Pack inspection
How do you check tilted caps and uneven cap height?
Check a cap against an approved bottle-and-closure reference around the full circumference, not from one viewing angle. Compare cap height, visible tilt, thread engagement, liner or tamper-feature position and bottle-neck condition. Manual gauges, height sensors or vision can support inspection, but the method must be proven with real acceptable and deliberately representative fault samples.
Inspection starting point
Define the approved cap-height reference before inspecting
Use the sequence below to define an approved pack datum, distinguish tilt from uniform high seating and choose inline or offline checks that match the actual fault. Do not treat cap presence alone as proof of correct thread engagement.
Use the same bottle datum, viewing orientation and accepted reference pack for every comparison. A cap that is uniformly high, locally tilted or visibly present can represent different conditions, so the inspection method must distinguish the failure that matters.
Start with an approved pack datum
Define what a correctly seated cap looks like and which surfaces provide a repeatable reference. The cap top may not always be flat or dimensionally suitable as the only datum, so the bottle shoulder, neck ring, cap skirt or tamper feature may also be relevant. Use fixed lighting and viewpoint for photographs so changes are comparable.
Inspect the cap around its circumference. A closure can appear level from the front while one side remains high because of crossed thread engagement, a trapped liner or a damaged neck finish.
Match the observed condition to the process
| Observation | Possible process areas | Confirm with |
|---|---|---|
| One side of cap consistently high | Uneven thread start, pre-threading angle, bottle tilt | Cap orientation, bottle guide position and thread inspection |
| Cap uniformly high | Incomplete tightening, obstruction, liner or tamper-feature interference | Cap height, removal torque and approved integrity check |
| Cap height varies randomly | Packaging variation, unstable bottle, inconsistent presentation or wear | Batch trace, run log and component comparison |
| Cap appears seated but band is distorted | Tamper-feature contact, neck geometry or overtravel | Band position and approved finished-pack reference |
| Fault appears after changeover | Guide, belt, wheel height or recipe setting not repeated | Recorded format settings and first-off checks |
Choose inline and offline checks deliberately
A presence sensor can confirm that a cap exists, while a height sensor or vision system may detect gross height or tilt if the pack presents a stable reference. Those checks do not automatically prove thread engagement, torque, liner seating or leak performance. Combine inline detection with offline sampling where necessary.
For automatic rejection, create good and fault samples that represent the actual range, prove detection and verify that the rejected bottle is removed and accounted for. Do not set a tolerance only from a perfect single sample.
Related guidance
Continue the investigation or specification
Reduce cross-threading
Review cap start, bottle stability and thread-engagement causes.
Capping quality control
Define the approved pack and combine inline with offline checks.
Specify capping inspection and rejection
Match cap-height faults with proven detection and reject logic.
Discuss cap-height inspection
Send approved and faulty samples for an inspection review.