoint may be correct while the actual temperature varies with the load, and a pot running below its setpoint produces the same symptoms as a low preheat. Recording the pot temperature at the start and end of a shift, and checking it against a calibrated probe periodically, prevents a slow drift from being attributed to the flux or to the board.

Board and Design Factors

Some wave soldering defects are designed in. A pad that is too large holds more solder than the joint needs and encourages bridging, while a pad that is too small produces a weak fillet. A thermal relief that is too restrictive prevents the hole from reaching temperature, and one that is too generous drains solder away from the joint. Components placed too close to each other shadow the wave and produce skips that no process setting can correct.

The orientation of the board on the conveyor is a design decision with the same weight. Parts that present a long row of leads to the wave are more likely to bridge than the same parts turned ninety degrees, and connectors that act as a dam create turbulence that produces icicles. Reviewing the layout for wave soldering, rather than only for placement, is the cheapest way to remove these defects. Wave soldering quality criteria give the inspection standard that follows.

Defect Response

When a defect appears, the response should follow the process rather than the symptom. Bridging suggests too much solder or too little flux activity, icicles suggest a thermal or a contamination problem at the exit from the wave, and insufficient fill suggests a hole that never reached temperature. Recording which side of the board and which location the defect appears at narrows the list quickly, because a defect that follows one component is a design or a shadowing issue while a defect that appears across the board is a process setting. That single distinction saves hours of trial and error.

Trend data turns these single investigations into control. A daily record of defect counts by type, together with the flux weight, the preheat temperatures and the pot analysis, shows whether the process is drifting and which variable moved first. Most wave soldering problems become obvious in that data long before they become obvious in the yield figure, provided somebody is looking at it.

FAQ

Why do skips appear on only part of the board? Usually uneven preheat, uneven flux coverage or shadowing by a tall component. Measure the top-side temperature at several points before changing the wave settings.

How often should the solder pot be analysed? On a defined interval that reflects the volume of plated through holes processed, and after any event that could contaminate the pot.

Does nitrogen help wave soldering? It reduces dross and improves wetting, particularly with lead free alloys, at the cost of gas and an additional variable to control.

Can bridging be fixed by lowering the wave? Sometimes, but if the pad geometry is the cause the defect will return. Check the design before adjusting the machine.