Plumbing and sanitary ware
Catch chrome plating defects, ceramic glaze pinholes, valve casting porosity, gasket seating, and threaded-connection geometry before fittings leave the line.
Automated quality inspection for plumbing fittings and sanitary ware lines, running on a refurbished iPhone alongside your polishing cell, plating tank, glaze booth, and leak-test rig.
What is automated quality inspection for plumbing and sanitary ware production?
AI defect detection for plumbing and sanitary ware uses a camera and a vision model to watch every fitting as it leaves the polishing cell, the plating tank, the glaze booth, the leak-test rig, and the pack station, and to flag non-conforming units before they reach the wholesaler. Instead of an operator at the inspection bench or rigid rule-based vision, the model learns the casting geometry, finish target, glaze colour, and gasket geometry of your SKU portfolio, and applies a consistent visual checkpoint across shifts, line speeds, and finish changeovers.
Plumbing fittings and sanitary ware are particularly hard to inspect at line speed because the brass casting reflects differently after polishing than after chroming, the matt-black PVD finish reads similarly to a clean dark casting under foundry lighting, and the glaze pinhole that ruins a wash-basin looks identical to a speck of overspray to a tired eye. Rule-based vision built around a single fitting reflectance breaks the moment you change finish, basin shape, or batch — which is why most search results for AI defect detection in plumbing return sewer-camera content rather than line-side QC. The category is genuinely under-served, and the iPhone-based deployment closes the gap.
Defects we catch on plumbing and sanitary ware lines
Chrome and PVD plating coverage
Plating coverage defects are the thin-spots, drips, and uneven gloss caused by rectifier-current drift, anode-spacing wear, or PVD chamber pressure variation. Thin-spots and edge skips fail the wholesaler's display sample and trigger pallet returns. Operators check polishing and plating at the rack but cannot watch every fitting under matched lighting, so the borderline cases pass the inspection bench. The AI model learns the in-spec finish for each SKU and flags drift as soon as the local reflectance crosses your tolerance, with the frames available so you can adjust the rectifier or rebalance the rack before a full batch ships out of spec.
Ceramic glaze pinholes and crawling
Glaze defects on wash-basins, WC pans, and shower trays include pinholes, crawling, blistering, and uneven gloss caused by glaze-spray viscosity, kiln-temperature drift, or substrate-cleanliness errors. Pinholes on the bowl interior fail at the showroom and trigger consumer complaints. Manual operators catch the obvious craters but miss the pinholes that pass the booth and only show under installer-side lighting. The AI model holds the visual signature of an in-spec glaze for each SKU and flags pinholes, crawling, and blister patches as soon as the local pattern deviates from spec.
Valve-body casting porosity
Casting porosity defects in brass and zinc valve bodies include shrinkage cavities, surface blowholes, and inclusions caused by foundry pour-temperature drift, mould-vent wear, or alloy-batch tolerance. Porosity at the threaded port fails the leak test on a building site and triggers a warranty return. Operators sample valves at break but miss the porosity that opens up only after machining. The AI model learns the in-spec casting signature and flags shrinkage, blowholes, and inclusions at the post-machining bench, with the frames available so you can adjust pour temperature or vent the mould before a full batch ships.
Gasket and O-ring seating
Seating defects are the missing, twisted, or trapped O-rings on cartridges, taps, and shower hoses, caused by feed-bowl misalignment, gasket-batch tolerance, or operator assembly drift. Missing or twisted O-rings fail leak-test and trigger first-pressure failures on the installer side. The defects ruin the warranty profile and fund the call-out cost. The AI model picks up the visual signature of an in-spec O-ring in a single frame and flags missing, twisted, or trapped rings at the assembly station, before the leak rig sees the part.
Threaded-connection geometry
Thread defects on tap unions, hose connectors, and supply pipes include wrong pitch, cross-threaded starts, burrs at the lead-in, and short or long thread length caused by cutting-tool wear, fixture drift, or alloy-batch tolerance. Cross-threaded fittings fail the installer's first attempt and trigger a returned package through the wholesaler. Manual operators check thread engagement at the inspection bench but miss the burrs that pass the gauge and damage the installer-side fitting. The AI model learns the in-spec thread profile for each SKU and flags pitch, lead-in, and length deviation at the post-machining bench.
Surface scratches and polishing marks
Surface defects include light scratches, polishing wheel marks, and tooling drag caused by polishing-belt wear, pad-pressure drift, or fixture misalignment. The worst offenders sit on the visible spout face and pass the inspection bench to fail at the showroom. Operators catch the obvious gouges but miss the light scuffs that pass the polishing cell and show under display lighting. The AI model learns the in-spec finish and flags scratches, drag marks, and polishing artefacts at the polishing-cell exit so the line can replace the belt or adjust the pad before a full batch ships.
The lighting setup that makes this work on a plumbing and sanitary ware line is a diffuse overhead light with cross-polarisation over the polishing cell and the plating rack to read finish, plus a low-angle spot light at the glaze booth and the leak-test rig to read pinholes and gasket seating. An iPhone Pro with macro and wide-angle lenses handles the seven defect families from a single inspection station per critical control point. We synchronise the rig with the conveyor encoder so that flagged fittings drive a downstream divert or hold decision. We spec the optics with you during onboarding.
How Enao runs on a plumbing and sanitary ware line
The full hardware rig costs less than €1,000 and consists of a refurbished iPhone Pro, a diffuse overhead light with optional cross-polarising filter and a low-angle spot light for glaze and gasket inspection, a USB-C cable, and a mount that clamps over the polishing cell, the plating rack, the glaze booth, the leak-test rig, or the pack station. PLC integration is not required for the first deployment, the rig fits in a flight case, and the line keeps running while you set it up.
Onboarding is self-serve. Your line team mounts the rig, opens the Enao app, and starts collecting reference frames at the next changeover. Day one returns 80% accuracy without any prior labelling, and by day fourteen the model is operating above the manual inspector on the defect families it has seen, improving with every flagged fitting that the line confirms or rejects.
Each line teaches its own model what its casting geometry, finish target, and glaze colour look like. When you swap to a different finish or basin shape on the same line, the model adapts in a single shift. When you bring a sister line online with a similar product family, the second model starts from the first model's experience and the marginal effort drops sharply.
Out-of-spec fittings stop reaching the pack station, scrap is logged at the inspection point rather than at the QC office, and your operators get back the hours of attention they need for the parts of the job that still need a human, including polishing setup, plating rack-out, and warranty-return analysis.
How Enao compares to manual inspection and traditional machine vision
For plumbing and sanitary ware producers the comparison sharpens around five dimensions.
Setup time on a sanitary ware line. — Manual visual inspection: hours of training per operator, ongoing labour. Traditional machine vision: three to nine months of integration with a system integrator, plus a rule set per fitting and finish. Enao: deployed in a week by your own team, day one at 80% accuracy.
Hardware cost per line. — Manual visual inspection: none upfront, ongoing labour cost. Traditional machine vision: €40,000 to €200,000 per line for industrial cameras, structured lighting, and integration. Enao: under €1,000 per line with a refurbished iPhone Pro, lamp, and mount.
Handling new finishes, basins, and glaze colours. — Manual visual inspection: re-train operators for every new SKU. Traditional machine vision: rewrite the rule set per finish, often outsourced to the integrator. Enao: re-teach the model on new fittings, finishes, and glazes in a single shift, no code to touch.
Detection accuracy on subtle plating drift and glaze pinholes. — Manual visual inspection: high at shift start, drops measurably after three hours. Traditional machine vision: strong on dimensional checks, weak on subtle plating drift and glaze pinhole detection. Enao: learns finish, glaze, and gasket signatures from reference frames and holds accuracy across shifts and runs.
Who runs it. — Manual visual inspection: trained operator at the inspection bench. Traditional machine vision: system integrator or a specialised vision engineer. Enao: your line team, no external specialist required.
Wholesalers and merchants change vendors over the cost of a returned pallet, and the cost of a chargeback or a quiet specification swap sits well above the cost of an iPhone-based inspection rig. Enao is built for that gap.
