Jul 03 2008

How do FBG sensors work?

Published by at 10:32 am under Sensors

FBGs are essentially reflectors built inside the core of an optical fiber. The reflectors are made by permanently altering the refractive index of the core. FBGs are off-the-shelf items today, and can also be made to order to fit particular applications (e.g., many FBGs in an array with irregular spacing on a single fiber).

Each FBG reflects a certain narrow slice of spectrum. Such slices typically have a smooth Gaussian shape. The center (i.e., top) of the reflected Gaussian peak is what is used to make measurements (aka, the “center wavelength”). The center wavelength is a digitally encoded zero point for each sensor — one that doesn’t change with time. This is one fundamental characteristic of FBG sensors that make the technology so valuable for long term monitoring of structures.

A FBG’s peak shifts (to a higher or lower center wavelength) when either the fiber is strained or its temperature changes. When the strain or temperature change is returned to the zero point, so does the sensor reading, i.e., there is no hysteresis.

FBGs are specially packaged to isolate the measurement property of interest. For example, the Micron Optics os3100 strain gage uses a steel carrier to transfer strain from the structure to the FBG, whereas the os4100 temperature gage isolates the FBG from strain. Used together, one can easily collect temperature-compensated strain measurements.

If you’re buying FBGs for your application, consider these specifications:

* FBG length of 10mm, >80% reflectivity, 3dB bandwidth of 0.25nm, 15dB isolation (that’s the “clean” region of the FBG peak), and a tensile strength of 150kpsi. This strength is equivalent to a proof strain of 15,000ue. Stronger and shorter FBGs are available, but you trade other properties to improve these characteristics.

* Also consider fiber coatings. Most fibers and FBGs are coated with acrylate. This dates back to FBG’s telecom roots. Polyimide coatings, however, have a higher temperature tolerance (250 C vs. 80 C for acrylate), and are far superior in transferring strain to the fiber core.

Tom Graver
Director, Optical Sensing
twgraver@micronoptics.com

10 responses so far

10 Responses to “How do FBG sensors work?”

  1. Optics » How do FBG sensors work?on 04 Jul 2008 at 9:14 pm

    [...] How do FBG sensors work?FBGs are essentially reflectors built inside the core of an optical fiber. The reflectors are made by permanently altering the refractive index of the core. FBGs are off-the-shelf items today, and can also be made to order to fit … [...]

  2. Johnon 07 Jul 2008 at 3:33 pm

    Very interesting points but also very basic. Would like to have more details and specifing, how they are manufactured, applications, etc
    John

  3. Tom Graveron 07 Jul 2008 at 4:29 pm

    John,

    For much more detail you could take a look at the Micron Optics Optical Sensing Guide at http://www.micronoptics.com/pdfs/Micron%20Optics%20Optical%20Sensing%20Guide.pdf

    It gets into the principles of operation and manufacturing techniques for FBGs.

    We most often use the strip-write-recoat technique for the FBGs in our sensors. This manufacturing approach offers a good balance of optical properties, coating choice, strength and cost.

    I’ll work to get specific application examples into the blog. There are so many, and folks keep thinking of more. Most applications really do leverage one of the inherent advantages of fiber optic sensors, like immunity to lightning or corrosion resistance, or high strain capacity.

    Do you have a particular application in mind? I’d be happy to share, in this forum, what I know about your specific application. Also, feel free to contact me directly to discuss your unique needs.

  4. jannahon 19 Aug 2009 at 9:51 pm

    hi..

    may i know what is the advantage of using FBG as a sensor compared to single mode fiber (smf)?
    what is the criteria needed when characterize FBG?

  5. Tom Graveron 20 Aug 2009 at 2:55 pm

    Jannah,

    Plain SMF or MMF (multi-mode fiber) is used as the sensing element in systems using Raman or Brillioun backscatter measurement techniques. These systems depend on the fact that the behavior of light in plain fiber is a function of strain and temperature in the fiber. Thus, plain fiber can be used as an effective sensor over many kilometers. Look at http://www.sensortran.com and http://www.omnisens.ch for more on these “distributed sensing” technologies.

    FBGs can be thought of as a quasi-distributed sensing approach. Each FBG focuses its measurement on a small and discreet sections of fiber, and many FBGs can be located on one fiber. For example, if you want to measure at 1kHz the strain at 100 critical points around the root of an airplane wing, one would likely choose FBG strain sensors. However, if you need to know the average strain every 10 minutes each meter along 80km of a pipeline, a distributed approach may be best.

    I’m not sure I understand your question about characterizing the FBG. Users of our FBG sensors really do not need to worry about optical characteristics. The sensors, measurement instruments and software make it easy to translate optical wavelength information into the engineering parameters of interest, e.g., strain, temperature, acceleration, and pressure.

    Tom

  6. jannahon 21 Aug 2009 at 1:10 am

    thank you Tom

    for your information, i’m a student n doing a research on optical fiber sensor. i’m just begin my study.
    i’m glad to have a feedback from you because i dont have anybody to discuss about optical fiber sensor except my supervisor.

    thanks again..

  7. Will Womackon 16 Dec 2010 at 5:46 pm

    Hello, Tom. I hope all is well. Do you have any sources for calculating strain on 80 um diameter fiber in pure tension if the tensile load is known? (Similar to the Corning stress calculator for 125 um diameter fiber)

  8. Tom Graveron 17 Dec 2010 at 11:50 am

    Hello Will,

    Assuming the material properties are the same, the strain be inversely proportional to cross sectional area of the fiber.

    So the stress on an 80um fiber will be higher by a factor of 62.5^2 / 40^2 or 2.441. Since Young’s Modulus is the same as a 125um fiber the strain (strain = stress/YM) will also increase by a factor of 2.441.

    In summary the Corning calculator could be used to convert tension to stress and strain for a 125um fiber, then multiply the resulting stress and strain by 2.441 for a 80um fiber.

    Tom

  9. PImfgon 13 Sep 2011 at 8:46 pm

    Great Post! Thanks Tom for the information that you have been posted, It is very detailed. By the way I just want to ask since that it is posted back in 2008, do we have an updated version of this one?

  10. Tom Graveron 26 Sep 2011 at 4:30 pm

    PImfg,

    The basic technology is unchanged, but there are advances in sensor, instrument and software.

    New sensor packages are coming on line regularly. Our latest release is a 50mm travel displacement gage, the os5100. It’s a rugged package that our civil structures customers have been seeking for a long time. They will integrate their new displacement measurements with strain, temperature, and acceleration all on the same fiber, all using FBGs.

    Instruments have expanding capabilities as well. For example, the new sm690 can sample FBGs at 2MHz. FBGs are already used in blast and ballistics measurement, now the higher sampling rates can help characterize these events with more precision.

    The ENLIGHT software has an expanding suite of data management tools. This is important as the numbers of FBGs and scan rates increase. ENLIGHT still makes it easy to move from optical properties to the real measurement properties of interest, e.g., strain and temperature.

    Tom