The use of mid-wavelength (also referred to as midwave) infrared (MWIR) light is crucial in many areas, such as gas absorption; thermal monitoring of homes and equipment; military enhanced-vision systems for imaging terrain, people and vehicles; and environmental gas monitoring.
Infrared in the MWIR range is even useful in the diagnosis of pregnancy in dairy cows, among other applications.
Like other types of infrared (IR) optical system, optical filters are essential in ensuring the successful operation of MWIR hardware. These filters facilitate the transmission of desired wavelengths while blocking others as required by the application in question.
From time to time, designers and engineers may deliver filter specifications which are more stringent than necessary, failing to consider what needs may already be fulfilled by other parts of a design.
As a consequence of these more stringent specifications, designs can become more difficult to implement, meaning that they then take longer to bring to market, and involve more expense than is perhaps necessary.
By seeking quality advice from the beginning and then implementing a series of straightforward amendments to the design process, companies can achieve their goals while simultaneously saving time and money.
An MWIR filter is an optical device which operates in the MWIR range – which extends from 3 to 6 microns. Filters are made up of substrates with multilayer thin film structures. These substrates can transmit required IR wavelengths whilst blocking others.
MWIR filters are normally edge pass or bandpass filters, and their wavelength selectivity enhances the signal-to-noise ratio for the detectors in an MWIR detection or imaging system.
Filters allow applications to receive the wavelengths they need to function correctly. Several critical characteristics exist which can affect the filter’s ability to meet the application’s needs.
These characteristics include transmitted and blocked wavelength ranges, surface quality, total amounts of IR energy passed by the filter, as well as the ability to preserve these characteristics within the application’s operating environment.
Manufacturers must properly match filter construction to the specifications of the application design, in order to tightly control filters’ characteristics.
Surface quality and the existence (or absence) of surface defects can affect image quality. In some cases, surface defects are less of a concern, for example when the application is designed to only measure the amount of MWIR light present in a particular setting.
Sometimes this impact is more important, however. For example, an imaging application that converts MWIR into an image which is designed to be viewed by the human eye. Defects may affect the quality of the image which, in this case is vital.
Designers are prone to applying surface-quality specifications that are more rigorous than required. As MWIR wavelengths are much longer than wavelengths in the visible light spectrum, defects which may be visible to the naked eye in ambient light would not be visible under MWIR light. This is because the wavelengths are too long to highlight the defects.
Even in applications which do require imaging capabilities, surface quality is frequently ‘over specified’. When filter surface quality is over specified, yields can decline substantially, increasing the final filter cost.
The sets of wavelengths either blocked or transmitted by the filter form the core of its intended functionality. Depending on the design of the application, bandpass filters may be needed to transmit a band of adjoining wavelengths within the MWIR range.
Additionally, some applications may require longwave (transmitting light over a certain wavelength) or shortwave (transmitting light below a certain wavelength) filtering.
As the wavelength range becomes narrower, the amount of light transmitted reduces. Manufacturers must regularly balance the filter’s particular wavelength transmission bands with the total amount of IR energy travelling through the filter to achieve a design’s requirements.
For example, a sensor must receive a specific amount of energy to react. If the filter is unable to transmit enough IR energy to meet the sensor’s specifications, then the design simply will not function.
Limiting the size of the transmission band, whilst supplying more wavelength selectivity, also decreases the total potential light that is able to pass through it. So, this can actually compromise detector’s functionality if not addressed.
It is imperative that filters are able to work within the desired application environment. For example, a coating that is stable at humidity levels or temperatures within a manufacturing facility may react very differently within the application’s final operating environment.
The production method used to coat filters has considerable impact on the ability to control a filter. Most MWIR filters are coated using an evaporation method. Within this process, the coating compound is heated until it is in a vapor state before this vapor is allowed to condense on the substrate.
Iridian utilizes another approach: energetic sputtering. While this technology is routinely employed in other wavelength ranges (such as within the visible or near-infrared (NIR) wavelengths), it is less commonly used to address needs in the MWIR range, because in these cases, it is important to use coating materials which transmit in the MWIR.
As a method, sputtering has some distinct advantages over evaporation. Filters coated using this method are environmentally robust, allowing manufacturers to apply a great deal of control over the sensor’s final spectral characteristics.
Common Filter Issues
Design choices can have unexpected consequences, even beyond any compromise in performance characteristics. While MWIR filters are popular in numerous industries, even the most experienced designers can create superfluous manufacturing challenges via their filter specifications and designs.
At an advanced level, designers may approach challenges like scientists, rather than engineers. They contemplate what would be the most ideal solution, without considering that real-world working conditions may be accepting of less-than-perfect choices. This may result in specifications which are too tight, leading to excessive costs.
For example, designers may ask for a tight defect specification on the surface of the filter. Where applications involve sensing and not complete imaging, then the quality specification relating to the surface may be unjustified, provided enough IR light can still reach the detector, with no compromise in the signal-to-noise ratio.
The correct design approach very much depends on the application being designed. An imaging system designed for military use, for example, may require a high image integrity as an operator will need to be able to discern and react to objects on a display. In other applications like MWIR gas sensing, this may not be a factor that needs to be considered.
Communication issues at the design stage can cause problems, especially if there is an assumption that cutting one large filter into several smaller filters is more cost-effective than ordering appropriately sized filters at the outset.
This may result in unnecessary requirements for standardized performance on a large part, carrying with it considerable cost implications. While properly sizing filters for the device design throughout the manufacturing process can result in additional processing costs, this may still be less than the costs attached to producing larger parts which are fully compliant. Even the cost of custom processing is factored in, this may result in much lower financial implications.
Other elements of the system’s hardware can effect filter design and its related requirements. An IR detector will only operate over a certain range, with light that falls outside that range not registering on the detector. These kind of limitations can become a positive feature of the filter’s design, however.
For example, should the detector not respond to wavelengths that are under 2 microns, it is not financially sensible to engineer the filter to block anything between 1 and 2 microns.
Thinking about the filter early on in the design process allows for redesigned optical path layouts which will then impact upon the filter’s cost, complexity and achievable performance.
Adjusting the placement and size of the filter within the optical path, or even reengineering the placement of detectors (and therefore the beam path), can alter the whole industrial design and also the device’s operation. As such, changes in detection techniques can result in a more efficient final product.
A complete understanding of an application’s genuine needs before designing filter solutions that balance both commercial and technical requirements can pre-empt and help avoid issues that can occur in filter manufacturing.
The manufacturing process itself can have a beneficial or adverse effect on the difficulty of meeting design specifications, as well as having a major impact on the turnaround time of custom designs or specifications. Iridian has a typical turnaround time from ordering to product delivery of 6 to 8 weeks.
Where filter manufacturers do not work in partnership with system designers or otherwise fail to explore the intricacies of the manufacturing process, there may be unforeseen delays in fulfilling a successful filter build. This can delay a whole project, in some cases by many months, thus increasing time to market.
Filter Design Solutions
Minor amendments or additions to the design process may help in the elimination of these issues.
Designers must consider filter requirements as early in the design process as it is practical to do so. Hardware used should not be comprised of a mix of black boxes that function independently – rather, a systems-engineering approach should be employed to explore how the interactions of these different elements can affect the whole project.
There could also be advantages in this approach, such as utilizing a detector’s specifications to limit the range of wavelengths that the filter should block.
It is also prudent to consider entirely different application designs. A variation in layout, an alternative detector, or the implementation of some other change may improve overall performance while lowering costs.
Above all, it is important to consider a filter vendor as a partner and collaborator. Consultations should commence at the very beginning of the design process. Working with experts in filter design and manufacture can prompt crucial questions based on their experience and expertise of the capabilities of different types of filters. It is always better to tackle problems before they arise, ensuring both time and money savings. While any design project will present its own challenges, elimination of avoidable issues can free up resources to focus on those that remain.
This information has been sourced, reviewed and adapted from materials provided by Iridian Spectral Technologies.
For more information on this source, please visit Iridian Spectral Technologies.