Sensors based on Bragg gratings (FBG, Fiber Bragg Grating) are valued for their robustness, the ability to multiplex many measurement points in a single fiber, and the encoding of information in wavelength rather than in signal power. We cover the basics of how they work in the article FBG sensors: how they work, and here we focus on how to design them.
The most important design question is: what spectrum should the grating have so that its shift can be measured reliably? The answer comes from a model in which you can change the grating parameters and observe the effect in the spectrum. OptiGrating from Optiwave is used for this.
How a Bragg grating works
A Bragg grating is a periodic modulation of the refractive index of a fiber core or waveguide. It reflects light whose wavelength satisfies the Bragg condition:
λ_B = 2 · n_eff · Λ
where n_eff is the effective refractive index of the mode and Λ is the grating period. Light at other wavelengths passes through practically unchanged. This makes the grating act as a narrowband reflector or filter.
Where the sensitivity to temperature and strain comes from
The Bragg condition explains why a grating is a sensor. If n_eff or Λ changes, the reflection wavelength changes.
- Temperature changes the refractive index of the glass and causes thermal expansion of the fiber, that is, it changes both
n_effandΛ. - Strain stretches the fiber, changing the grating period
Λand, indirectly, the refractive index as well.
The shift of the Bragg wavelength is therefore the measurement signal. In typical silica fibers around 1550 nm, the sensitivity is on the order of 10 pm per degree Celsius and about 1 pm per microstrain, but the exact values depend on the fiber, the grating, and how it is packaged. That is why the sensitivity is worth determining for a specific construction rather than taking it from a table.
OptiGrating: how it models gratings
According to Optiwave, OptiGrating uses coupled-mode theory to describe light in the grating and enables both analysis and synthesis of gratings. For analysis, the grating is approximated by a sequence of uniform segments, which are combined with the well-known transfer matrix method.
This way of working makes it possible to model gratings with a varying profile, for example with apodization, because each fragment of the grating can be described by separate parameters. The result is a reflection and transmission spectrum that can be assessed against the sensor requirements.
Apodization: a cleaner spectrum, a better measurement
The spectrum of a grating with constant modulation has pronounced sidelobes, which make it harder to determine the position of the main peak and can interfere with neighboring channels in multiplexed systems. Apodization, that is, a gradual change of the modulation amplitude along the grating, suppresses these sidelobes.
In its OptiGrating materials, Optiwave shows sidelobe suppression using apodization. For a sensor this means a cleaner reflection peak and a more accurate determination of the peak position, and therefore a more accurate measurement of temperature or strain.
Long-period gratings
Besides classical Bragg gratings, in which light is reflected into the counter-propagating mode, there are long-period gratings (LPG), in which coupling to cladding modes occurs. They have a different spectral characteristic and often a greater sensitivity to the refractive index of the surroundings. Optiwave points to support for long-period gratings with coupling to cladding modes, and to their use as filters for flattening the EDFA characteristic.
From grating to sensor: a parameter as a signal
Designing a sensor comes down to the question of how the grating spectrum changes with the measured quantity. In the model this can be checked by sweeping parameters: changing the period and the refractive index to correspond to a change in temperature or strain, and observing the shift of the spectrum.
With this approach, it is worth paying attention to several things:
- the width of the reflection peak and its effect on the accuracy of determining the position,
- the sidelobe level,
- the reflectivity, that is, the signal strength,
- the range within which the spectrum shift stays within the band of the readout system,
- distinguishing the effects of temperature and strain, which shift the spectrum in the same direction and need to be separated, for example by using a second, reference grating.
Integration with OptiSystem and OptiInstrument
Optiwave states that OptiGrating is integrated with other tools from the company, such as OptiSystem and OptiInstrument. In practice this enables three stages of work.
Grating design. In OptiGrating you design the grating and check its spectrum.
System model. In OptiSystem the grating can be used as an element of the readout system: source, grating, detector, or spectrum analyzer. For more on system-level analysis, see the article Simulation of broadband fiber-optic networks.
Measurement. In OptiInstrument you can automate the spectrum measurement of the manufactured grating and compare it with the model. We describe this in the article Automating photonic measurements: OptiInstrument.
Example workflow for designing an FBG sensor
1. Define the measured quantity and range
Determine whether the sensor measures temperature, strain, or both, and over what range.
2. Define the spectrum requirements
Establish the required center wavelength, peak width, reflectivity, and acceptable sidelobe level.
3. Design the grating in OptiGrating
Select the grating period, length, modulation amplitude, and apodization profile.
4. Check the reflection and transmission spectrum
Verify that the spectrum meets the requirements and, if necessary, modify the parameters.
5. Sweep the parameters
Study how the spectrum shifts when the period and the refractive index are changed to correspond to a change in temperature and strain.
6. Build a model of the readout system
Use OptiSystem to model the source, the grating, and the detector or spectrum analyzer.
7. Study multiplexing
If several gratings are to work in one fiber, check that their spectra do not overlap across the whole measurement range.
8. Measure the manufactured grating
Automate the spectrum measurement in OptiInstrument and compare it with the model.
Which tool for which task
| Task | Tool |
|---|---|
| Design of a Bragg grating and its reflection spectrum | OptiGrating |
| Apodization and long-period gratings | OptiGrating |
| Model of the readout system and the whole system | OptiSystem |
| Automating the grating spectrum measurement | OptiInstrument |
The most common mistakes in FBG sensor design
Skipping apodization
A spectrum with strong sidelobes makes it harder to determine the peak position and reduces the accuracy of the sensor.
Not separating temperature and strain
Both quantities shift the spectrum, so a sensor without compensation can confuse one with the other.
Sensitivity from a table instead of from the model
The real sensitivity depends on the fiber, the grating, and the packaging, so it is worth determining for a specific construction.
Overlapping spectra in a multiplexed system
If the spectra of gratings in one fiber overlap when they shift, the measurement is ambiguous.
Neglecting the readout system
Measurement accuracy also depends on the source noise and the analyzer resolution, not only on the grating.
Summary
A sensor based on a Bragg grating is simple in principle, but its quality depends on the details of the spectrum: peak width, sidelobe level, and reflectivity. These parameters can be designed and checked on a model.
OptiGrating lets you design a grating using coupled-mode theory and the transfer matrix method, introduce apodization, and check the spectrum. OptiSystem lets you place it in a model of the readout system, and OptiInstrument lets you automate the measurement of the manufactured grating. This makes the path from a sensor idea to a calibrated device shorter and better controlled.
Manufacturer resources
OptiGrating product page: Optiwave: OptiGrating
Overview of OptiGrating capabilities: Optiwave: OptiGrating Overview
Measurement and automation with OptiInstrument: Optiwave: OptiInstrument