Selective Soldering Nozzle Selection for Through Hole Assembly

Selective soldering solders the through hole joints of a board that also carries surface mount devices, using a small nozzle of flowing alloy rather than a full wave. It is selected when a connector or terminal block cannot survive a second full thermal cycle, or when the assembly is too dense for a pallet to shield the top side. Results depend far more on three settings than on the machine itself: the selective soldering nozzle, the dwell time, and the thermal profile that supports them. This article explains how those variables interact and how to choose a nozzle for a given through hole connector.

Where Selective Soldering Fits

Selective soldering exists because different areas of the same board have incompatible thermal limits. A board may carry fine pitch devices on the top side and a heavy connector on the bottom, so the assembly is reflowed first and then soldered from below with a local heat source. Only the through hole joints see the second thermal event, which keeps the surface mount side inside its own process window.

The alternative is a wave soldering machine with a pallet that masks the surface mount area. That works when the board has enough free space for the pallet to seal against, and when the connector geometry lets the wave reach every pin. Selective soldering wins when pin counts are moderate, when the board is large, or when the masking pallet becomes so elaborate that its handling risk exceeds the benefit it provides.

Nozzle Geometry and Wave Shape

A selective soldering nozzle is a tube with a defined inner diameter and an orifice shape, and it sets both the contact patch and the flow pattern at the joint. A small diameter produces a narrow, high velocity stream that suits single pins with tight spacing, while a larger diameter produces a broader, calmer pool that can cover several pins at once. Too large an orifice on a fine pitch connector floods adjacent pads and bridges them.

The wave height must be set so the alloy touches the joint without climbing the pin. When the nozzle sits too high the wave wraps the component body and leaves solder on the shoulder, and when it sits too low the joint is only partially wetted. Nitrogen blanketing raises wetting force and reduces oxide, which is useful on a narrow stream where the surface to volume ratio is large, but it changes the thermal behaviour and must be included when the profile is developed.

Dwell Time and Contact Behaviour

Dwell time is the interval during which the wave is in contact with a given joint, and it controls how much energy the joint receives. Short dwells leave an incomplete fillet or an unfilled barrel because the alloy never reaches the far side of the board. Long dwells overheat the laminate, lift the mask and grow an unnecessarily thick intermetallic layer. A useful starting range for a standard pin is two to four seconds, adjusted upward as thermal mass rises.

Dwell is not the same as cycle time. A board with twenty joints can be soldered in a single pass with a multi nozzle head or joint by joint, and the per joint contact time is what must be controlled. Copper thickness, barrel plating quality and the surrounding copper area all change the heat a joint needs, which is why the same recipe rarely transfers between board families without verification.

Flux Application and Wetting

Flux application sets the surface chemistry before the alloy arrives. A droplet or micro spray system deposits a controlled volume on the joint area only, which limits residue and keeps flux away from parts that cannot be cleaned. Dipping the whole board is simpler but wets areas that do not need flux and leaves more residue to remove later.

The volume matters as much as the method. Inadequate flux leaves oxides on the pad and the pin, so the alloy beads instead of wetting and a joint that looks full may have no metallurgical bond. Excess flux is not self correcting either: it creates spatter, generates voids and, with a no clean chemistry, can leave an activated residue that later attacks the joint. The target is a thin, complete film across the joint area, confirmed by inspection rather than by the number on the panel.

Thermal Profile for the Assembly

A selective soldering thermal profile has a preheat phase and a solder phase, and the two must be developed together with the dwell. Preheat brings the joint and the surrounding laminate to a temperature where the alloy wets quickly without a long contact, and it must be gradual enough that the board does not distort. Bottom side heaters and top side convection both contribute, and their balance decides whether the joint reaches temperature from the pin or from the pad.

Verification follows the same logic as reflow profiling. Thermocouples attached to real joints, including the coldest and the hottest ones on the board, show the actual heat history, which is the only thing that matters. The techniques used for reflow oven profile verification apply directly here, and a profile validated on a bare test coupon usually understates the load that populated joints place on the heat source.

Fixturing and Board Support

Support decides whether the measurements taken during programming remain valid in production. A board that is free to bow changes its distance to the nozzle, so wave contact varies from one joint to another even though the recipe is unchanged. Dedicated fixtures with local supports under connectors and adjustable pins control that distance and make the process repeatable.

Warpage is the usual cause of a joint that solders correctly at the start of a batch and poorly at the end. Design features that keep the board flat, described in PCB warpage control, reduce the problem at its source, and the fixture then only has to compensate for the remaining variation instead of a curve that changes with temperature.

Programming, Verification and Defect Signatures

Programming a selective machine means teaching positions, then correcting them against a sample soldered under production conditions. The first article should be judged against acceptance criteria for fill, fillet and wetting, including a look at the barrel on a cross sectioned sample rather than only a visual check. Criteria for that judgement are set out in solder joint acceptance criteria.

Defects point back to specific parameters. Bridging indicates an oversized orifice or a wave that is too high, incomplete fill indicates insufficient dwell or preheat, and solder balls or spatter indicate too much flux or a wet surface. Recording the parameter set next to each defect and reviewing it weekly is how a selective line stays stable, and it is the same discipline the gopcb process team applies when a customer transfers a connector assembly into volume production.

A first article check confirms that the process and the drawing agree on the points listed above, and that the coupon data supports the values used in the design.

Where a measurement falls outside the expected window, the sample is retained so that the cause can be established before the balance of the batch is released.

FAQ

Can one nozzle cover every joint on a board? Rarely. A nozzle wide enough for a heavy connector usually floods fine pitch pins, and a narrow nozzle is too slow for a large connector. Most programs use two or three nozzles with separate recipes.

How often should the dwell recipe be revalidated? After every change of alloy supplier, flux chemistry, board finish or fixture, and at least once a year otherwise. Alloy and flux changes shift wetting behaviour enough to move the optimum dwell by a second or more.

Is nitrogen always necessary? No. It improves wetting and reduces dross on narrow streams, but a well controlled process with adequate flux can meet the acceptance criteria without it. Decide on measured fill and defect rate, not on the specification sheet.