Common Problems

1. The lock-in does not remain stable. Or the lock-in but does not seem to track the resonance.

   You may see a big fluctuation in either lock-in signal or resonance frequency. This effect can be caused by several things.

   First, it is possible that the frequency to which you are locked is not the true resonance, or you are completely out of resonance region. The solution is again to find manually the resonance, looking for the most pronounced signal.

   The problem could also be caused by a wrong lock-in amplifier phase. Please refer to Section .

   Finally, it could be that the PI Feedback box is improperly set. Adjust the gains until the lock-in becomes more stable. Admittedly this method is not very quantitative. Based on experience, at high polarization ($>$40%), both gains should be adjusted anti-clockwisely nearly to the end. At lower polarization ($\sim$20%), both gains should be adjusted by about 4$\sim$5 turns clockwisely. These numbers may also vary at a higher or lower EPR D2 signal amplitude.

   You may also optimize the PI circuit gains by observing how fast the circuit follows the resonance. When counter reading is stablized, change Wavetek frequency manually by 0.04 MHz (which is close to the real jump during sweeping, at a target polarization $\sim$ 40to resonance in roughly 3$$5 seconds, then the circuit is working well. If less than 2 seconds and counter reading is not stable, decrease both the relative and absolute gain. If longer than 6 seconds or lose the signal, increase both the relative and absolute gain.

2. The lock-in does not track the resonance when doing AFP flips. This is common, since the frequency shift during the spin flip can be quite large, on the order of 20-40kHz.

   First, it is possible that the PI feedback is not strong enough for the circuit to follow the resonance shift. Try increasing the absolute gain of PI circuit if possible. Second, it could be that the modulation DS 345 amplitude is too small. The size of this amplitude determines how far from the central frequency the circuit looks for the resonance. Try increasing the modulation amplitude (but do not increase too much, usually it is less than 0.8 Vpp). This will cause the counter reading less stable and you need to compromise between stablizing the circuit and following the frequency shift.