Proportional-Derivative-Integral controller for laser stabilization at HU Berlin.
23.08.2019
Successful laser locking with this PID. Coefficients used for locking : Kp = 0.2, Ki = 0.2, Kd = 0.0.
10 dB attenuator used just after DAC.
Laser currnent controller used in locking experiment: the one above, HU rev 7.2017
26.08.2019 The sharpest peak were achieved with coefficients: Kp=0.1, Ki =0.05. Adding derivative part only broadened thee peak.
28.08 Kp 0.0025 Ki 0.001
29.08 Network analyzer used to determine PID's transfer function. Every part of PID controller behaves as expected when given input across wide range of frequency. The experiment is being conducted wiith parameteres as follows: sweep range: 50 Hz - 10 kHz amp - 10 dBm input: only A - 20 dB attenuator was used for calibration, otherwise NA's input was overloaded
Following coefficients should be set to gather data:
Kp 1
Ki 0
Kd 0
Kp 0
Ki 1
Kd 0
Kp 0
Ki 0
Kd 1
/ x3
Setting, wich was the best for locking laser:
Kp 0.0025
Ki 0.001
Kd 0
/x5
Kp [0.0025 0.025 0.25 2.5]
Ki [0.001 0.01 0.1 1 ]
Kd [0 0.01 0.1 1 ]
2.09 Implementation of PID controller with sampling using only MU units successful (hope so. it seems to work, but id doesnt seemt to improve bandiwdth) Starting measuring PID controller - firstly PID from a week before (one which uses sampling to voltages), followed by measuring PID implemented today. NA' settings: sweep range: 50 - 10000 Hy amp: 1 V (without any attenuator) input: only A