Most stencil enquiries arrive with exactly one number attached to thickness: "0.12 mm, same as last time." That is understandable, but it skips the actual decision. Foil thickness is not a quality dial where thinner means finer and thicker means more robust. It is a value squeezed from two directions at once — pushed down by the finest pitch on the board, pushed up by the largest paste-hungry pad.
This article covers that one decision variable only. How area ratio works arithmetically is covered in aperture design, area ratio and paste release; material and frame tension are in stencil material and frame tension selection; the construction of multi-level stencils is in step stencil multi-level aperture design; the fabrication routes are compared in laser cutting versus electroforming. What follows is the question that comes after all of those: does your specific component mix still have a solution at a single foil thickness?
1. Two effects that pull in opposite directions
Thickness acts in two places simultaneously, and that is the entire conflict.
Upward — paste volume. Printed volume is aperture area multiplied by foil thickness. For the same aperture, 0.15 mm delivers half again as much paste as 0.10 mm, purely arithmetically. Large connector pads, shielding frames and power components need that volume to form a load-bearing joint.
Downward — release capability. Whether paste leaves the aperture at all is described by area ratio. Our SMT stencil fabrication page defines it as aperture area ÷ wall area, with the IPC rule of thumb requiring at least 0.66. Wall area grows with thickness. The thicker the foil, the lower the area ratio at an unchanged aperture — and below 0.66 paste stays in the aperture, producing insufficient paste or open joints.
The same number that supplies volume for the connector takes release away from the fine-pitch pad. That is why "thinner is finer" is a badly formed statement. Thinner is release-friendly and volume-poor; thicker is volume-rich and release-critical.
2. The lower bound is arithmetic, not taste
From the two already-published figures quoted above — the definition of area ratio and the value 0.66 — the minimum aperture follows directly. The transformations below are pure arithmetic from those two inputs, not Sunking measurements and not company-specific limits:
- Round aperture, diameter D, thickness t: area ratio = D ÷ (4t). Requiring ≥ 0.66 gives D ≥ 2.64 · t.
- Rectangular aperture, width W, length L: area ratio = (W · L) ÷ (2t · (W + L)).
This moves the discussion from opinion to calculation. A case that recurs in practice: a 0.20 mm round aperture, as found on very small BGA ball patterns, requires t ≤ 0.20 ÷ 2.64 ≈ 0.076 mm by this calculation. Our thinnest standard step is 0.08 mm, so the aperture already sits below what this geometry and the 0.66 guideline allow. At that point no choice of thickness helps; the conversation has to move to aperture shape, aperture wall condition or fabrication method — which is exactly what the engineering review before cutting exists for.
Conversely, for an elongated fine-pitch aperture of, say, 0.25 × 1.30 mm, the calculation gives an upper bound near 0.158 mm, so 0.15 mm still works and 0.20 mm does not. At 0.20 × 1.20 mm the same calculation lands near 0.13 mm, making 0.12 mm the last step that satisfies the condition.
Please read these numbers as worked examples of the method, not as release values. Our fabrication page states explicitly that exact thresholds, aperture sizes and foil thickness are confirmed during engineering review against the customer data and equipment, and that the released stencil data after engineering review governs the actual apertures.
3. The steps we actually build
Thickness selection does not happen on a continuum. It happens on a ladder. Per our SMT stencil fabrication specifications we build foil thicknesses of 0.08 / 0.10 / 0.12 / 0.15 / 0.20 mm. Stainless steel foil is standard; nickel applies to electroformed stencils, and Kapton or PET to hand printing.
The practical consequence: if your calculation yields an upper bound of 0.158 mm, the answer is 0.15 mm, not 0.158 mm. And when the calculation lands between two steps, the deciding question is which direction of error hurts less. Too little paste produces open joints; too much produces bridging. Which of those two failure modes dominates in your layout is a property of your board, not of the stencil.
4. The conflict: one board, two irreconcilable requirements
Now the case this article is really about. A typical industrial or control board carries, at the same time:
- a fine-pitch device whose apertures force an upper bound around 0.12 mm by the calculation in section 2;
- a large connector or shielding frame whose joint only becomes load-bearing at substantially more paste volume.
At a single foil thickness there is no value that satisfies both. This is not an optimisation problem where you meet in the middle. A compromise at 0.15 mm violates the release condition at the fine pitch; a compromise at 0.10 mm starves the connector. Both produce failures, at different locations on the same board.
For exactly this case, the scenario table on our fabrication page lists multilayer / step stencil against "mixed paste volumes on one board": different thickness zones on one stencil let large connector pads take more paste while fine-pitch areas take less. That is the constructive answer to a conflict that a single number cannot resolve.
The second option is to split the print: two stencils for the same side, in two print passes. That costs cycle time and requires a clean separation of component groups in the data. Our specification shows that we build top, bottom, or top and bottom sides — on one stencil or on separate ones. Which route is right in your case depends on quantity, line cycle time and the spatial distribution of the component groups, and is confirmed after data review.
5. What thickness cannot fix
A significant share of print problems is reported as a thickness problem while being something else. The defect table on our fabrication page sorts this out:
- Bridging arises when apertures are larger than the pads, when there is a gap between stencil and board, or when separation speed is too high. The countermeasure given there is to keep apertures 5–10% smaller than the pads for standard components, check support and print force, and slow the separation. A thinner foil does not close a gap.
- Paste offset from the pads traces back to unclear or missing fiducials, or to the absence of fiducial-based alignment. That is a data issue, not a thickness issue.
- Uneven deposit thickness can come from declining stencil tension, a worn squeegee, or indeed from mixed paste volumes on one board. Only the last of those three relates to the thickness decision.
- Insufficient paste is the case where thickness genuinely participates: area ratio below 0.66, rough aperture walls, or paste residue in the apertures. The listed countermeasures are a thinner foil or larger apertures, supplemented by electropolishing or nano-coating and shorter wipe intervals.
The distinction matters because changing foil thickness means a new stencil. Answering an alignment or tension problem with a thickness change buys a tool and keeps the defect.
6. What thickness co-decides, though it does not look like it
Three side effects of the thickness decision are routinely missed.
Aperture reduction and thickness are coupled. The "5–10% smaller than the pads" rule for standard components reduces aperture area — and therefore area ratio too. Deriving thickness from nominal pad size and then shrinking the aperture can silently drop you below 0.66. The correct order is: fix the aperture you actually intend to cut, then check thickness against it.
The aperture wall is part of the calculation. Release depends not only on geometry but on wall condition. That is why our specification lists mechanical polishing as standard and electropolishing as an option that improves paste release, plus the optional nano-coating that reduces residue in the apertures. If your calculation sits close to 0.66, that is the lever to examine first, ahead of the next thickness step.
The frame limits the usable field. Our common frame sizes are 370×470, 420×520, 450×550, 550×650 and 584×584 mm. For a panel that runs close to the edge, the frame participates in deciding framed versus frameless — framed stencils are intended for repeat printing at higher volumes, frameless foils for small batches and prototypes.
7. What we need for a defensible thickness recommendation
So that thickness selection does not default to "same as last time", the engineering review needs the following. Four items live in your data; two you have to state actively.
- Gerber RS-274X with paste, solder-mask and outline layers, ODB++, or DXF — the input formats listed on our page. Without a paste layer there is no aperture discussion.
- The finest pitch on the board and its pad geometry, since that produces the upper bound on thickness.
- The most paste-hungry components — connectors, shielding frames, power devices — as the lower bound on volume.
- Whether both groups sit on the same side. Only that determines whether a section 4 conflict exists at all.
- Usage frequency and quantity, which drive framed versus frameless.
- Fiducials on the paste layer and your printer's alignment mode.
Engineering review happens before any cutting: we extract the paste layer from your Gerber or ODB++ data, cross-check it against pads and solder-mask openings, and flag missing fiducials or outline edges. Area ratio and aspect ratio are reviewed group by group against component pitch. Aperture issues therefore come back to you at that stage, not after the print.
If the aperture design also needs checking against what the board itself can carry, minimum trace width, drilling and outline tolerances are in our manufacturing capabilities and tolerance standards. For the link between stencil and inspected assembly, our quality assurance page is the reference, and board material questions are covered in the material overview. For a concrete thickness recommendation, submit your data through request a quote; an overview of available board types is under products.
FAQ
Which foil thickness is right for my SMD stencil?
There is no universally correct thickness, only one that fits your board. The upper bound comes from the finest pitch through the area ratio ≥ 0.66 condition; the lower bound comes from the most paste-hungry component. We build 0.08 / 0.10 / 0.12 / 0.15 / 0.20 mm; which of those works for your data is confirmed during engineering review before cutting.
Is a thinner stencil always better for fine pitch?
For release yes, for volume no. A thinner foil raises area ratio and helps small apertures, but it simultaneously reduces printed paste volume at every other pad. On a mixed-technology board that starves the connector while the fine pitch looks good.
What happens if area ratio falls below 0.66?
Paste stays in the aperture. The result is insufficient paste and incomplete release, up to open joints. The countermeasures listed on our fabrication page are a thinner foil or larger apertures, supplemented by electropolishing or nano-coating and shorter wipe intervals.
My layout has a 0.4 mm pitch device and a large connector. Can one thickness do both?
In many of these cases, no. When the release condition at the fine pitch forces a thin foil and the connector needs more volume, no single value satisfies both. The multilayer or step stencil exists for exactly this scenario — different thickness zones on one stencil. The alternative is two stencils and two print passes.
Will a different thickness solve my bridging problem?
Usually not. Bridging typically comes from apertures larger than the pads, a gap between stencil and board, or excessive separation speed. The first thing to address is keeping apertures 5–10% smaller than the pads for standard components, checking support and print force, and slowing the separation.
What data do you need to determine thickness?
Gerber RS-274X with paste, solder-mask and outline layers, ODB++, or DXF. Also useful: the finest pitch, the most paste-hungry components, quantity and usage frequency, and your printer's alignment mode.
Does thickness change the frame decision?
The two decisions are independent but confirmed together. The frame question follows usage frequency: framed stencils for repeat printing at higher volumes, frameless foils for small batches and prototypes. Common frame sizes are 370×470, 420×520, 450×550, 550×650 and 584×584 mm.