October 5-9, 2014


P2.13 Energy reconstruction in events detected in TES X-ray detectors

M.Teresa Ceballos (Instituto de FĂ­sica de Cantabria (CSIC-UC))

M.T.Ceballos (IFCA (CSIC-UC), B. Cobo IFCA (CSIC-UC), N. Cardiel (UCM), J. van der Kuur (SRON)

The promising performance (high energy resolution with imaging capabilities) that the x-ray calorimeters based on Transition Edge Sensors (TES) technology can offer, have made them the perfect bet for the high spectral resolution instrument (X-IFU) to be proposed for the recently approved ESA L2 mission ATHENA. The new technology in development for these devices, requires a different approach for the x-ray event detection and reconstruction: the electrical pulses which are the response of the device to the abrupt increase in the thermometer resistance after the absorption of an x-ray photon, must be detected in the noisy signal, then graded (quality determination) and finally their energy must be reconstructed. The energy content of the pulses is obtained through a two-step process: optimal filtering (which ends up in a value for the photon "pseudo energy") and energy calibration (to obtain the "real" values of the photon energy). Finally, the energy resolution of the detector is determined. Here we present the details of the energy calibration algorithm that we have designed for the Event Processing Software package that our group at IFCA is developing for the X-IFU Onboard Event Processor. This algorithm relies on three main basic ideas: the photon Energy and Pseudo-energy (pre-calibrated, estimated energy) exhibit a linear (or quadratic) relationship, the Energy distribution of photons follows a normal law and, as the area must be preserved, the distribution of photon Pseudo-energies can be expressed as a modified-Gaussian function which depends on the linear/quadratic parameters and the variance of the photon Energy distribution. In our approach, instead of fitting the histogram of Pseudo-energies to obtain the best linear/quadratic parameters and the variance (energy resolution), we fit the pseudo-energies Cumulative Distribution Function (CDF), avoiding the uncertainties associated to a histogram construction due to the optimal bin selection. Some results of the performance of the method based on pulse simulations are presented.

Mode of presentation: poster

Applicable ADASS XXIV theme category: Data Analysis / Pipelines