Solder mask and via treatments look independent on a design drawing. In the fab they often collide. Tight pad spacing that leaves little room for a reliable mask dam, order notes calling for via tenting while the Gerber opens the same holes, resin-plug requirements that clash with open-window data, or paste-layer openings that never show up in the solder-mask layer — these are the problems that keep generating engineering queries during CAM review.
This page is for PCB designers, layout engineers, and manufacturing engineers who want to understand why these conflicts show up, what process limits actually drive the CAM decisions, and how clearer documentation can stop most of them from reaching the review stage. Everything here comes from patterns we see when real customer data hits the manufacturing sequence.
Here is a quick map of the four recurring EQ categories this article covers.
| EQ Category | Typical CAM Query | Main Trigger or Source of the Issue | Typical CAM Consideration / Resolution |
|---|---|---|---|
| Solder-mask bridge feasibility | Can the bridges between these SMD/IC or BGA pads be retained, or do we open the windows? | Pad-to-pad spacing too tight for a reliable dam | Confirm open-window decision and notify customer of the change |
| Via opening vs covering | Order flag says cover/tenting but Gerber shows openings or mixed treatments on the same board | Mismatch between order-system selection and solder-mask Gerber data | Clarify which instruction takes precedence; confirm if selective treatment is intentional |
| Resin via plugging scope | Which vias need resin plug, including guard/fence holes? Plug required but mask still opens them | Unclear via population list; process sequence conflict between fill and open mask | Define exact via list and resolve open-versus-plug conflict before proceeding |
| Paste/assembly vs solder-mask layers | Extra openings exist in paste or placement data — do they need to be copied into the mask layer? | Design-tool openings present in assembly layers but missing from official solder-mask Gerber | Confirm which openings are intentional fabrication features versus assembly-only data |
How Solder Mask and Via Treatments Interact in PCB Fabrication
Solder mask does two jobs at the same time. It protects copper that should stay insulated and it opens the areas that need solder or probe access. Via treatments — open, tented, resin-plugged, or capped — happen at different points in the process. Once a via is filled and planarized, the mask that follows has to work with that surface. If the mask data already opens a via, tenting becomes hard to pull off.
These choices lock in at specific steps. Plating finishes first. Resin fill and planarization, if needed, come next. Solder-mask coating, imaging, and development happen after that. Any mismatch between the intended via treatment and the mask openings stays invisible in the design tool. It only shows up when CAM checks the layers against the real process sequence.
That is why these conflicts surface in CAM review and why the same themes keep appearing.
Solder Mask Function and Manufacturing Sequence
Once the copper pattern is done and after any via fill, liquid photoimageable solder mask gets coated, imaged, developed, and cured. The imaging and development steps set the practical limit on how narrow a dam can sit between pads and how tightly an opening can sit around a via.
Because the mask goes on late, earlier variations — copper registration, drill position, plating thickness — can stack up and shrink the window the mask process has to work with. Surface-finish needs add another constraint: areas that need ENIG, OSP, or immersion tin are normally left open. On fine-pitch parts and BGAs the remaining mask still has to keep neighboring pads isolated, so the geometry that reaches the mask stage needs enough clearance.
Via Treatment Methods and Process Order
The four common treatments sit at different points in the sequence and interact with the mask differently:
- Open vias leave mask openings so the barrel stays exposed on one or both sides.
- Tented vias let the mask close over the hole.
- Resin-plugged vias get filled, cured, and planarized before the mask goes on; the mask then usually covers the leveled surface.
- Capped vias get a copper cap after fill; the mask goes over the cap.
Resin fill and planarization lock the geometry before the mask ever sees the board. After that step, an open-window instruction in the mask layer has limited ability to restore an open barrel. If the mask data already opens a via, tenting becomes difficult. The intended treatment needs a clear declaration — ideally in both the order notes and the Gerber data — before the job reaches CAM. Ambiguity between an order-system flag and the actual mask layer is one of the most common reasons a job pauses for clarification.
This table shows where each treatment sits in the process and where the usual conflicts appear.
| Via Treatment | Typical Process Timing | Interaction with Solder Mask | Common Conflict Risk |
|---|---|---|---|
| Open | After plating, openings defined in mask | Mask apertures required over the via | Order says "cover" while Gerber opens the holes |
| Tented | After plating, mask closes over the hole | Mask forms a tent | Gerber opens the via while notes call for tenting |
| Resin-plugged | Fill + planarize before mask coating | Mask usually covers the planarized surface | Plug required yet mask layer still opens the via |
| Capped | Copper cap after fill, then mask | Mask applied over the copper cap | Cap present but open-window data remains in the mask layer |
Solder Mask Bridge Failures in Fine-Pitch and High-Density Areas
A solder-mask bridge is the narrow strip of mask left between two adjacent pads. Its job is to keep solder from bridging those pads during reflow. When the finished copper spacing falls below what the imaging and development process can hold, the bridge becomes hard to form reliably. CAM then faces a practical choice: open the windows or hold the job.
This limit shows up most often on SMD and IC footprints and inside BGA arrays. The same pattern keeps appearing in review: pad spacing is too tight to produce a bridge, or the bridges in the BGA region are hard to retain. In most cases the outcome is the same — the bridges get omitted and the pads share a continuous open area at the mask level.
Pad Spacing Limits That Prevent Reliable Bridges
Bridge feasibility depends on the finished copper geometry and the minimum feature the mask process can hold. Registration between the copper pattern and the mask image further reduces the usable window. When the nominal clearance is already near the process edge, normal variation can thin the bridge, break it, or leave residue on the pads. At that point the usual CAM decision is to open the windows rather than risk incomplete coverage.
Why Bridges Collapse or Are Omitted in BGA and Fine-Pitch Zones
BGA and fine-pitch layouts deliberately shrink pad-to-pad spacing to hit density targets. The combination of tighter copper, tighter mask registration, and the need to keep openings aligned with the pads leaves little margin. When retention looks impractical, CAM converts the bridges to open windows and notifies the customer. Trying to force a sub-minimum dam introduces yield risk most fabricators prefer to avoid.
Effects of Converting Bridges to Open Windows
Once the bridges are gone, neighboring pads share an open area at the mask level. During assembly the chance of solder bridging goes up, especially under fine-pitch parts or BGAs where paste volume and reflow already sit near the edge of control. The change needs documentation and communication; otherwise the assembly house may still expect the original isolation. Incomplete notification just moves the problem downstream.
Via Opening Versus Covering Conflicts Encountered in CAM
Another frequent finding is the mismatch between an open-window requirement and a cover requirement on the same vias. It can show up as an order-system flag that selects "cover oil" while the Gerber shows openings, or as mixed treatments on one board — some vias opened on both sides, others covered. These get flagged because the two instructions are hard to execute on the same feature.
Order Specification Versus Gerber Data Mismatch
The order system and the Gerber data are independent sources of truth. When they disagree, software has limited ability to decide which one wins. Typical patterns include:
- The order interface is set to via tenting or cover oil, yet the supplied mask layer contains openings over those vias.
- Fabrication notes call for open vias while the mask layer is closed.
- One side is specified open and the opposite side covered, but the data does not reflect the split.
Resolution needs a clear customer answer. The CAM engineer has little room to invent a middle ground that satisfies both contradictory instructions.
Mixed Via Treatments on a Single Board
Selective opening or covering is sometimes intentional — test vias left open, signal vias tented, thermal vias plugged. The rationale can be valid, but the data needs to declare the selection clearly. When the same region contains both open and covered vias with no clear rule, the fabricator has trouble deciding which treatment applies to which hole. A layer-by-layer or via-by-via definition removes the guesswork before the job reaches plating and mask.
Resin Via Plugging Scope and Process Conflicts
Resin plugging is a multi-step sequence: drill, plate, fill, cure, planarize. The leveled surface then receives solder mask. Because the fill happens before the mask, any later open-window instruction on a plugged via creates a direct process conflict. Extra friction appears when the plug scope is unclear — especially whether guard or fence vias around a BGA or connector are included — and when an explicit VIA Plug layer contradicts cover or open notes elsewhere in the package.
Defining Which Vias Must Be Plugged
The decision to plug is driven by reliability, planarity, or assembly needs. Guard vias or via fences are often meant to stay open or tented, yet the order note may simply say "all vias resin plugged." Without a clear population list or a dedicated plug layer that identifies the intended holes, CAM needs to ask. Over-plugging adds cost and time; under-plugging leaves the design intent unmet.
Plugging Versus Solder Mask Opening Incompatibility
Once a via is filled and planarized the surface is essentially solid. An open-window instruction in the mask layer has limited ability to recreate an open barrel; at best it leaves a shallow depression. The two requirements therefore conflict on the same feature. The usual CAM path is to hold the job and ask which instruction takes priority — plug and cover, or open and leave unplugged.
VIA Plug Layer Versus Cover/Open Requirements
An explicit VIA Plug layer is useful when it is the single authoritative list of holes that receive resin. When that layer exists yet other notes or the mask data still call for open or tented treatment on the same holes, the conflict reappears. Consistency across order notes, fabrication drawing, plug layer, and mask layers is the most practical way to keep the query from appearing.
Layer Consistency Problems Between Paste/Assembly Data and Solder Mask
Solder-paste and pick-and-place layers are generated for assembly; they are not fabrication layers. When those layers contain openings that do not exist in the official solder-mask Gerber, CAM needs to decide whether the extra openings are intentional mask features that were omitted or assembly-only data that should be ignored. Leaving the mismatch unresolved risks either missing mask openings or exposing copper the designer never intended to expose.
Extra Openings in Paste or Assembly Layers
Design tools sometimes place paste apertures or assembly outlines that are larger than, or offset from, the corresponding mask openings. In other cases a via or test point is opened in the paste layer but left covered in the mask layer. If the extra openings get copied into the mask data without review, the finished board may expose copper that was meant to stay covered. If they are ignored, the assembly process may expect openings that are not there.
Synchronization Requirements Across Layers
Whether an opening that appears in the paste or assembly layers needs to be added to the solder-mask layer is a judgment call. If the opening is meant to stay on the finished board — over a via, test point, or thermal pad — it belongs in the mask data. If it exists only for stencil or placement purposes, it is better left out of the mask layer.
A short note in the fab documentation is usually enough. Stating that paste openings are for reference only, or listing the specific openings that should be synchronized, keeps the CAM engineer from having to guess the designer's intent. Without that clarity the job often pauses for a clarification email.
What CAM Review Actually Flags and How Conflicts Are Resolved
The four themes above produce a recognizable pattern of engineering queries. CAM engineers typically ask for confirmation of bridge retention or open-window preference in fine-pitch areas, clarification of via treatment when order flags and Gerber disagree, definition of the exact via population that requires resin fill, and resolution of any mismatch between paste-layer openings and the solder-mask data. Each query holds the job until the customer answers.
Common Clarification Requests Raised by CAM
The questions that come up most often are straightforward:
- Should the solder-mask bridges between these SMD or BGA pads be kept or opened?
- Which vias are intended to be open, tented, or resin-plugged when the order flags and Gerber data disagree?
- Are the guard or fence vias included in the resin-plug requirement?
- Do the extra openings that appear in the paste or assembly layers need to be added to the solder-mask layer?
These are not formalities. The instructions as written are hard to follow at the same time, so the fabricator needs to ask.
Resolution Paths and Their Manufacturing Implications
Once clarification arrives, the path is usually one of three options: accept the data as-is with the residual risk noted, modify the Gerber or notes to a consistent treatment, or request a design change. Leaving the issue half-resolved leaves yield or assembly risk on the table. The longer the clarification loop runs, the more it pushes the schedule. Aligning design intent with the actual process sequence before the data is released remains the most practical way to avoid most of these loops.
Engineering Considerations That Reduce Solder Mask and Via Failures
Most of the conflicts described above are preventable. The common root is a gap between the designer's intended treatment and the data that actually reaches CAM. Closing that gap takes deliberate checks at the design-to-process interface and consistent documentation.
Design-to-Process Alignment Checks
Before releasing the data, check that the copper pad geometry leaves enough spacing for any required solder-mask bridges, that the vias intended for plugging are clearly identified, and that the solder-mask layer matches the final open or cover decision for every via. Also confirm that paste-layer openings do not introduce unintended mask features. These checks work best when they are done against the real process sequence rather than a generic design-rule checklist.
Documentation Practices That Minimize Ambiguity
A single, consistent statement of via treatment — ideally in both the fabrication notes and an explicit layer — removes most open-versus-cover conflicts. When selective treatment is needed, a clear rule or a dedicated plug layer eliminates guesswork. Notes that state whether paste openings are for reference only, or need to be synchronized, stop the layer-consistency questions before they start. Vague language such as "cover oil where possible" or "plug all vias" without a population list simply hands the decision to CAM and invites delay.
FAQ
Q1: Why do solder-mask bridges fail more often in BGA areas?
A1: BGA pad pitch is usually the tightest on the board. The tighter spacing plus normal registration tolerance leaves little room for a reliable mask dam. When the bridge is hard to form cleanly, CAM converts the area to open windows.
Q2: What is the difference between via tenting and via opening?
A2: Tenting lets solder mask cover the via and form a closed tent over the hole. Opening leaves a mask aperture so the via barrel stays exposed. The two treatments are hard to apply to the same via; the data needs to declare one or the other.
Q3: Can a resin-plugged via also have a solder-mask opening?
A3: In practice the two conflict. Resin plugging and planarization happen before solder mask is applied. Once the via is filled and leveled, an open-window instruction has limited ability to restore an open barrel.
Q4: Should openings that appear in the paste layer be copied into the solder-mask layer?
A4: Only if those openings are intentional fabrication features. Paste-layer data is generated for assembly; extra apertures that do not belong in the solder-mask Gerber are better left as assembly-only data. A clear note prevents CAM from having to interpret the intent.
Q5: How can designers reduce the number of solder-mask and via engineering queries?
A5: Align the design intent with the actual process sequence early, declare via treatments consistently across all documentation, and verify that solder-mask, plug, and paste layers do not contradict one another before the data is released for fabrication.
