On this page
A ceramic via opening may be flush after filling, recessed after drying, disturbed when a face conductor is printed, or changed again during firing. Looking only at the finished cap cannot identify which event created the depression. This guide establishes repeatable local datum planes at four process states and uses matched witness vias to decide whether a refill is needed before subsequent printing. It complements broader via cleaning, firing and electrical-diagnosis pages by focusing on cap-height history and the refill decision.
Key design decisions
- Measure cap height against nearby ceramic at filled, dried, post-print and fired states.
- Keep a protected witness group out of the later print pass to isolate pull-out or displacement.
- Authorize refill only from a state-specific rule tied to final capture-pad and continuity needs.
Name the via state before describing the defect
Use separate labels for wet-filled, dried, refilled, face-printed and fired conditions. Photograph both openings and retain panel coordinates at each state. A depression measured after firing cannot be called drying shrinkage unless the prior state was observed. Likewise, paste found on a screen after face printing does not prove that it came from a via without a mapped cap change.
Define the intended construction: through-filled, wall-metallized or another drawing-specific form. This guide addresses a filled opening with a surface cap that interacts with later printing. It does not convert the geometry into a universal via design recommendation. Record fill direction, backing or support, number of passes and which face is inspected.
Measure cap recession from a local ceramic reference
Establish an annular ceramic reference zone outside the paste meniscus and any capture-pad print. Measure signed cap height relative to a fitted local plane: positive for protrusion and negative for recession. Keep the measurement window and filtering method constant. Large substrate bow or nearby printed layers can corrupt a single global plane.
Sample more than the visual center. A cap can tilt, form a central dimple or retain a raised rim. Report the minimum and representative area rather than one cursor reading. Optical methods can lose data on steep or reflective surfaces; contact methods can deform an unfired deposit. Select the method for the state and declare its limitations.
hcap = zcap - zceramic,local
- hcap: signed via-cap height
- zcap: measured cap surface level at the stated location
- zceramic,local: fitted nearby ceramic reference level
The local ceramic zone is clean, accessible and unaffected by a later face print; state-specific measurement uncertainty is reported.
Capture recession during drying before any refill
Measure the same vias after filling and after the specified drying interval. Record elapsed time from fill to drying, orientation, load condition and paste identity. Solvent loss and particle rearrangement can change volume and surface shape, but the observed change belongs only to the actual formulation and geometry. Do not infer an internal void from a shallow top depression.
Calculate the cap-height change for each via rather than comparing unrelated group averages. A panel-position pattern may indicate fill transfer or drying airflow; random large changes may point to inconsistent fill volume or local hole condition. Keep an unmeasured companion group if the measurement technique could disturb the wet deposit.
Δhdry = hcap,dry - hcap,fill
- Δhdry: state change attributed to the defined drying interval
- hcap,dry: cap height after drying
- hcap,fill: cap height at the documented filled state
Same via, registered location and non-disturbing measurement; later handling or printing has not occurred.
Set a refill rule from downstream geometry
A refill can restore cap volume but can also create protrusion, trap material at the rim or change how the face conductor transfers. Define the refill trigger from the permitted surface condition before the capture pad is printed. Include measurement uncertainty so borderline vias are not alternately refilled and accepted by repeated readings.
Compare no-refill and controlled-refill groups across the observed recession range. Record refill side, tool, paste condition and redrying. Do not assume two partial fills equal one complete fill; interfaces and trapped air may differ. The decision must follow through face printing, firing, physical inspection and electrical continuity before it becomes a drawing-specific process instruction.
Isolate disturbance from the subsequent print pass
Create matched groups with equivalent dried cap heights. Protect one group from the face-conductor print while exposing the other to the normal pass. Measure both afterward. A new depression, smear or material tail only in the exposed group supports print interaction. If both groups change during the same hold, continued drying or measurement drift remains a competing explanation.
Inspect the screen and squeegee path without claiming transferred material originated from a via until location correspondence is established. Record print direction relative to the opening. A raised cap may contact the screen differently from a recessed one, and a capture-pad aperture can either bridge the depression or accentuate it depending on the actual paste and geometry.
| Observed change | Leading hypothesis | Control |
|---|---|---|
| Fill to dry recession | Volume loss or redistribution during drying | Same-via registered height history |
| No-print witness also changes | Continued conditioning or measurement drift | Time-matched protected group |
| Only printed group loses cap material | Screen or squeegee interaction | Matched cap-height exposure pair |
| Refilled cap becomes proud | Excess refill or rim retention | No-refill and graded-refill groups |
| Finished cap looks sound but path is high | Internal or interface issue | Defined electrical path and section |
Follow cap geometry through firing and refiring
Record cap height and surface morphology after the via firing and after any face-conductor or later layer firing. The final shape reflects more than drying. Binder removal, sintering, material compatibility and repeated thermal exposure can change the surface and interfaces. Use the temperature history of the loaded specimen, not a generic paste peak, in the traceability record.
Keep fired observations separate from unfired refill control. A smooth fired cap does not prove a void-free conducting section, while a shallow depression does not automatically mean an open path. Where destructive analysis is justified, select sections from both electrical outliers and matched controls and document whether polishing introduced pullout.
Connect surface history to the final electrical boundary
Measure a path that identifies the via and its face connections. Use documented current, contact positions and specimen temperature, with four-wire sensing where the geometry and resistance level justify it. A capture pad may mask cap recession optically while still producing an acceptable path, or it may bridge a weak surface temporarily and become intermittent later.
Compare electrical results with the exact cap-history group: no refill, refill, protected witness and print-exposed. Do not divide a chain resistance by via count when one defect can dominate or face traces are not removed. The surface study decides when geometry changes; electrical evidence decides whether the defined path meets its drawing-specific requirement.
Validate a refill sequence with traceable witness vias
A focused trial spans the observed dried-recession range and includes no-refill, controlled-refill, no-print witness and face-print exposure groups. Measure cap geometry at each state, inspect both faces and record fired continuity. Balance groups across panel positions so a fill-direction or support pattern cannot masquerade as a refill benefit. Preserve missing and unmeasurable data.
The released instruction should state via geometry, fill direction, paste and support identity, drying state, cap-height method, refill trigger, redrying, face-print sequence and final electrical boundary. Reopen the review when hole geometry, substrate thickness, paste, backing, drying, capture-pad artwork or screen changes. The result is conditional on the evaluated stack and is not a general via-fill capability claim.
Send the via cap-height history
Provide the state-by-state evidence needed to review drying recession and refill.
- Ceramic grade and thickness, hole geometry, intended fill construction and both-face artwork.
- Fill direction, support, paste identity, drying conditions, refill method and all later print/firing steps.
- Registered cap-height measurements and images for filled, dried, post-print and fired states.
- Via electrical boundary, test conditions, affected panel positions and drawing-specific acceptance rule.
The drawing-upload form loads as you reach this section.