Comparing Cesium and Potassium magnetometers

Abstract

Geometrics staff of engineers and physicists have reviewed the article Development of a high sensitivity potassium magnetometer for near surface geophysical mapping published in the May issue of First Break. We have found the article contains inaccuracies and some theoretical misstatements that we feel should be addressed. To that end we offer the following comments: 1. Equation (4) (on page 83) is incorrect in that the author has ignored the effect of bandwidth on system performance. It is not the total noise that is of concern but the noise density (typically in nanovolts per root Hz) that is at issue. Components in the noise that are far from the Larmor frequency are easily filtered out in the electronics. However, there is no way to effectively remove noise that is very near the Larmor frequency. This is why, in spite of its much narrower line width, the performance of the Potassium is not better than that of a Cesium magnetometer. We can verify this through direct comparison of published specifi- cations. 2. The Cesium magnetometer is capable of 0.0004nT/√Hz- RMS. It should be noted that there are several Cesium magnetometer array installations using high performance counting systems wherein low noise data is being collected at 100Hz or 120Hz on up to 7 or 8 sensors simultaneously (Naval Research Labs and Oakridge National Labs MTADS systems). All statements referring to a maximum of 10Hz sample rate and low sensitivity specifications (0.05nT) are in error.