In short: Hand soldering suits repair and prototypes, wave soldering suits through-hole parts, and reflow soldering suits dense surface-mount boards, and each process leaves recognisable defects (bridges, cold joints, tombstones) that trace back to heat, wetting or paste volume.
Why soldering process choice matters
A solder joint is both an electrical and a mechanical connection. A board with a 0.4 mm pitch BGA, a handful of connectors and a few hundred 0402 passives cannot be built the same way as a single-sided power supply with big electrolytic capacitors. Choosing the wrong process is a classic source of yield loss, and it is a favourite interview topic because it ties together component packaging, metallurgy and DFM for PCB rules. Layout decisions made early, as described in the PCB design process steps, decide which process is even possible.
All three methods rely on the same physics: molten solder must wet a clean metal surface, form an intermetallic layer (Cu6Sn5 for tin-copper joints), and then freeze into a smooth, concave fillet. Flux removes oxides so wetting can happen. Without wetting, you get a ball of solder sitting on the pad, not a joint.
Hand soldering
A technician applies a heated iron tip to the pad and lead together, then feeds wire solder to the joint rather than the tip. Typical tip temperature for lead-free SAC305 (melting about 217 °C) is 320 to 370 °C, with 2 to 4 seconds of contact. Leaded Sn63/Pb37 melts at 183 °C and is far more forgiving, which is why training labs often start with it.
- Best for: prototypes, rework, connectors added after reflow, odd parts that cannot survive an oven.
- Weak point: operator variation. Two people produce two different joints.
- Technique: heat the pad and the lead, add solder to the opposite side of the iron, remove solder first, then the iron.
Key fact: the heat transfer is controlled by thermal contact area, not tip temperature alone. A small tip at a high setting often gives worse joints than the right-sized tip at a moderate setting.
Wave soldering
In wave soldering the populated board rides on a conveyor over a pumped standing wave of molten solder at roughly 250 to 265 °C for lead-free alloys. The sequence is flux spray, preheat to 100 to 130 °C on the top side (to activate flux and limit thermal shock), then the wave, with a contact time of about 2 to 4 seconds. Through-hole leads draw solder up the barrel by capillary action, which is why plated through-hole quality, covered in via types in PCB, matters so much.
Surface-mount parts can be wave soldered only if glued to the underside and oriented properly, since tall neighbours shadow small parts from the wave. Many shops now use selective wave soldering, where a small nozzle solders individual connectors, avoiding the shadowing problem entirely.
Reflow soldering
Reflow is the dominant process for surface-mount assembly. Solder paste, a suspension of 20 to 38 micron alloy spheres in flux, is stencil-printed onto pads, parts are placed by pick-and-place machines, and the whole board passes through a multi-zone oven. A lead-free profile has four stages: preheat ramp (1 to 3 °C/s), soak (150 to 200 °C for 60 to 120 s), reflow with peak at 240 to 250 °C and 45 to 90 s above liquidus (217 °C), then controlled cooling. A deeper look at the thermal curve is in our design flow overview and in the dedicated reflow profile article coming to the PCB track.
Because every joint on the board reaches temperature together, reflow gives the best repeatability for BGAs, QFNs and 0201 passives. The downside is that thermal mass varies: a big inductor lags a tiny resistor by 10 to 20 °C, so profiling with thermocouples on the coldest and hottest parts is mandatory. Packages for BGA escape layouts, discussed in BGA routing techniques, depend on reflow because the balls hide under the package.
Defects and what causes them
Bridging is solder joining two adjacent pads or pins. Causes: too much paste, a stencil aperture ratio that is too generous for fine pitch, slumping paste, or in wave soldering a lead that is too long or a conveyor angle that is wrong. Cold joints look dull, grainy and often have a convex shape with a poor contact angle; they come from insufficient heat, movement during solidification, or contamination. Tombstoning is when a small chip part stands up on one end. During reflow one pad wets first, and its surface tension pulls the part upright before the other pad melts. Unequal pad sizes, uneven heating and misplaced parts are the usual triggers.
Key fact: tombstoning force scales with the wetting difference between the two pads, so symmetric pad geometry and a gentle ramp rate are the cheapest cures.
Other defects worth knowing for interviews: solder balling (flux outgassing from wet paste), voiding in BGA and thermal pads (above about 25 percent area is typically rejected under IPC-A-610 Class 2 or 3 criteria for critical joints), head-in-pillow (BGA ball and paste never merge due to warpage), and insufficient hole fill in wave soldering.
Animation: how a tombstone forms
Which process suits which component
| Component type | Best process | Why |
|---|---|---|
| 0201 to 0805 passives, QFP, QFN | Reflow | Printed paste volume is precise |
| BGA, CSP, LGA | Reflow only | Joints are hidden under the body |
| Large through-hole connectors, transformers | Wave or selective wave | High throughput, good barrel fill |
| Heat-sensitive parts, repairs, prototypes | Hand or hot-air rework | Local heat, no oven stress |
Mixed-technology boards commonly use reflow first, then selective wave or hand soldering for the through-hole leftovers. Dense boards that depend on controlled impedance and clean power delivery also need joints that are electrically sound; see signal integrity in PCB for how poor joints show up as intermittent faults. If you want hands-on training for lab and industry settings, ChipXpert runs hardware-focused programmes.
Practical rules that come up in interviews
- Keep paste stencil area ratio above 0.66 and aspect ratio above 1.5 for reliable paste release.
- Match thermal mass and pad size on both ends of chip parts to avoid tombstoning.
- Use nitrogen reflow for fine pitch to reduce oxidation, accepting higher cost.
- Inspect with AOI for bridges and X-ray for BGA voids; ICT catches opens and shorts afterwards.
- Design tools from Cadence, Siemens (Xpedition, Valor) and Altium can run DFM checks for pad spacing, solder mask dams and wave-shadowing before release.
Ready to go from theory to board bring-up? Explore the PCB design track on our CourseTron courses page and build boards that survive real assembly lines.
FAQ
Why does lead-free soldering produce more defects than leaded?
Lead-free alloys such as SAC305 melt about 34 °C higher and wet more slowly, so the process window between wetting and component damage is narrower.
Can I wave solder surface-mount parts?
Yes, but only chip components and small packages glued to the underside, and shadowing from taller parts is a risk. Selective soldering is preferred for mixed boards.
How do I tell a cold joint from a good joint?
A good joint is shiny (for leaded alloys) or slightly satin (for lead-free), with a concave fillet and low contact angle. A cold joint is dull, grainy or lumpy with a high contact angle.
What prevents bridging on fine-pitch parts?
Use smaller stencil apertures, thinner stencils (100 micron or less), solder mask dams between pads and accurate placement.