FMCG/Magnetometer optimization

The final part of magnetometer setup involves adjusting the pump and probe laser detunings to optimize the optical pumping and Faraday rotation, respectively. This will assume that the magnetometers have been set up and nulled.

Pump Optimization

While not absolutely necessary, it is generally good practice to tune the pump laser near resonance (8.4 kΩ) and re-null the fields at that detuning. Else the optimal laser detuning will likely point toward whatever detuning the magnetometers were nulled at.

1. With the magnetometers nulled on resonance, apply a low-frequency (5-20 Hz) sine wave to the Y room coils. Adjust the amplitude so the applied field does not drive the magnetometers out of the linear regime of the dispersion shape. Because sweeping the pump detuning often changes the DC value of the field (due to changing AC-Stark shifts), it's also wise to reduce the gain of the I-V converters somewhat (5-20 μA/V).
2. Return to the Laser Scan v1.1.vi program, or re-open from the FPGA Magnetometer.lvproj project it if it's been closed.
3. Most of the parameters will remain the same as the absorption scan, with two notable exceptions:
   1. Click the light next to Scan Pump to tell the program to scan the pump laser rather than the probe
   2. The Differential? button should be enabled (depressed). In differential mode, the program will record a short segment of the sine wave and record the peak-to-peak value of that segment.
   3. The temperature range [5-10 kΏ], Number of points [200], dwell time [100 ms], and PreScan Dwell [10 s] can remain unchanged.
4. Click Run to run the program. The program will ramp the laser detuning and record the peak-to-peak value of the sine wave at each detuning. Far off resonance, the pumping rate will be very low and the response will be small. As the laser approaches resonance, the response will increase. Right on resonance, depending on laser intensity and rubidium density, the response may drop as too many photons are absorbed at the front, leaving unpolarized atoms at the back. Moving slightly away from resonance, the response will likely increase, followed by a decrease as the pumping rate again becomes small. Again, a file containing the data will be recorded at the location indicated by Path Out.
5. Determine a detuning that gives good response for all channels. It's unlikely that all cells will experience optimum performance at the exact same detuning, but small compromises on each channel can be made such that they all perform reasonably well. Set the laser detuning to that value (first by hitting the Utility/Local button, and then TEC Control - Set and entering the value using the keypad). Be sure to re-null the fields after this detuning has been selected, as the nulling fields required (especially in the z direction) will depend on the laser detuning (again, due to AC-Stark shifts).

Probe Optimization

Probe optimization is done very similarly to the pump optimization. The only difference is the Scan Probe light must be selected in the Laser Scan v1.1.vi program.

The probe response is a convolution of the faraday rotation signal (which peaks near the resonance line) and the probe absorption, which hits a minimum near the resonance line. Generally this function will optimize near roungly 75% transmission. Again, it is unlikely all magnetometers will optimize at exactly the same point in detuning space, but choose a detuning that allows all channels to perform adequately. Hit the Utility/Local button to disable computer control, then set the detuning appropriately.