Many techniques for measuring neutron star radii rely on absolute flux measurements in the X-rays. As a result, one of the fundamental uncertainties in these spectroscopic measurements arises from the absolute flux calibrations of the detectors being used. Using the stable X-ray burster, GS 1826-238, and its simultaneous observations by Chandra HETG/ACIS-S and RXTE/PCA as well as by XMM-Newton EPIC-pn and RXTE/PCA, we quantify the degree of uncertainty in the flux calibration by assessing the differences between the measured fluxes during bursts. We find that the RXTE/PCA and the Chandra gratings measurements agree with each other within their formal uncertainties, increasing our confidence in these flux measurements. In contrast, XMM-Newton EPIC-pn measures 14.0$\pm$0.3 % less flux than the RXTE/PCA. This is consistent with the previously reported discrepancy with the flux measurements of EPIC-pn, compared to EPIC-MOS1, MOS2 and ACIS-S detectors. We also show that any intrinsic time dependent systematic uncertainty that may exist in the calibration of the satellites has already been implicity taken into account in the neutron star radius measurements.
↧