To investigate the effect of moisture content, excitation frequency and normalisation model on image reconstruction with a fluidised bed dryer, an electrical capacitance tomography (ECT) sensor was mounted near the bottom of the drying chamber. An ECT system based on an HP4128 impedance analyser was used to measure capacitance and loss conductance between the electrode pairs in the sensor. It has been found that the capacitance depends on not only the particle moisture but also the excitation frequency. With a low moisture content, the relationship between capacitance and frequency is simple and linear. With a high moisture content. however, the relationship becomes more complex and non-linear. For image reconstruction, different normalisation models have been used: series, parallel, Maxwell and Bottcher models. The results show that with a low moisture content, these models give nearly the same image errors. With the increase in moisture content, the difference between these models becomes more and more obvious. With different gas-solids How patterns, the four models also give slightly different images. In the end of this paper, solids distribution and averaged solids concentration profile for dynamic test with different excitation frequencies in a fluidised bed dryer are given. (c) 2009 Elsevier B.V. All rights reserved.