Dear Mohammed,

There can be many explanations to the fact. But first of all can you acquire the spectrum and have a look at it? If the spectrum is not stable for example has interference modulation on it then the peak wavelength will be clearly unstable. This effect can be produced if you have cleaved the fibre too near to the FBG (5 cm or less). We call it here Etalon effect. Another source of noise which you can consider is birefringence. Each fibre shows a little birefringence and this means that your spectrum is in reality two spectra very close each other. For various reasons if the stress in the fibre is changed during measurement the peak wavelength moves and the effect can be as great as 20 pm. Third possibility are temperature fluctuations grater than 0.1 Celsius durning the measurement due to insufficient thermal isolation and/or stabilisation. Hope this will help.

Thank you for your explainations.

I think you should accurately measure the temperature during each time you go in your experiment. Note that the temperature greatly affects the spectrum by about 15 pico meter (may be more) for just one degree celsius of change.

The temperature response of the Bragg wavelength is caused by two factors, the change in the effective refractive index of the glass due to the thermo-optic effect, and the thermal expansion of the glass, which in turn changes the grating pitch. The thermal expansion coefficient of the fiber is approximately 0.55×10^-6 /°C for fused silica), whereas the thermo-optic coefficient is about 8.3×10^-6 /°C for germania-doped silica-core fiber. For silica fibers, the thermo-optic effect is the dominant factor accounting for approximately 95% of the observed shift in the Bragg wavelength.

Thank you Taha, That what I did during the last tow days. The temperature is very effective in the measurements of the wavelengths. Thank you again.