FPGA-Based Laser Frequency Stabilization

Abstract

The goal of this work was to use the Red Pitaya and a suitable software solution to improve the stability of the laser in order to obtain a fast, cost-e"ective solution as an alternative to expensive classic measurement equipment. First, the necessary theoretical basics for control engineering, the isotope 133Cs, the spectroscopy method, the operation of a lock-in amplifier, Fourier transformation and Allan variance were covered. This was followed by a detailed explanation of the locking process using saturation spectroscopy and EIT. The stabilized laser was overlapped with a reference laser, and the resulting beat signal was recorded and analyzed by its FFT. Subsequently, the Allan variance was calculated for the measured frequencies. In summary, locking the laser with the help of the Red Pitaya and the Lock-In+PID software worked. The bandwidth of the laser becomes narrower using the correct tuning parameters of the PID control system.The results of a series of measurements were presented in the evaluation. After locking the laser to the optical transition of 133Cs, the variance of the laser frequency was determined to be (5 ± 0.1) MHz. After optimizing the tuning parameters, a minimum variance of (2.5 ± 0.1) MHz was calculated. It is therefore possible with the setup described to reduce the fluctuations of the laser frequency in the MHz range. Depending on the experimental use of the stabilized laser and the required stability, the Red Pitaya with the help of the lock-in software can be a good alternative to expensive classical measuring instruments. Other locking techniques, such as the Pund-Drevor Hall method, would also be possible with the same software and could lead to better results. If it would be possible to implement the error signal as a function over time graphically in the software, this would reduce the tuning e"ort considerably. The reaction times of the Red Pitaya fluctuate from time to time, which makes the locking process even more di!cult. In general, it was possible to minimize the bandwidth of the laser by a few MHz to improve the initial situation for working with optical memories. There are other software solutions for the Red Pitaya, which were not tested in this work due to time constraints. In the following, these could be tested and compared with the previous results. In conclusion, the Red Pitaya board has once again proven to be a good, cost-e"ective alternative to expensive classical measuring devices.