Compensated neutron-gamma density (NGD) is a nuclear density measurement that can be made without a chemical gamma ray source. The measurement uses fast neutrons from an electronic pulsed neutron generator (PNG). These fast neutrons penetrate the formation and collide with and excite atomic nuclei. The nuclei rapidly return to their original energy state and emit gamma rays, some of which are scattered back to the detectors. The count rate of detected gamma rays, compensated for fast neutron transport effects, depends on formation electron density.

This new measurement is desirable where operating with a chemical source is prohibited or the risk of loosing the drill string in hole is high. The NGD measurement was first deployed on logging-while-drilling (LWD) equipment to minimize the impact of stuck-pipe events and to provide a density measurement with greater depth of investigation into the formation. The first implementation of the measurement used a ruggedized PNG as a source and a single detector, but the single detector configuration rendered the measurement sensitive to standoff. A caliper input was required to correct this effect, but introduced additional uncertainty. The measurement evolved to a standoff-compensated density using two detectors at different spacings from the PNG. The two detector responses are processed by means of a spine-and-ribs algorithm to correct for borehole effects, much like traditional cesium-based density measurements. This compensated NGD measurement provides standoff- compensated formation bulk density for formation evaluation, without using a chemical source.

The compensated NGD algorithm has been tested on various field datasets. The results are compared against an earlier uncompensated NGD algorithm and traditional gamma-gamma density (GGD) measurements. This paper provides an overview of the current compensated algorithm's advantages and highlights limitations that are currently being addressed through ongoing development.

You can access this article if you purchase or spend a download.