In surface-mount technology assembly, the stencil printing process is statistically responsible for approximately 60 to 70 percent of all downstream soldering defects. Insufficient paste volume, misregistration, bridging and skipped deposits are root causes that propagate invisibly through reflow and only manifest as measurable faults during automated optical inspection or functional test — at a point where rework costs are significantly higher. Solder paste inspection, commonly abbreviated as SPI, addresses this problem by measuring the printed paste deposit immediately after the stencil printer and before component placement, closing a feedback loop that classical 2D optical inspection could never provide.
The concept of in-line SPI is not new, but the precision, throughput and actionable data output of modern 3D SPI systems have transformed it from a supplementary check into a mandatory process control element on competitive SMT lines. The following overview explains the underlying measurement physics, the key metrics evaluated, integration into the production line and the requirements that determine system selection.
Measurement Principles: From 2D Optical to Structured-Light 3D
Early SPI systems operated exclusively in two dimensions, capturing top-down greyscale or colour images of the printed board to assess paste area and registration offset. While computationally simple and fast, 2D SPI cannot determine paste height or volumetric fill, leaving the most critical parameters unmeasured. A paste deposit that appears correctly shaped in plan view may still be insufficiently tall, entrapping a void or exhibiting a collapsed profile that will cause a solder joint defect after reflow.
Contemporary 3D SPI systems rely predominantly on structured-light triangulation, also referred to as phase-shift moiré or fringe projection. A projector casts a series of sinusoidal stripe patterns onto the board surface at a defined angle; one or more cameras, positioned at a known offset, capture the resulting fringe images. Local height at every pixel is computed from the phase shift of the observed fringe relative to the projected reference. By acquiring multiple phase-shifted fringe sets and applying phase-unwrapping algorithms, the system resolves absolute height across the full field of view with sub-micrometre repeatability under controlled conditions. Some platforms combine fringe projection with colour imaging in a single pass, acquiring both volumetric and 2D texture data simultaneously to maintain line throughput.
Key Metrics and Acceptance Criteria
The primary measurement outputs of a 3D SPI system are paste height, paste area and calculated paste volume for each individual deposit. Volume is considered the most robust indicator of print quality because it integrates height distribution across the entire deposit footprint and is least sensitive to minor shape irregularities. The measured volume is compared against the theoretical aperture volume derived from the stencil design — typically expressed as a percentage. Industry practice, guided by IPC-7711 and process capability studies at leading EMS providers, commonly defines acceptance windows in the range of 50 to 150 percent of nominal volume, though tighter windows down to 70 to 130 percent are applied for fine-pitch and area-array components.