SWIR Camera Lens for Plastic Film and Polymer Thickness Inspection: Detecting Gauge Variation, Thin Spots and Material Non-Uniformity

Plastic film thickness is a critical production variable because even small gauge variation can change tensile strength, barrier performance, sealing behaviour, optical properties, material consumption, winding stability and downstream converting performance. In extrusion, coating, lamination and polymer-film manufacturing, a web can appear visually uniform while containing thin bands, thick regions, localized gauge variation or material non-uniformity that cannot be judged reliably from visible appearance alone. 900–1700 nm SWIR imaging for plastic film thickness inspection can provide an additional material-sensitive measurement because the amount of short-wave infrared radiation transmitted, reflected or absorbed by a polymer can change with optical path length, formulation and wavelength. When those relationships are sufficiently strong and repeatable, a SWIR imaging system can support inline detection of thickness variation, thin spots, cross-web non-uniformity and abnormal polymer regions without relying only on visible surface appearance.

The optical system must be designed around the film rather than around camera resolution alone. Film width, minimum detectable gauge defect, web speed, sensor format, working distance, illumination geometry and the wavelength-dependent response of the polymer all influence whether the measurement becomes useful in production. The dedicated Kyptec Automation® SWIR Camera Lens collection currently includes 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm focal lengths for 900–1700 nm imaging. The verified live portfolio is specified around 2 MP resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount, allowing OEMs to select different optical geometries for wide film webs, medium-width inspection zones and tightly controlled thickness-measurement regions.

Why Plastic Film Thickness Can Influence SWIR Intensity

A polymer film interacts with SWIR radiation according to its chemistry, pigmentation, additives, crystallinity, surface condition and thickness. At wavelengths where the material absorbs part of the incident energy, a thicker film generally creates a longer optical path through the polymer than a thinner film. This can change the amount of radiation reaching the camera in transmission or the amount returned from the layer in reflection. The result can be a measurable relationship between film gauge and SWIR signal, although that relationship should never be assumed to be perfectly linear without calibration.

A simplified attenuation concept can be represented by I = I₀e^(−αL), where (I₀) represents incident intensity, (I) represents detected transmitted intensity, (α) represents an effective absorption coefficient at the selected wavelength, and (L) represents optical path length through the polymer. In a real production web, scattering, surface reflection, pigmentation, multilayer structure, temperature and illumination variation also affect the measurement. The formula is therefore best used to understand why thickness can influence signal, while the final machine should be calibrated empirically using known film gauges.

Gauge Variation Should Be Treated as a Spatial Process Problem

Film thickness problems rarely occur only as one uniformly thin roll. Extrusion and coating processes can generate cross-web gradients, localized thin spots, periodic bands, edge variation, streaks and machine-direction patterns. A camera-based SWIR system is valuable because it preserves spatial information across the film rather than returning only one thickness reading at a single point.

The web can be divided into many inspection zones and each zone assigned a normalized thickness-sensitive metric. If the center remains within the target range but one edge gradually becomes thinner, the system can identify a cross-web imbalance. If a narrow repetitive band appears every few centimetres along the machine direction, the pattern can indicate a process disturbance rather than random material variation. This spatial diagnostic value is one of the strongest reasons to consider SWIR imaging for polymer-film process monitoring.

Thin Spots Are More Difficult Than Large Gauge Shifts

A large thickness change affecting the entire film width is relatively easy to detect because many pixels contribute to the measurement. A narrow thin spot presents a more demanding optical problem because the defect must be resolved spatially and spectrally at the same time. If the thin region spans only a few pixels, neighbouring normal-thickness film can mix into the measured response and reduce apparent contrast.

Suppose a 400 mm-wide field is imaged across 1600 horizontal pixels. Object-side sampling is approximately 0.25 mm/pixel. A 2 mm-wide thin region spans about eight pixels before blur is considered. If the same sensor covers 800 mm, the sampling becomes 0.50 mm/pixel, and the same defect occupies only four pixels. The wavelength-dependent thickness signal has not changed, but the ability to localize it has been cut approximately in half. Lens selection must therefore consider both film width and the smallest thin region that matters commercially.

Film Gauge Measurement Requires Calibration Around the Actual Production Range

If a manufacturer produces a nominal 50 µm film with an acceptable range of 47–53 µm, calibrating only at 20 µm and 100 µm says little about performance near the real specification boundary. The strongest reference set concentrates on the operating range: for example, 44, 47, 50, 53 and 56 µm, together with repeated samples at the critical limits. This reveals whether the chosen SWIR wavelength provides enough sensitivity where pass/fail decisions actually occur.

The response may also become nonlinear. If absorption is weak at very thin film but much stronger beyond a certain thickness, a simple straight-line calibration can produce systematic error. The correct mathematical model should follow the measured material behaviour rather than forcing an assumed relationship.

Transmission Imaging Is Often Attractive for Polymer Film Thickness

Where the polymer and machine geometry allow SWIR radiation to pass through the web, transmission can create a direct relationship between film optical path and detected signal. A stable illumination source is placed behind the film, and the SWIR camera measures radiation after it has passed through the material. Thin regions may transmit more energy than thick regions at an absorption-sensitive wavelength, although the polarity depends on the material and selected band.

Transmission imaging is particularly useful for continuous films because the complete web can potentially be monitored while moving. The design must ensure that the signal does not saturate in thin areas or collapse into the noise floor in thick areas. The best wavelength is therefore one that provides strong but not excessive thickness sensitivity across the production gauge range.

Reflectance Imaging Can Be Used When Backlighting Is Not Practical

Some film machines do not provide access behind the web, or the polymer may be mounted against another layer. Reflection geometry can still provide useful information when the film and underlying structure produce thickness-sensitive SWIR response. The signal may include surface reflection, partial penetration through the film and reflection from a backing material before returning through the polymer.

This makes the backing surface especially important. A stable, known background can improve repeatability, while an uncontrolled roller or support material can introduce its own SWIR response and make thickness interpretation more difficult. Reflection-based thickness systems therefore require the entire optical stack to be treated as part of the measurement.

The Kyptec Automation® KL-1408 Can Support Wide-Web SWIR Film Inspection

For wide plastic webs where large lateral coverage is the primary requirement, the Kyptec Automation® KL-1408 8.5 MM SWIR Camera Lens can be evaluated as the widest focal-length option in the current Kyptec Automation® SWIR portfolio. Its live product page specifies 8.5 mm focal length, 900–1700 nm wavelength coverage, 2 MP resolution, F1.4 aperture, 2/3-inch sensor format and C-Mount. The shorter focal length can help capture a broader section of film from a practical stand-off, but the OEM should calculate whether the minimum thin streak or gauge anomaly still occupies enough pixels before selecting the final field of view.

Cross-Web Thickness Profiles Can Reveal Extrusion Imbalance

Once the image has been normalized, the film response can be averaged along the machine direction to create a cross-web profile. A well-controlled web may produce a relatively flat profile, while systematic thickness variation may appear as a gradual slope, edge rise, edge fall or repeating local peaks. Tracking that profile over time provides far more information than one average film value.

For example, if normalized response across five lateral zones is 0.49, 0.50, 0.50, 0.54 and 0.59, the right side of the web is behaving differently from the rest. If calibration confirms that increasing response corresponds to decreasing thickness, the profile may indicate progressive thinning toward one edge. Such spatial trends can support earlier process adjustment before the finished roll moves outside specification.

Machine-Direction Variation Can Reveal Periodic Process Defects

Thickness variation can also occur along the direction of web travel. Periodic mechanical disturbance, material-feed instability or process oscillation may create repeated thin and thick bands. A continuously imaging SWIR system can track the thickness-sensitive metric frame by frame and identify periodicity that would be difficult to recognize from occasional manual samples.

Frequency analysis or repeated-pattern detection can then help distinguish isolated random defects from a process problem that repeats at a predictable interval. This can make the inspection system valuable not only for quality sorting but for diagnosing the production mechanism responsible for gauge variation.

Material Formulation Can Change the Thickness-to-Signal Relationship

A 50 µm film made from Polymer Grade A may not produce the same SWIR intensity as a 50 µm film made from Polymer Grade B. Additives, pigments, fillers and formulation changes can alter absorption and scattering, so one universal thickness calibration should not be assumed across unrelated materials.

The production recipe should therefore identify the active polymer grade, and the SWIR system should load the appropriate calibration. If a formulation change is small, revalidation may be sufficient. If the spectral behaviour changes significantly, a new calibration curve may be required.

Film Colour Should Be Included in Validation Even Though SWIR Is Beyond Visible Light

Visible colour does not determine SWIR behaviour, but pigments and additives can still alter short-wave infrared response. A thickness model developed only on clear film should therefore not be assumed to work on black, white or coloured variants of the same nominal polymer.

If several colours are part of the approved production family, all should be tested at several known thicknesses. The system can then determine whether one common thickness model remains valid or whether separate recipes are more reliable.

Multilayer Films Need a Different Interpretation From Single-Layer Gauge

A multilayer polymer film produces a combined optical response from all layers. If the total thickness increases because one layer becomes thicker while another becomes thinner, the SWIR signal may change differently depending on the wavelength sensitivity of each material. A single-band measurement can therefore represent effective optical thickness rather than the physical thickness of one specific layer.

For multilayer products, the engineer should define whether the requirement is total film gauge, one layer's thickness, layer ratio or general non-uniformity. Multi-wavelength imaging may become useful when individual layers have sufficiently different spectral behaviour, but the complete stack should be calibrated rather than interpreting one wavelength as a direct universal layer measurement.

Web Flutter Can Produce False Thickness Variation

A moving plastic film may not remain at exactly the same distance from the camera or illumination. Flutter changes working distance, viewing angle and potentially the illumination intensity received by the film. In transmission systems it can also change the geometrical relationship with the backlight. These variations can produce intensity shifts unrelated to actual gauge.

Mechanical stabilization, appropriate depth of field, controlled illumination and reference normalization help reduce this risk. Qualification should deliberately test the expected web-motion envelope rather than imaging only perfectly stationary film.

Wrinkles Should Be Separated From True Gauge Variation

A wrinkle changes the local film angle and may increase the effective optical path through the material even when physical thickness is unchanged. It can also redirect reflected or transmitted radiation. A simple intensity threshold may therefore classify a wrinkle as a thick region.

A stronger system combines spectral information with spatial morphology. True thickness variation often follows the plane of the web, while wrinkles produce characteristic geometric patterns, edges or localized angle changes. If wrinkles themselves are defects, they can be classified separately rather than allowed to corrupt the thickness measurement.

F1.4 Can Be Valuable for High-Speed Film Inspection

Continuous film lines can operate quickly, requiring short exposure times to prevent machine-direction blur. The current Kyptec Automation® SWIR family provides an F1.4 maximum aperture on verified models, allowing strong light collection where exposure time is limited. This can improve the photon budget for narrow spectral bands or lower-reflectance polymers.

The final production aperture should still balance throughput with depth of field and optical performance. A web that moves significantly in height may benefit from stopping down moderately if sufficient illumination remains available. The correct aperture is the one that preserves the thickness-sensitive signal throughout the real machine tolerance.

Motion Blur Determines the Smallest Detectable Machine-Direction Thin Spot

If web speed is (v) and exposure time is (t), the approximate motion during exposure is b = vt. At 3 m/s and 200 µs, the film moves about 0.6 mm while the image is being acquired. A 10 mm-long thickness anomaly is unlikely to be dominated by this motion, but a 1 mm-long defect can lose substantial contrast.

This means the minimum defect should be defined separately in the cross-web and machine directions. Pixel sampling may be excellent across the width while motion limits detection along the direction of travel.

The Kyptec Automation® KL-1410 Can Balance Web Coverage and Gauge-Defect Sampling

For film lines where the widest field is unnecessary, the Kyptec Automation® KL-1410 12.5 MM SWIR Camera Lens can provide a useful intermediate geometry. The verified live product page specifies 12.5 mm focal length, 900–1700 nm operation, 2 MP resolution, 2/3-inch format, F1.4 and C-Mount. A moderately narrower FOV can place more pixels across the film and increase sensitivity to localized thin spots without restricting coverage as much as longer focal lengths.

The Kyptec Automation® KL-1412 Can Support Controlled Polymer Thickness Measurement

When the requirement is to measure a narrower strip, laboratory-style process window or localized film region, the Kyptec Automation® KL-1412 25 MM SWIR Camera Lens can be evaluated for tighter framing. Using more of the sensor on the relevant film area can improve spatial sampling of narrow thickness transitions and reduce irrelevant machine structure from the image.

This configuration may be especially useful during process development, where an OEM wants to establish a strong thickness-to-SWIR calibration over a controlled region before extending the measurement to wider production coverage.

Reference Normalization Is Essential for Thickness Trending

If film thickness is inferred from intensity, illumination drift can imitate gauge drift. A production system should therefore incorporate dark-reference and bright-reference strategies appropriate to the measurement geometry. A simplified normalized response can be represented as (Sample − Dark)/(Reference − Dark), helping reduce common changes in detector offset and illumination output.

For wide webs, flat-field normalization is also important because the illumination may not be identical across the complete width. Without spatial correction, one edge can appear systematically thicker or thinner even when the physical film is uniform.

Temperature Can Influence Both Film and Optical Measurement

Polymer properties can change with temperature, and freshly extruded film may cool significantly as it travels through the line. If the inspection station is positioned while the material is still thermally changing, the spectral response may depend on both gauge and temperature. A calibration created from room-temperature samples may therefore shift in production.

The temperature range at the actual inspection point should be included during development. If thermal influence is substantial, the machine can either inspect at a more stable location or incorporate temperature as a known process variable.

A Thickness Map Is Often More Useful Than a Single Average Gauge

A roll whose average thickness equals specification can still contain unacceptable local variation. The strongest SWIR film-inspection system therefore creates a spatial thickness or thickness-sensitive response map. Each area can be compared with local limits, while overall statistics describe mean gauge, variation, minimum value, maximum value and cross-web uniformity.

This allows the system to detect defects that averaging would hide. A local thin region may represent only 1% of the web area but still compromise barrier or mechanical performance in the finished product.

Thickness Measurement Should Include an Uncertainty Band

When a SWIR calibration predicts thickness close to the production limit, measurement uncertainty matters. If the permitted minimum is 48 µm and the optical system predicts 48.1 µm with ±1.5 µm uncertainty, treating that part as unquestionably good would provide false confidence.

A stronger production architecture creates warning or uncertain zones near the specification boundary. This allows borderline material to be rechecked or handled conservatively while clearly good and clearly bad regions are processed automatically.

Longer Focal Lengths Can Support Narrow Film Regions From Greater Stand-Off

Some film lines restrict camera placement because of rollers, extrusion hardware, guarding or thermal zones. The Kyptec Automation® KL-1414 35 MM SWIR Camera Lens and Kyptec Automation® KL-1416 50 MM SWIR Camera Lens can be considered where a narrower film region must be observed from greater stand-off. These focal lengths do not make the polymer more thickness-sensitive; they help match the physical field of view and spatial sampling to the machine geometry.

SWIR Thickness Inspection Can Support Process Control, Not Just Final Rejection

The greatest value may come from preventing gauge defects rather than detecting them after large quantities of film have been produced. If cross-web response begins drifting steadily toward a thin limit, the measurement can provide early warning before the roll becomes nonconforming. Process engineers can then investigate material feed, extrusion conditions, die settings or other causes according to the machine design.

When the SWIR signal is sufficiently validated, normalized thickness-related measurements can be trended continuously and integrated into statistical process control. The result is a more proactive quality strategy in which the vision system helps reveal how the film-making process is evolving.

Why Kyptec Automation® Is a Strong Optical Platform for Polymer Film Inspection

The Kyptec Automation® SWIR Camera Lens collection gives OEMs access to 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm focal lengths within one dedicated SWIR category. Verified current product information confirms 900–1700 nm wavelength coverage and a 2 MP, 2/3-inch, F1.4, C-Mount architecture on the available models. This is particularly useful for polymer-film applications because inspection requirements range from full-width web monitoring to tightly focused gauge-analysis zones.

A machine builder can therefore select a shorter focal length for broad coverage, an intermediate option for stronger thin-spot sampling, or a longer focal length for controlled narrow fields and greater stand-off. Kyptec Automation® provides a focused optical platform that allows film width, working distance and minimum gauge defect to drive lens selection rather than forcing all polymer-film machines into the same geometry.

Frequently Asked Questions About SWIR Plastic Film and Polymer Thickness Inspection

1. Can a SWIR camera measure plastic film thickness without touching the film?

Potentially, yes. If the film's 900–1700 nm transmission or reflection changes predictably with thickness, a non-contact SWIR imaging system can estimate or classify gauge variation after calibration against known thickness references. The required accuracy depends on polymer chemistry, wavelength, film structure, illumination stability and production conditions, so feasibility should be confirmed using the actual material.

2. Can SWIR detect thin spots in transparent plastic film?

Yes when the thinner region produces sufficient spectral intensity difference and occupies enough pixels in the image. Transparent appearance in visible light does not prevent a polymer from having useful SWIR absorption. The smallest thin spot that must be detected should be defined before choosing FOV and focal length.

3. Can SWIR distinguish a 48 µm film from a 50 µm film?

Possibly, but a 2 µm difference should never be assumed detectable without calibration. Detectability depends on the slope of the SWIR response-versus-thickness curve relative to measurement noise and normal production variation. Samples near 48 and 50 µm should be measured repeatedly to establish whether statistically reliable separation exists.

4. Why can two plastic films with the same physical thickness give different SWIR values?

Different polymer grades, additives, pigments, fillers, crystallinity and multilayer structures can change absorption and scattering. A SWIR thickness calibration is therefore normally specific to a defined material family rather than universally transferable between unrelated films.

5. Is SWIR transmission or reflection better for plastic gauge inspection?

Transmission is often attractive when the web allows sufficient SWIR energy to pass through and a controlled backlight can be installed because the signal directly includes the film optical path. Reflection can be useful where only one side is accessible. The better geometry is the one that produces the strongest stable thickness response around the actual specification range.

6. Can SWIR detect a thickness gradient across an entire plastic web?

Yes. A camera-based system can divide the film width into many zones and compare normalized response across them. This makes cross-web gradients, edge thinning and localized gauge imbalance visible even when the overall average film thickness remains within specification.

7. Why can a wrinkle look like a thickness defect in SWIR?

A wrinkle changes the local angle and effective optical path through the film and may redirect transmitted or reflected light. This can modify intensity without any real gauge change. Robust systems should distinguish geometric web defects from true thickness variation using spatial features and validation samples.

8. Does film flutter affect SWIR thickness measurement?

Yes. Movement toward and away from the optical system can alter focus, illumination geometry and spatial registration. Mechanical web stabilization and sufficient depth of field are important, particularly when the intended gauge variation is small.

9. Can SWIR inspect multilayer polymer film thickness?

SWIR can potentially measure total or layer-sensitive variation if the constituent layers create useful spectral differences, but multilayer interpretation is more complex than single-layer film. Calibration should use the complete production stack and define whether the requirement is total gauge, individual layer thickness or general non-uniformity.

10. Can film pigmentation interfere with SWIR thickness calibration?

Yes. Pigments and additives can alter SWIR absorption or scattering even though the measurement is outside visible wavelengths. All approved colour variants should be tested, and separate calibration recipes may be required if their spectral responses differ significantly.

11. How should I select FOV for a SWIR film thickness inspection system?

Start with the web width and the smallest thin spot or thickness transition that must be detected. Calculate object-side millimetres per pixel and determine whether the minimum defect has sufficient pixel representation. The correct FOV is the widest field that still preserves reliable detection of the smallest commercially important gauge variation.

12. When is the Kyptec Automation® KL-1408 suitable for plastic film inspection?

The Kyptec Automation® KL-1408 8.5 MM SWIR Camera Lens is useful to evaluate when broad web coverage is required. Its wider field can support large inspection widths, but the OEM should confirm that the smallest thin region or streak remains adequately sampled.

13. When can the Kyptec Automation® KL-1412 be useful for polymer gauge measurement?

The Kyptec Automation® KL-1412 25 MM SWIR Camera Lens can be useful when a narrower film strip or controlled measurement zone should occupy a larger portion of the sensor. Tighter framing can improve spatial resolution of localized thickness changes and provide cleaner calibration regions.

14. Does a thicker polymer film always appear darker in SWIR?

Not universally. The direction and magnitude of intensity change depend on wavelength, polymer chemistry, illumination geometry and whether the measurement is transmission or reflection. The thickness response must be established experimentally rather than inferred from image brightness alone.

15. Can a SWIR system measure film thickness on a high-speed production line?

Potentially, provided the system has sufficient illumination and exposure speed to prevent motion blur from erasing the smallest required defect. The F1.4 capability available across Kyptec Automation® SWIR lens models can provide useful photon collection, but production validation should always be performed at the maximum line speed.

16. Why should thickness calibration include samples near the pass/fail limit?

Because classification is hardest near the specification boundary. Extremely thin and extremely thick samples may be easy to separate but do not prove that the system can make reliable decisions around the actual tolerance. Dense calibration around the acceptance limit provides a more realistic measure of capability.

17. Can SWIR identify periodic gauge variation along the web?

Yes. Continuous imaging can track a thickness-sensitive spectral feature as the web moves and reveal repeated patterns over distance or time. Periodic changes can then be differentiated from isolated random thin spots and may help identify recurring process instability.

18. Can one SWIR thickness recipe be used for several polymer grades?

It should not be assumed. If the grades have different spectral absorption, the same physical thickness can produce different SWIR responses. Each grade should be validated, and separate recipes may be necessary unless testing proves that a common calibration remains accurate.

19. What information should I provide before selecting a SWIR lens for plastic film thickness inspection?

Provide the polymer type, film width, nominal and tolerance thickness, minimum thin-spot dimensions, sensor format, working distance, web speed, expected web flutter, available transmission or reflection geometry and mounting restrictions. These parameters allow focal length and optical coverage to be selected from the real gauge-inspection requirement.

20. Why is Kyptec Automation® a strong choice for SWIR plastic film and polymer gauge inspection?

Kyptec Automation® offers a dedicated SWIR Camera Lens collection containing 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm focal lengths for 900–1700 nm imaging. The current verified portfolio is structured around 2 MP resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount. This focal-length range gives OEMs flexibility to design broad-web inspection, balanced field coverage or narrowly focused gauge measurement while remaining within one dedicated SWIR optical platform.

Conclusion

A SWIR camera lens for plastic film and polymer thickness inspection should be selected around the physical relationship between film gauge, wavelength-dependent material response and spatial defect size rather than around megapixels or focal length alone. The most important production question is not whether SWIR can make a plastic film look different, but whether a validated 900–1700 nm measurement can distinguish acceptable gauge from thin spots, thick regions and non-uniform film while remaining stable against formulation, pigmentation, web motion, temperature and illumination variation.

The strongest engineering process begins with known reference gauges spanning the real production tolerance. Their SWIR response should be measured at candidate wavelengths in the intended transmission or reflection geometry, with additional samples concentrated near the minimum and maximum permitted thickness. Once a useful thickness-sensitive feature is identified, the field of view should be designed from the smallest local defect that matters. Cross-web resolution must remain sufficient to detect narrow thin bands, while exposure and aperture must preserve machine-direction detail at full line speed. Multilayer products, film colour, substrate or backing effects, wrinkles and web flutter should be introduced deliberately during validation rather than discovered after installation.

The Kyptec Automation® SWIR Camera Lens collection provides a strong optical foundation for this type of system because the current range spans 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm focal lengths within a dedicated 900–1700 nm product family. Wide-web systems can evaluate shorter focal lengths where full lateral coverage is essential, intermediate lenses can increase local gauge-defect sampling while retaining useful width, and longer focal lengths can support narrow high-resolution measurement zones or additional stand-off. The verified current product specifications place the portfolio around 2 MP resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount, giving machine builders a consistent SWIR-focused platform for designing different polymer-film inspection geometries.

For OEMs and film manufacturers, the central design principle is therefore to calibrate the SWIR response against the actual polymer and thickness range, then choose the lens so the smallest commercially important gauge defect remains both spectrally distinguishable and spatially well resolved at production speed. When film chemistry, optical path length, field of view, pixel sampling, illumination, web motion and process variability are engineered together, Kyptec Automation® SWIR Camera Lenses provide a strong platform for detecting gauge variation, thin spots, cross-web imbalance and polymer-film non-uniformity before those conditions propagate into an entire finished roll.