Why is my longpass leaking blue/green—spec or setup issue?

Most 'blue/green leak' from a longpass filter is a setup or geometry issue, but it can also be a specification limit or a physical defect. A series of quick diagnostic checks can help you determine the root cause and how to fix it fast. Seeing unwanted blue or green light leaking through your longpass filter can be frustrating. The good news is that the cause is usually a simple setup issue rather than a faulty filter. Most often, the problem stems from the filter's angle, the beam's geometry, or stray light paths. However, the leak can also be due to a specification limit or a physical defect. This guide provides a fast diagnostic checklist to help you identify the root cause and implement a solution.

1. Reduce the Angle of Incidence (AOI):
Place the filter perpendicular to the light beam or move it into a collimated section of your optical path. If the leak significantly decreases, the problem is related to your setup's geometry and the angle of incidence.

2. Stack a Second Longpass Filter:
Add a second, similar longpass filter in the beam path. If the leak is reduced by a factor of 10 to 10,000, your original filter's optical density (OD) is likely too low for your application. If there is little change, the cause is probably a stray light path.

3. Block Stray Light Paths:
Use a black aperture that is slightly smaller than the filter's clear aperture, tape the edges, or add baffles around the mount. If this reduces the leak, stray light was bypassing the filter entirely.

4. Flip the Filter:
Some filters have coatings optimized for a specific direction of light travel. If flipping the filter changes the amount of leakage, you have found a specification quirk related to its orientation. Ensure it is mounted in the manufacturer's recommended direction.

5. Use a Polarizer:
Place a polarizer in the beam and rotate it. If the amount of leaked light varies with the rotation, the issue is related to AOI and polarization dependence in your setup.

6. Test with a Monochromatic Source:
Use a blue or green LED as your light source. If your filter's datasheet specifies OD 4 blocking at that wavelength but your camera still registers a signal, consider your system's dynamic range. An OD 4 filter still transmits 0.01% of light, which can be visible with a bright source or long exposure time.

If the quick checks point to a specific cause, the following details explain why it happens and how to fix it permanently.

This is the most common category of problems.

• Angle of Incidence (AOI) Blue-Shift: Interference longpass filters are designed for a specific angle of incidence, typically 0 degrees. At oblique angles, the cut-on edge of the filter shifts to shorter wavelengths. This blue-shift can be estimated with the formula:
  λ(θ) ≈ λ(0)√(1 - (sinθ/n_eff)²) An angle of 30-45° can shift the edge by 5-12%, potentially moving it into the green or blue spectrum and allowing those colors to pass. The fix is to use the filter at near-normal incidence, place it in a collimated section of the beam, or specify a filter with a longer cut-on wavelength to provide more margin.
• Fast Beams or Wide Cone Angles: In imaging systems with fast optics (e.g., f/1.4-f/2), the light rays converge or diverge at a wide range of angles. Each ray hits the filter at a different angle, 'smearing' the sharpness of the cut-on edge and allowing some shorter wavelengths to leak through. To fix this, move the filter to a collimated space in your optical path, use a slower optical setup (stop down), or choose a filter specifically designed for use in fast cones.
• Polarization: At angled incidence, the cut-on edge for s-polarized and p-polarized light can separate. One polarization state may shift more than the other, causing it to leak at shorter wavelengths. You can resolve this by reducing the AOI, using a depolarizer before the filter, or selecting a polarization-insensitive filter design.
• Stray Paths and Edge Light: Light can find paths around your filter through gaps in the mount, or it can travel through the edge of the substrate itself (light piping). Ghost reflections between optics can also re-enter the beam path after bypassing the filter. To fix this, add black masks or baffles, use a field stop that is slightly undersized, tilt the filter by a few degrees to direct ghost reflections out of the system, and use filters with blackened edges.

This is often the second most likely culprit.

• Finite Blocking (Optical Density): No filter provides perfect blocking. A filter with an OD 3 to OD 4 specification will still transmit 0.1% to 0.01% of out-of-band light. For a very bright light source or a sensitive detector with a long exposure time, this small amount of transmission can be significant. The solution is to order a filter with higher blocking (OD 5-OD 6) in the stopband or to stack an absorptive filter (like a colored glass shortpass or notch filter) to increase the total optical density.
• Secondary Passbands: The periodic nature of thin-film coating stacks can create ripples or secondary transmission bands at wavelengths far from the primary cut-on edge. Unless the filter is specifically designed with 'extended blocking,' it may have a passband in the blue or green region. To fix this, choose a filter that explicitly specifies OD across the entire blue/green spectrum or pair your filter with an absorptive colored glass filter that blocks these passbands.

While less common, these physical issues can be the cause.

• Pinholes, Scratches, and Voids: Physical imperfections in the coating can act like tiny holes, allowing light to pass through the blocking region. These often appear as bright dots or can contribute to a general haze. The solution is to inspect the filter with a bright light and contact the manufacturer for a replacement if significant defects are present.
• Surface Films and Contamination: Oil from fingerprints or other surface contaminants can create micro-etalon effects (thin-film interference) or scatter light, reducing performance. The fix is to properly clean the optic using approved methods and solvents for precision optics.