A stencil quote looks like one number, but it is the sum of five choices:
| Driver | Range (SKPCB published) | When it is worth paying for |
|---|---|---|
| Structure | Framed (glue-in) / unframed foil | Volume runs, fixed printer, repeat builds |
| Foil material | Stainless steel (standard) / nickel (electroformed) / Kapton·PET (hand printing) | Fine pitch, low area ratio |
| Foil thickness | 0.08 / 0.10 / 0.12 / 0.15 / 0.20 mm | Set by the finest part and the largest pad together |
| Step | Single thickness / multilayer step | Fine pitch and large pads on the same board |
| Finishing | Grinding polish (standard) / electrolytic polish / nano-coating | Unstable release, frequent wiping |
One more cost never appears on the quote: a remake after a data change. It is often more expensive than any option.
1. Structure: framed or unframed
- Framed stencils: the foil is tensioned and permanently glued into a frame. Frame and tensioning are a one-time cost that buys stable tension and automatic printer alignment, so they suit volume production and repeat builds. SKPCB publishes frame sizes of 370×470, 420×520, 450×550, 550×650 and 584×584 mm.
- Unframed stencils: only the foil is delivered and mounted in a reusable tensioning frame. No frame cost, suited to prototypes and small batches.
- Cost logic: if a program prints only a few times, the frame cost is never spread out. If the job runs continuously, repeated mounting and alignment of a loose foil eats the saving. The frame must also match the printer; the wrong size means a remake.
2. Foil material and thickness
- Stainless steel is the standard material, laser cut, and covers most boards.
- Nickel (electroformed) forms the apertures by electroforming, so the walls are replicated from a mandrel. It earns its place in fine-pitch areas where the area ratio is low; the price is a longer process route.
- Kapton / PET is for hand printing and temporary prototypes, not machine production.
- Thickness itself moves the price little but moves rework a lot. It is constrained from both ends: the finest-pitch part needs a thin foil for area ratio and release; large pads, connectors and shield frames need enough paste. When the two conflict, a single thickness sacrifices one side, which is why step stencils exist.
3. Aperture process and aperture count
- Laser cutting cuts each aperture in turn; more and smaller apertures mean a longer cutting path.
- Electroforming is not timed per aperture, but it needs a mandrel and an electroforming cycle. It applies where laser-cut walls cannot give reliable release. The wall differences between the two routes are covered in the linked article; here the point is cost: electroforming is worth paying for only when the area ratio really falls short.
- Common waste: treating the whole board to the requirement of the finest part. Most boards have only a few areas that need high-spec apertures; the rest is fine with standard laser cutting.
4. Step stencils
A step stencil has zones of different thickness on one foil, so large pads get more paste and fine-pitch areas get less. It adds a thinning or thickening operation and adds design constraints at the squeegee and step edges.
- Worth it: the board has fine-pitch BGA/QFN alongside power parts, connectors or shield frames, and a single thickness gives either bridging or insufficient paste.
- Not worth it: the problem sits on a few pads and can be solved by changing aperture size or shape.
5. Finishing: electrolytic polish and nano-coating
- Grinding polish is standard.
- Electrolytic polish smooths the aperture walls and improves paste release.
- Nano-coating reduces wipe frequency and extends service life.
- Cost logic: both improve release; neither changes aperture geometry. If the area ratio is too low, change the foil or apertures first and treat finishing as a supplement.
6. Hidden cost: remakes and wrong revisions
Outside the quote, the most expensive stencil is the one that cannot be used:
- The paste layer was not final, and the PCB revision changed the pads.
- No paste layer was supplied, so apertures were cut 1:1 from the copper layer.
- Frame size or fiducials did not match the line printer.
- Thickness was set for one part family and yield dropped on another.
7. Configuration by stage
| Stage | Structure | Foil / aperture | Step | Finishing |
|---|---|---|---|---|
| Prototype | Unframed foil | Stainless laser | Avoid; adjust apertures instead | Standard polish |
| Pilot batch | Unframed or framed, depending on repeat builds | Stainless laser; fine-pitch areas reviewed for area ratio | Review for mixed boards | Decide from release behaviour |
| Volume | Framed, matched to the line printer | Laser or electroformed by area ratio | As needed for mixed boards | Electrolytic polish / nano-coating by wipe frequency |
8. Data to freeze before ordering
- Final paste, solder mask and outline layers (Gerber)
- Printer model and frame size
- Location of the finest-pitch parts and the largest pads
- Target foil thickness or range; whether a step is acceptable
- PCB revision, mapped one-to-one to the stencil revision
During data review SKPCB confirms stencil type, foil thickness, apertures and area ratio. Laser or electroforming route, polishing and coating, tensioning and framing are confirmed against the Gerber data, pad layout, material and equipment. Stencils can be scheduled together with a PCBA order.
FAQ
Q: What is the biggest cost item in an SMT stencil?
A: Usually the structure (framed or unframed) and the aperture process route. Thickness moves the price little but has the largest effect on rework.
Q: Do I need a framed stencil for prototypes?
A: Usually not. An unframed foil in a reusable tensioning frame is enough; switch to framed once the job runs repeatedly.
Q: Is an electroformed stencil always better?
A: No. Electroforming solves release for low-area-ratio apertures. Where the area ratio is sufficient, laser-cut stainless steel already does the job.
Q: Can nano-coating replace aperture changes?
A: No. Coating improves release and wipe frequency but does not change aperture geometry. If the area ratio is too low, change the foil or apertures first.
Q: How do I avoid a stencil remake?
A: Freeze the final paste layer and PCB revision before ordering, give the printer frame size, and mark the finest-pitch and largest-pad areas.