my colleague was improve the system by reducing the reflected wave. and also found the maximum condition for the timing. the microwave trigger should be delayed by 167 us and have a phase 180.

then, i get the polarization signal in FID area be 340 unit or up! compare to BG, which is just 25 unit. But in order to measure the polarization dependence or any dependence, i hope to have a small fluctuation on measurements.

when i try to fix the NMR coil by using optics mounting, i found that the input impedance shown on the network analyzer is very sensitive to position, only a very tiny different can change the impedance alot. That is why the NMR signal is not so reproducible, or with a large fluctuation.

when the experiment going well, suddenly, the signal gone, it only give signal like background. i checked the coil, put it out and found it is quite distorted. fix it in a good shape and do again. nothing.

then i thought, may be the network analyzer. then i restart. but don’t know how to operate…..

finally, it is the tuner circuit broken.

anyway, today found a way to stabilized the NMR coil. as long as moving it slowly, it can reproduce similar input-impedance with 5 ohm different. that result is done with a coil-before-fixed. Now the coil is in nice shape. it goes and leaves the sample more smoothly.


  1. setup the network analyzer and use it to get back the NMR signal
  2. testing the stability of the signal
  3. fin the crystal orientation.