Can You Stack Longpass and Shortpass Filters to Build a Custom Passband?

Optical engineers designing new infrared sensors frequently consider whether stacking existing components isolates a custom passband effectively. In laboratory settings, placing two distinct optical edges in the light path serves as a rapid method to verify detector response and isolate a target wavelength. However, transitioning this physical stacking strategy into OEM production introduces severe optical and mechanical flaws. 

BoDian Optical manufactures precision thin-film components for analytical instruments and industrial gas sensors. BoDian’s engineering teams routinely replace makeshift stacked assemblies with unified single-substrate solutions to recover lost transmission, eliminate spectral leakage, and ensure environmental stability. This technical guide examines the mechanics of stacked bandwidth generation, the resulting photometric penalties, and the engineering logic behind specifying a single unified filter for precision instrumentation.

Target Passband Generation Mechanics

Creating a custom passband through physical stacking relies on the intersection of two opposing transmission curves. When light passes through sequential filters, the final output spectrum is almost the mathematical product of their individual transmission profiles.

Overlapping Transmission Zone Definition

Placing two distinct optical filters in series restricts the light reaching the detector. Only the spectral region where both components exhibit high transmittance will survive the journey through the optical system. This overlap effectively extracts a custom passband out of the broader infrared spectrum.

Longpass Edge Cut-On Parameters

To define the starting boundary of the desired bandwidth, an Infrared Long Wave Pass Filter rejects the shorter wavelengths. Utilizing thin-film interference principles, a precision edge filter employs symmetrical periodic layer structures to achieve steep spectral transitions. The physical transition zone from optical density 3 (OD3) blocking to 50% peak transmission can be controlled to within 1% to 2% of the center wavelength. This steep edge acts as the precise lower limit for the target bandwidth.

Shortpass Edge Cut-Off Control

A shortpass component is placed sequentially to block the longer infrared tail. By truncating the transmission window opened by the longpass component, the shortpass filter defines the upper boundary. The spectral region trapped between these two engineered edges becomes the functional measurement bandwidth.

Stacked System Signal Attenuation Factors

While overlapping curves logically create a specific custom passband, the physical reality of propagating light through multiple independent substrates causes an immediate drop in photometric efficiency.

Fresnel Reflection at Air-Substrate Interfaces

Infrared optics typically utilize high-index substrates like Silicon (Si, index n ≈ 3.4) or Sapphire. Bare surfaces suffer from significant Fresnel reflection. A single uncoated Silicon surface reflects approximately 40% of incident infrared energy. Stacking two independent filters introduces four discrete air-glass interfaces into the system. Without specific optical matching, the compounded transmission loss severely degrades the signal-to-noise ratio, starving the detector of essential photons.

Internal Multi-Reflection Ghosting Effects

When two reflective substrates are mounted parallel to each other, they inadvertently form a Fabry-Perot cavity. Light bounces back and forth between the inner parallel surfaces before finally exiting the stack. This internal multi-reflection generates optical ghosting. Secondary signals arrive at the detector out of phase, creating false peaks and baseline drift in sensitive analytical instruments.

Anti-Reflection Coating Recovery Capabilities

If spatial constraints force the use of multiple discrete components, bare interfaces are photometrically unacceptable. Applying broadband Infrared Anti-Reflection Filters or equivalent AR coatings to the intermediate surfaces becomes mandatory. These coatings utilize alternating low and high-index thin-film interference layers to induce destructive interference for reflected waves, recovering the transmitted energy and mitigating Fresnel reflection.

Hidden Risks in Physical Filter Stacking

Beyond the immediate loss of photons, the mechanical separation between components introduces environmental and spatial vulnerabilities that threaten the stability of the custom passband.

Optical Path Length Distortion

Adding a second substrate inherently doubles the physical glass thickness. In convergent or divergent optical beams, this extended optical path length induces focal shift and spherical aberration. Compact NDIR gas sensors and miniaturized thermopile detectors lack the spatial volume to accommodate thick, stacked assemblies without compromising the focal spot size on the active detector area.

Angle of Incidence Blue-Shift Dynamics

Thin-film filters are highly sensitive to the angle of incidence. As the incident beam angle increases, the effective optical thickness of the internal film layers decreases, causing the transmission spectrum to blue-shift toward shorter wavelengths. If stacked components are not mounted perfectly parallel, or if a converging beam strikes each filter at slightly different angles, the longpass and shortpass edges shift independently. This asymmetrical shift warps the shape of the custom passband.

Mechanical Instability in Harsh Environments

Vibration and thermal cycling threaten physically stacked optics. For instance, optical sorting sensors mounted directly on mining excavators operating a heavy-duty Earth bucket face massive continuous mechanical shock. Under such extreme stress, stacked components shift, altering the internal air gap and tilt angles, which ultimately destroys system calibration. Executing systematic optimization of NDIR gas sensor optical paths prevents early prototype failures caused by mechanical shock and misalignment.

Single-Substrate Narrow Bandpass Solutions

To achieve high-fidelity spectral isolation without the penalties of physical stacking, optical engineers consolidate the required thin-film layers onto a single substrate.

Single-Substrate Interface Reflection Elimination

By depositing the longpass stacks, shortpass stacks, and spacer layers onto one continuous piece of Silicon or Sapphire, the intermediate air gap is physically eliminated. This single-substrate approach immediately recovers the 20% to 30% energy loss caused by internal Fresnel reflection in stacked setups.

Integrated Multi-Cavity Structure Benefits

Modern vacuum deposition techniques allow the design of multi-cavity Fabry-Perot structures. Instead of relying on two separate edge filters, a unified design utilizes coupled cavities to shape the custom passband. This results in a flatter transmission peak and deeper, continuous out-of-band blocking exceeding OD4, avoiding the risk of mismatched edge curves.

Signal-to-Noise Ratio Optimization

For OEM manufacturing, specifying an Infrared Narrow Bandpass Filter provides optimal optical performance. A unified bandpass maximizes the signal-to-noise ratio by ensuring high in-band transmission and absolute mechanical stability, outperforming stacked prototypes in every photometric metric.

Custom Optical Passband Specification Guidelines

Moving from a stacked laboratory prototype to a unified thin-film component requires precise optical specification to ensure the final product meets detector requirements.

Center Wavelength and Bandwidth Definition

Rather than quoting cut-on and cut-off points from separate filters, engineering teams should define the target center wavelength (CWL) and the full width at half maximum (FWHM). Specifying the exact bandwidth and the required optical density for out-of-band blocking ensures the coating design fully isolates the target signal without spectral leakage.

Substrate and Coating Detector Matching

The choice of substrate dictates the long-wave transmission limits and environmental durability. Silicon is excellent for mid-infrared transmission up to 7 microns, while Sapphire is preferred for shorter near-infrared wavelengths due to its mechanical hardness and resistance to thermal shock. The substrate thickness must match the detector package’s specific optical path length constraints.

OEM Manufacturing Collaboration

Transitioning a laboratory concept into a scalable instrument requires precision coating capabilities. Our facility deposits complex multi-cavity designs directly onto your chosen substrate, ensuring your custom passband remains stable across all environmental conditions.

Transitioning from a stacked prototype to a high-transmission single-substrate filter requires careful review of your center wavelength, bandwidth, and blocking requirements. If your engineering team is experiencing signal loss, ghosting, or poor signal-to-noise ratios in a current optical assembly, discussing your system geometry with thin-film specialists clarifies the upgrade path. To review your specific spectrum requirements or request technical drawing evaluations, please contact our optical design team.

FAQ

Can I use stacked filters for initial laboratory prototyping?

Yes, physically stacking separate longpass and shortpass components is a practical method for early-stage proof of concept. It allows engineers to quickly test different custom passband configurations without waiting for a dedicated coating run. However, due to severe interface Fresnel reflection and the resulting drop in signal-to-noise ratio, this method is strictly for validation and should not be implemented in final OEM instrument production.

How does the angle of incidence affect a custom passband?

Thin-film interference filters shift their transmission spectrum toward shorter wavelengths as the angle of incidence increases. In a stacked configuration, if the two separate substrates are not perfectly parallel, or if a converging beam strikes them differently, the longpass and shortpass edges blue-shift by varying amounts. This asymmetrical shift warps the target bandwidth and introduces measurement errors.

What is the primary advantage of a single-substrate narrow bandpass filter?

A unified narrow bandpass filter eliminates the internal air-glass interfaces inherently present in stacked systems. This single-substrate approach drastically reduces internal reflection and optical ghosting, maximizing the transmitted infrared energy and maintaining strict mechanical stability within a very short optical path length.