ISO 9001 Certified Manufacturer 360° Factory Tour
Applications

Bandpass Filter Selection for LiDAR

· 6 views · 3 min read · OOTEE Optics Editorial

Original content by the OOTEE Optics editorial team, drawn from first-hand production line and inspection data. Please credit the source when republishing.

Given limits in expertise and test conditions, the theories, parameters and conclusions here carry no guarantee of absolute accuracy. For technical learning and exchange only - never the sole basis for design, acceptance or selection. Corrections welcome.

Bandpass Filter Selection for LiDAR

What the receiver filter does

A LiDAR bandpass filter admits the laser return while attenuating out-of-band sunlight and stray light. In a pulsed time-of-flight receiver, its practical value is preserving weak returns without exhausting detector dynamic range. The narrowest available passband is not automatically the best choice.

Common architectures use 905 nm with silicon detectors or 1550 nm with indium gallium arsenide detectors. Their spectral responses differ, so the blocking specification must change with the receiver. All numerical requirements below are illustrative engineering targets, not product guarantees.

Spectral filtering in the receive path
Fig. 1 Spectral filtering in the receive path

Parameters that matter to the return

Thin-film interference creates a transmission window. Fig. 1 shows its location in the receive path; light bypassing the filter still requires mechanical baffling. Center wavelength (CWL) locates the window, while full width at half maximum (FWHM) describes its width at half the peak transmission. FWHM is not a guaranteed high-transmission interval.

Optical density (OD) specifies attenuation using dimensionless transmittance $T$. OD6 means a transmitted fraction of $10^{-6}$, or 0.0001%, within a stated blocking range.

$$\mathrm{OD}(\lambda)=-\log_{10}T(\lambda)$$

Increasing angle of incidence (AOI) generally shifts an interference passband toward shorter wavelengths. An effective-index approximation supports an initial budget:

$$\lambda(\theta)\approx\lambda(0)\sqrt{1-\frac{\sin^2\theta}{n_{\mathrm{eff}}^2}}$$

With $n_{\mathrm{eff}}=2.0$, a 905 nm normal-incidence center shifts to about 901.6 nm at 10°. Actual coating behavior, polarization splitting and converging beams require measurement. Trace both chief and marginal rays at the filter: receiver field of view alone does not specify AOI.

Turn the wavelength budget into acceptance criteria

Measure the laser spectrum across temperature, drive current and pulse conditions. Then include filter tolerances and angular shifts. Fig. 2 isolates the bandwidth trade-off: both ideal curves have the same center and peak, but a narrow window loses transmission faster when the return wavelength moves.

Ideal normal-incidence Gaussian model; 90% peaks
Fig. 2 Ideal normal-incidence Gaussian model; 90% peaks

For a hypothetical laser envelope of 902–908 nm, start by requiring at least 85% transmission throughout that interval under the agreed operating conditions. Only then assess a nominal CWL of 905 nm and FWHM of 20 nm. Several nanometers of angular shift may justify a pre-offset center or a different filter location.

Budget itemIllustrative requirementPurpose
Useful return≥85% across 902–908 nm at operating extremesProtect return throughput
Angle and temperatureAOI 0–10°; −20 to 70 °CCheck combined extremes
BlockingOD ≥5 over 400–880 nm and 930–1100 nmStarting point for a silicon receiver
Assembly and testClear aperture, thickness tolerance, test spot and polarizationMake acceptance reproducible

These targets require a feasibility review. Specify transition regions between blocking and transmission; an OD5 requirement cannot extend directly to a high-transmission boundary. A 1550 nm receiver needs blocking limits based on its own detector response.

Ask for transmission at the actual laser wavelengths rather than relying on nominal bandwidth. In the ideal 20 nm Gaussian example, transmission at 902 nm and 908 nm is about 84.6%, already below the illustrative 85% requirement even before angular or manufacturing effects. A flatter passband, additional width or revised optical geometry may therefore be needed. This is why a nominal CWL and FWHM pair cannot replace an interval-based throughput requirement.

For sample evaluation, record the measurement spot, spectral resolution and incident polarization alongside each curve. Use the same conditions when comparing samples, and distinguish guaranteed limits from typical data. Include assembly tolerances when deciding which angular and temperature combinations represent the operating extremes.

Background rejection depends on integrated leakage, not one OD value. For background spectral irradiance $E_{\mathrm{bg}}$ at the filter, effective collection area $A$ and detector responsivity $\mathcal{R}$, the approximate background photocurrent is:

$$I_{\mathrm{bg}}=A\int E_{\mathrm{bg}}(\lambda)T(\lambda)\mathcal{R}(\lambda)\,\mathrm{d}\lambda$$

Use consistent spectral units, such as irradiance per nm with integration in nm. A square-root bandwidth benefit applies when background shot noise dominates, the background spectrum is locally flat and signal throughput stays constant. It does not directly describe dark-noise-limited or saturated operation.

Different receivers need different compromises

A long-range vehicle receiver must retain weak returns under strong daylight, including at field edges. Outdoor robots also need checks for window tilt, scattering and light bypass. Wide-field detector arrays require transmission assessment across the image plane because pixels can see different incidence angles.

A narrow-field industrial rangefinder with controlled angles and limited laser drift may support a narrower passband. A wide-field design may benefit more from angular tolerance or optical relocation. Coherent receivers introduce local-oscillator and polarization constraints beyond this pulsed-receiver example.

Common specification traps

  • Requesting OD6 without a wavelength range, measurement floor or pinhole check.
  • Accepting only a room-temperature, normal-incidence curve instead of the agreed angle, temperature, polarization and aperture coverage.
  • Expecting rejection of another LiDAR operating inside the passband; timing, coding or spatial discrimination may still be necessary.
  • Choosing bandwidth before checking return throughput at operating extremes.

Technical reference: LiDAR source and detector pairing. Acceptance should link spectral measurements to return signal, background current, saturation margin and false-trigger results in the actual receive path.

Related Reading

View all

Related Products

View all products

Need a custom optical filter?

Send us your drawing and spectral targets — an engineer replies with a feasibility check and quotation within one business day.

Source Factory In-house coating lines, no middlemen
Ready to Ship Standard items dispatched same or next day
Quality Guaranteed 100 % inspected, spectral report included
Express Delivery DHL / FedEx worldwide