Research on multi-probe energy response compensation for X/γ dose rate meter

Dose rate meters operating in pulse counting mode usually encounter the problem of uneven energy response. While current methods of coating a single probe with a metal layer can effectively improve the flatness of the energy response, the energy range of the flat response is limited. In this research, a multi-probe energy response compensation method was proposed to solve the problem of uneven energy response of a CsI(Tl) dose rate meter. In this method, different relative energy response curves were obtained by adding different hardware compensations to a probe. Then, two to four relative energy response curves were selected and weighted to obtain a flatter response within the energy region of interest. Specifically, first, 51 compensation schemes were obtained by changing the geometry and material parameters of the CsI(Tl) probe compensation layer. Second, the relative energy response curves of CsI(Tl) probes with 51 compensation schemes were obtained by MCNP simulation. Finally, the weight coefficient of each relative energy response curve was determined by overdetermined equations, and the combination with the smallest relative energy response deviation was selected. Within the energy range of 80-1500 keV, the optimal two-probe compensation scheme was selected from one to three probe compensation schemes. After the dual probe combination compensation, the relative energy response deviation ranged from -23.0% to 5.0%. Within the energy range of 50-3000 keV, the compensation schemes of one to four probes were traversed. The optimal three-probe compensation schemes were selected. After combined compensation, the relative energy response deviation ranged from -27.3% to 15.3%. Furthermore, the compensation effect of multiple probes was better than that of single probes in both energy regions of interest. Simulation results demonstrated that our proposed method can significantly improve the flatness of the energy response of dose rate meters based on CsI(Tl).