Skip to main content

4.5 SPC & Dashboards

SPC and dashboards bring discipline to the production floor by turning raw process data into clear, real-time guidance. Instead of waiting for defects to surface, live charts track the vital signs of printing, placement, and reflow so small drifts are corrected before they escalate. The key is simplicity: a handful of metrics, split by family where it matters, presented in a way operators can act on without hesitation. With alarms tied to predefined corrections and every change logged against time and product, dashboards become a stable feedback system rather than a noisy distraction.

4.5.1 What SPC is (in factory words)

Statistical Process Control =(SPC) is the operational discipline that converts raw machine data into watchpredictive intelligence. By continuously charting the processvital while it’s runningsigns, not justof the partsSMT after.line, YouSPC pick a fewenables live signals, chart them simply, and agree what happens when they drift. The goal isn’t pretty graphs—it’s early, small corrections thatto keepprocess drift before defects fromever forming.occur. Dashboards are the visual interface for SPC, providing

real-time guidance to operators and engineers. The success of an SPC system is measured by its simplicity and the immediate, effective action it compels when a metric drifts.

4.5.1 SPC Fundamentals in Production

ThinkSPC ofinvolves itmonitoring key performance indicators (KPIs) to distinguish common cause variation (inherent, acceptable process noise) from special cause variation (a sudden, specific problem like a car:clogged nozzle or broken sensor).

  • SpeedometerProcess Control: SPC SPIis transferabout efficiency (are we printingcontrolling the rightmean volume?(µ) and the variation (σ) of critical parameters like Solder Paste Transfer Efficiency (TE).
  • FuelThe lightGoal: The Mounteraim starvation/miss rates (are feeders aboutis to bitemaintain us?)
  • Engineprocesses temp → Reflow TAL/peak/ΔT (are we cooking the board right?)

If any gauge moves, you nudgein a knobstate of nowstatistical control, notensuring afterthat AOIthe screams.process is consistently

capable


(C
pk
≥ 1.33
) of meeting specifications (Chapter 4.1).
  • Visual Control: Simple charts must be prominently displayed so operators can read the process status at a glance—like a fuel gauge or a thermometer.
  • 4.5.2 TheKey “few,Process vital” metricsMetrics (byBy station)Station)

    StartEffective withSPC these.requires Addtracking morethe only"few, ifvital" theyparameters changethat decisions.predict downstream failure, rather than monitoring everything.

    Station

    Key Metric

    WhyIndicator it mattersof...

    IfResponse it drifts… first correctionTriggered

    Printer (SPI)

    TE CpkCpk by familyFeature Family (chips, QFN edge, BGA, thermal total)

    Predicts bridges, tombstones, bridges,HIP HIP, voidsrisk.

    Immediate cleaning cycleClean sooner;or easesqueegee bladepressure pressure; slow separation; tweak hot apertures next build


    First-Print Pass %

    Measures recipe stability

    Look for paste age, bead size, support pinsadjustment.


    Wipes/Board & Reprint RateTrend

    Spot paste/stencilStencil/paste hygiene issuesdrift (Chapter 1.5).

    Adjust cleaning interval;interval; checkinspect stencil nano-coat; refresh beadcoating.

    PnP

    Placement time/board & Starvation minutesMinutes/Hour

    BottleneckFeeder healthmanagement discipline (Chapter 2.3).

    Splice earlier; rebalance split; loadLoad spare feeder; adjust splice alarm threshold.


    Miss/Retry rateRate by feederFeeder

    Catches badBad tape, peel,worn nozzle, or vision instability.

    Fix peel angle; clean nozzle;path; quarantine reelfaulty component reel/feeder.

    Reflow

    TAL / Peak Mean

    Solder (meanwetting +completeness range)and thermal margin.

    Wetting and HIP control

    Small belt/belt speed or zone nudge;temperature confirmnudge paste(Chapter limits3.2).


    ΔTCross-Board at peak∆T

    EvenConvection heatinguniformity indicatorand thermal profile stability.

    Adjust blower speedRaise blower;or lengthensoak soak;zone addsetpoint supports(Chapter 3.1).

    AOIInspection (AOI/AXI)

    False calls/boardCalls/Board &(by Top 3 refs

    Touch-time and noiseRefDes)

    Programming noiseTighten lightingand profile;manual tweakverification 3costs rules, not 30(OpEx).


    EscapesProgrammer found later

    Safety net reality check

    Raise limits for Class A only; fix process if trend

    AXI

    Void % (QFN/BGA)review &and collapsetuning flags

    Hidden-jointof quality

    Adjustspecific windowing/soak;lighting/ROIs consider(Chapter N₂4.4).




    4.5.3 SimpleDashboards chartsand thatAlarm operatorsTiers

    actually

    Effective readdashboards use visual alerts to enforce a standardized response protocol. Every deviation must be tied to a documented action (the "if this, then that" playbook).

    A) Visualization

    • Control Charts: The I-MR (Individuals &and Moving Range) chart is standard for tracking continuous data like SPI TE and Reflow TAL/peakPeak, pershowing panel.the process mean (µ) and the variation (σ) over time.
    • p-chartsTraffic Lights: forGreen/Yellow/Red AOIstatus falseindicators calls/board.
    • Runclearly chartscommunicate (sparklines)the forhealth starvationof minutes,each misses/retries, wipes/board.parameter.

    B) Alarm Tiers

    KeepAlarms must be tiered to prevent nuisance alerts from overwhelming the visualsoperator, dumb-clear:ensuring trafficonly lightscritical (green/yellow/red),issues lasthalt 50 panels, and a vertical line labeled “change made” whenever someone tweaks a recipe or cleans optics. If people can’t read it at 2 meters, it’s too fancy.production.




    4.5.4 Alarm tiers (decide once, use everywhere)

    Tier

    What triggers itCondition

    WhatMES happensResponse

    Management automaticallyAction

    InfoGreen

    FirstAll yellowmetrics hitwithin (e.g., SPI TE outside center but inside spec)3σ.

    LogContinuous &flow.

    No showaction amber badge on station; no stoprequired.

    WarnYellow (Warning)

    2 of 3 panelsconsecutive indata yellowpoints outside 2σ or trenda run of 6 points trending up/downdown.

    Printer:Log add clean; PnP: check feeder; Reflow: nudge belt/zone; AOI:warning, send to small review queue

    Stop

    Any red (spec violation) on critical family or repeat warn with no improvement

    Block WIP; open short NCR; require quick sign-offalert to resume

    supervisor,

    Make the MES or line PC enforce this—humans forget when the line is hot.




    4.5.5 Closing the loop (the “if this, then that” map)

    Tie each dashboard needle to a small, specific action:

    • SPI TE low on chips (Cpk <1.0) → reduce blade pressure one notch and shorten clean interval; recheck next 3 panels.
    • SPI area creep near fine pitch → slow separation 10–20%; if persists, wet+vac, then inspect apertures at next stencil spin.
    • PnP miss spikes on Feeder 12 → verify peel path, run 20 dry picks; if still bad, swap feeder; quarantine reel lot.
    • Starvation minutes climbing on constraint → pull next splice earlier (countdown threshold), load spare pre-threaded feeder.
    • TAL short on cold TC → slow belt a touch (time first), then re-aim late zone by +3–5 °C.
    • ΔT at peak > target → add soak 10–20 s via mid-zone setpoints; tick blowers up one step.
    • AOI bridges rising in one region → check printer separation and cleaning; don’t widen AOI until print is fixed.
    • AXI QFN voids trending up → verify paste lot/date; lengthen soak slightly; consider N₂ A/B run.

    These moves should be printed next to the screen. No guessing.




    4.5.6 Roles & views (one dashboard does not fit all)

    • Operator view (per station): 3–5 gauges, next-action button (“Add clean now”, “Check Feeder 12”), and clear pass/stop banner.
    • Lead / Supervisor: line balance (cycle time per machine), red/yellow counts per hour, top 3 issues to clear this shift.
    • QE/PE weekly: Cp/Cpk trend by family, escapes vs AOI/AXI limits, before/after of any recipe or stencil change.
    • Management: OEE (Availability/Performance/Quality) plus first-pass yield. No microscope photos.




    4.5.7 Make data trustworthy (plumbing that matters)

    • Everything gets an ID: recipe revs (printer/PnP/oven), stencil ID, paste lot, feeder IDs, operator badge, WO/panel/SN.
    • One time base: all stations sync to the same clock so panel histories align.
    • Auto-attach evidence: SPI/AOI/AXI images and reflow plots stored against the panel ID; no screenshots in chat apps.
    • Annotate changes: pressing “changed belt speed +5 mm/min” drops a vertical line and a note on charts.




    4.5.8 Start small (a rollout that sticks)

    1. Pick one product and wire SPI TE Cpk, wipes/board, PnP misses, TAL/peak, AOI false calls.
    2. Agree alarm tiers and the first corrections (Section 10.5.5).
    3. Run two weeks; collect “we touched X, chart did Y” stories.
    4. Freeze the layout; clone to the next product with only limit tweaks.




    4.5.9 Common traps (and quick escapes)

    • Too many charts. Result: no one looks. Trim to the five that drive actions.
    • Alarms with no playbook. Result: people mute them. Add the button that does the first fix.
    • Averages across families. Result: masked trouble. Always split by feature family (chips/QFN/BGA/thermal).
    • Goal-post moving. Result: pretty charts, bad product. Change the process, not the spec, unless customer agrees.




    4.5.10 Pocket checklists

    Designing the dashboard

    • 5–7 metrics total; split by feature family where relevant
    • Green/Yellow/Red bands defined; MES enforces Stop tier
    • One-clickrecommend first correction per(e.g., alarm"Wipe Stencil Now").

    Supervisor enforces immediate corrective action.

    DailyRed use(Stop)

    Any

    • Operatorsdata clearpoint warns withexceeds the buttonedSpecification action;Limit notes(USL/LSL) auto-loggedor a sustained violation of the Cpk floor.

      Hard stop of the line, mandatory engineering sign-off to resume.

      Open Non-Conformance Report (NCR) and conduct root cause analysis.

      4.5.4 Closing the Loop (The Corrective Cycle)

      The primary function of the SPC system is to enable the closed-loop feedback that drives continuous process improvement.

      1. Detection: An alarm tier (Yellow or Red) is triggered by a metric drift (e.g., SPI TE µ drops 5 points).
      2. Correction:Leads reviewThe topoperator 3executes issuesthe atpredefined break;First assignCorrection quick(e.g., fixesincrease squeegee speed slightly). This must be a simple, single-parameter fix.
      3. Annotation:QE samplesThe escapesaction vstaken AOI/AXIis andautomatically feedslogged backagainst tothe processtime/product ifon trend

    Changethe control

    • Anychart, recipe/stencilcreating change =a vertical line +labeled comment"Process Change." This logging is critical for auditing the process.
    • Verification: The system monitors the next 3–5 panels to verify the correction successfully brought the mean back toward the center without increasing the variation.
    • Engineering Fix: If repetitive corrections fail, the issue is deemed Special Cause (e.g., worn stencil, faulty feeder) and requires Engineering/Quality approval to implement a permanent solution, which updates the Golden Recipe.
    • Final Checklist: SPC Implementation

      Requirement

      Control Point

      Management Focus

      Metrics

      Select and track the fewest possible predictive metrics (TE Cpk, TAL/Peak, Miss Rate).

      Prevents data overload; ensures focus on chartshigh-risk areas.

    • Action Plan

      Weekly Cp/Cpk & Pareto review; onlyEvery smallYellow/Red rulealarm edits,tier notmust broadbe relaxtied to a documented
      First Correction playbook.

      Eliminates human guessing and wasted time during drift events.

      Data Integrity

      All machines must sync to a single time base; every touch and recipe change must be auto-logged and annotated on charts.

      Ensures data is trustworthy for root cause analysis.

      Visualization

      Dashboards use simple control charts and traffic lights that clearly show the control limits.

      Promotes immediate, effective operator response.




      When SPC focuses on the few signals that matter, enforces consistent alarm responses, and feeds results back into process control, it keeps production calm and predictable. The outcome is fewer surprises, steadier yields, and inspection systems that spend less time firefighting and more time confirming that the line is already on track.