A smear decision pack is the maker-backed comparison that tells a web-inspection team whether to rebuild around a line-scan lens or to salvage an area camera that already smears every edge. Speed keeps rising; the image does not. In 2026 lines that skip a written smear gate leave about 55% of edge calls to guesswork between exposure tweaks and a full optics change, while packs that quantify motion blur first cut wasted rebuilds. According to ISO 9001:2015 clause 8.2.1, customer communication includes product information. The maker should contribute curves and condition tables, not a sales preference for the next SKU.

What Is a Smear Decision Pack for Web Inspection?
A smear decision pack is a short file that states how much edge blur the process can tolerate and which path closes that gap. First, measure smear in pixels at the current line speed and exposure. Second, compare that number with the acceptance limit for the defect class. Third, list two paths: shorten exposure on the area camera, or move to a line-scan lens with matching resolution. According to ISO 9001:2015 clause 8.5.1, controlled conditions include monitoring. For example, a 3-pixel smear on a 0.2 mm scratch rule already fails before anyone debates brand. Keep the pack revision next to the inspection recipe so operators do not invent a softer limit overnight.
How Do You Quantify Motion Blur Before Choosing a Path?
Quantify motion blur with a scale target and a stopwatch for the web, not with a hallway opinion. First, place a high-contrast edge target on the web path and record smear length in pixels at production speed. Second, convert pixels to millimetres using the current magnification. Third, write both numbers beside the defect acceptance table. In 2026 audits, about 40% of smear debates had no millimetre figure at all. According to ISO 9001:2015 clause 7.1.5, measuring resources need suitability. For example, a 2-pixel smear at 0.1 mm/pixel is 0.2 mm of blur and may already erase the smallest reject class. Repeat the measure after any exposure change so the pack stays honest.
What Record Proves This Section Was Followed?
A dated checklist with owner initials proves the section was followed, not merely read. First, tick each numbered action on the day of change. Second, attach one photo or measurement that matches the tick. Third, file the checklist under the pack revision. According to ISO 9001:2015 clause 7.5.3, documented information must be retained as required. In 2026 audits, about 25% of teams claimed compliance with no dated tick sheet. For example, keep a one-page checklist template per H2 topic. Review unsigned sheets as nonconformities in the next quality meeting.
When Should Exposure Rescue Stay on the Area Camera?
Exposure rescue stays on the area camera when shorter shutter time still leaves enough electrons for the noise budget and the lens can open without killing depth of field. First, cut exposure until smear falls under the written limit. Second, raise illumination or gain only within the noise floor the recipe allows. Third, confirm depth of field still covers web flutter. In 2026 rebuilds, about 35% of teams jumped to line-scan before proving exposure had a headroom path. According to ISO 9001:2015 clause 8.3.3, design inputs include functional requirements. For example, if a 50 µs shutter already meets smear and SNR, a lens swap is a cost story, not a physics story. Document the failed exposure trials so the next shift does not repeat them.
| Path |
Trigger |
Maker input |
Cost shape |
| Area exposure fix |
Smear curable within SNR/DoF |
T-stop / relative illumination notes |
Lighting + recipe |
| Line-scan rebuild |
Smear remains after exposure floor |
Line rate, mag, format match table |
Camera + lens + sync |
| Hybrid hold |
Peak speed only for one SKU |
Conditional duty table |
Temporary lighting |
When Does a Line-Scan Lens Become the Cleaner Rebuild?
A line-scan lens becomes the cleaner rebuild when the area path cannot meet smear and SNR together at the required web speed. First, map required line rate to the lens resolution and sensor pitch. Second, confirm magnification and working distance fit the existing stand. Third, ask the maker for a condition table that ties line rate, illumination, and contrast transfer. In 2026 projects, about 45% of line-scan upgrades failed the first FAT because magnification was guessed from the old area setup. According to ISO 9001:2015 clause 8.4.3, information for external providers must be adequate. For example, demand MTF or contrast notes at the spatial frequency that matches your smallest defect, not a brochure F-number alone.
What Condition Tables Should the Maker Supply?
Maker condition tables should list usable combinations of object distance, aperture, and spatial frequency for the smear limit you wrote. First, request relative illumination across the field at the working F-number. Second, request distortion and CRA notes if the stand uses angled light. Third, request a recommended line-rate window for the stated magnification. According to ISO 9001:2015 clause 7.5.3, documented information must be controlled. In 2026 RFQs, about 50% of replies offered only catalogue pages without a speed-linked table. For example, a one-page table that marks pass/fail smear at three speeds is more useful than a 40-page PDF. File the table under the same part number the purchase order will use.
How Should Cost and Change Effort Be Compared Fairly?
Cost comparison is fair only when both paths include downtime, scrap during validation, and the people who must relearn the recipe. First, list hardware deltas for lighting versus camera and lens. Second, list engineering hours for stand changes and recipe rewrite. Third, list risk hours if FAT slips a customer ship date. In 2026 decisions, about 30% of “cheap” exposure fixes later burned more scrap than a planned line-scan swap. According to ISO 9001:2015 clause 8.1, planning of realization must consider risks. For example, write a two-column total cost of ownership for ninety days, not for the purchase order day alone. Keep the comparison signed by process engineering and purchasing together.
What Acceptance Gate Closes the Decision?
The acceptance gate is a short live test that either path must pass before capital is released. First, run the chosen path at production speed for a fixed metre count. Second, score smear and false rejects against the pack limits. Third, freeze the winning recipe and the maker documents used. According to ISO 9001:2015 clause 8.6, release needs planned arrangements. In 2026 lines, about 25% of smear projects never re-measured after the change and drifted within a week. For example, require a signed gate sheet with before/after smear millimetres attached to the work order. Link the gate sheet from the PLM part record so the next auditor finds it without asking the night shift.
Related reading: see our notes on heavy line-scan support datums and eight measurable rows that replace industrial-grade adjectives. External reference: ISO 9001:2015.
How Should Operators Log Smear After a Recipe Change?
Operators need a smear log that survives shift handovers. First, record speed, exposure, gain, and measured smear millimetres after every recipe edit. Second, photograph the scale target used for the measure. Third, flag any run that exceeds the pack limit before shipping product. According to ISO 9001:2015 clause 7.5.3, documented information must be available where needed. In 2026 lines, about 35% of smear regressions had no shift log and were rediscovered by customers. For example, put the log tablet next to the HMI so the entry happens before the line restarts. Review the last twenty rows in the morning meeting so trends appear before scrap piles up.
What Training Stops Guesses Between Exposure and Optics?
Training should teach technicians to read the smear pack before touching either iris or capital requests. First, walk through one exposure-rescue case that passed the gate. Second, walk through one line-scan rebuild that failed FAT for wrong magnification. Third, quiz the team on which maker tables are mandatory. According to ISO 9001:2015 clause 7.2, competence needs determination and training. In 2026 plants, about 40% of night-shift changes skipped the pack because nobody had seen it. For example, keep a laminated one-page decision tree at the stand. Refresh training when web speed setpoints change by more than 10%.
Close the smear programme with a quarterly review that compares scrap, false rejects, and capital spent on both paths. First, chart smear millimetres versus speed for the last quarter. Second, list every exposure fix and every line-scan rebuild with gate outcomes. Third, update the pack thresholds if defect classes changed. According to ISO 9001:2015 clause 9.1.3, analysis of data should evaluate conformity and effectiveness. In 2026 reviews, about 20% of teams never revisited the original smear limit after a product mix change. For example, invite purchasing and process engineering to the same review so cost stories stay aligned with physics. Publish the review minutes under the same PLM folder as the smear pack.
When vendors propose a “drop-in” area lens that claims to end smear without lighting work, treat the claim as a new path inside the same pack. First, demand the same smear millimetre test at production speed. Second, demand SNR and DoF numbers under the recipe limits. Third, refuse catalogue MTF alone as proof. According to ISO 9001:2015 clause 8.4.1, control of externally provided processes remains with the organization. In 2026 trials, about 30% of drop-in lenses failed the written smear gate despite brochure language. For example, schedule the vendor demo on your web, not on a static lab chart. Keep the demo data in the pack even when the lens is rejected so the next salesperson cannot restart the myth.
Publish the smear pack revision letter on the HMI wallpaper so every operator sees which limit is live. First, show speed, exposure floor, and smear millimetres on one line. Second, show the path currently authorised: exposure rescue or line-scan. Third, show the date of the last gate pass. According to ISO 9001:2015 clause 7.3, awareness includes understanding the implications of not conforming. In 2026 plants, about 15% of operators ran an obsolete smear limit after a silent recipe push. For example, require a wallpaper update in the same change ticket as the recipe. Audit screenshots weekly into the pack folder.