Recent advances in the photon counting, imaging readout for microchannel plate (MCP) detectors has led to a substantial improvement in their spatial resolution. The spatial accuracy (similar to 7-10 um) of an MCP detector with a cross strip (XS) readout has been shown to be limited by the MCP pore size (< 10 um). In this paper, we study the ultimate resolution limits of the XS readout itself. The model allows us to determine the requirements on the anode's geometry and the signal processing electronics in order to reach a particular spatial resolution. The optimal detector parameters, such as the width of the charge footprint at the anode (determined by the distance and the field between the MCP and the anode), and the gain of the detector can also be found with the help of our model. The model indicates that the optimum full-width at half-maximum of the charge footprint distribution at the anode is a factor of -1.6 larger than the anode period. Given a noise of charge sensitive amplifiers of 350 electrons rms each we predict that the MCP gain can be as low as 2.5 x 10(5) for this detector to resolve similar to 7 um features.

Results of our modeling also indicate that the accuracy of the position obtained for center of gravity centroiding of the charge distribution is inferior to fitting a Gaussian-like analytical function, providing the geometry of the anode is accurate enough. The model predictions are compared with the experimentally measured images and reveal the critical parameters (anode's geometric accuracy and amplifier noise), which can be improved in future detectors.

The combination of a photocathode, microchannel plate stack and photon counting, imaging readout provides a powerful tool for high dynamic range, high spatial resolution, and high timing accuracy detectors for the X-ray to visible light spectral range. Significant improvements in spatial resolution, quantum efficiency, and background rate have made these devices attractive for many applications and are being applied to completely new research areas. We describe the latest developments in high-resolution cross-strip (XS) anode readout for photon counting imaging with microchannel plates (MCPs). We show that the spatial resolution of an MCP detector with the cross-strip readout is now limited only by the MCP pore width. For the first time, we show images of resolution test masks illustrating the exceptional spatial resolution of the cross-strip readout, which at the present time can resolve features on the scale of similar to7 mum with MCP gains as low as 6 x 10(5). Low-gain operation of the detector with high spatial resolution shown in this paper is very beneficial for applications with high local counting rate and long lifetime requirements. The spatial resolution of the XS anode can be increased even further with improved anode uniformity, lower noise front-end electronics, and new centroiding algorithms. This could be important when MCPs with smaller than existing 6 mum pores become commercially available, thus improving the detector resolution down to the few micrometer scale.

Micromachined silica microchannel plates (MCPs) under development have a number of advantages over standard glass MCPs and open completely new possibilities in detector technologies. In this article we present the results of our studies on the thermal properties of silica microchannel plates (sMCPs). Similar to standard glass microchanel plates the resistance of silica MCPs was measured to change exponentially with temperature with a negative thermal coefficient of -0.036 per degreesC, somewhat larger than that of standard glass MCPs. The resistance also decreases linearly with the applied voltage, with the voltage coefficient of -3.1x10(-4) V-1. With the knowledge of these two coefficients, our thermal model allows the calculation of the maximum voltage, which can be applied to a given MCP without inducing a thermal runaway. A typical 25 mm diam, 240 mum thick sMCP with 6 mum pores has to have the resistance larger than similar to30 MOmega to operate safely at voltages up to 800 V. With this model we can also calculate the time required for a given silica MCP to reach the point of thermal equilibrium after a voltage increase. We hope that the ongoing efforts on a proper modification of the sMCP semiconducting layer will lead to the production of new MCPs with a small negative or even a positive thermal coefficient, reducing the possibility of thermal runaways of low-resistance MCPs required for high count rate applications. (C) 2004 American Institute of Physics.