HETGS Line Profiles  

MARX vs. ChaRT

 

MARX vs. ChaRT

The internal HRMA model in marx is a simple ray tracing model with multi-conic sections. On the other hand, ChaRT (driven by SAOSAC) uses a finite-element method to perform ray traces more rigorously. Supposedly the ray trace with ChaRT could produce more articulate PSF. So we check to see if there are any noticeable difference between marx and ChaRT.

Comparison of Integrated Line Profiles

HEG

Comparison of MARX and ChaRT line profiles (HEG +1)
Comparison of MARX and ChaRT cumulative line profiles (linear) (HEG +1)
Comparison of MARX and ChaRT cumulative line profiles (log-log) (HEG +1)
Figure 15: Comparison of MARX (red) and ChaRT (black) integrated line profiles for HEG +1 order. The line simulation is performed at E = 1.5 keV.: (top) comparison of integrated line profiles; (middle) comparison of cumulative line profiles in a linear scale; (bottom) the same as the middle plot above in a log-log plot.

MEG

Comparison of MARX and ChaRT line profiles (MEG +1)
Comparison of MARX and ChaRT cumulative line profiles (linear) (MEG +1)
Comparison of MARX and ChaRT cumulative line profiles (log-log) (MEG +1)
Figure 16: Comparison of MARX (red) and ChaRT (black) integrated line profiles for MEG +1 order. The line simulation is performed at E = 1.5 keV.: (top) comparison of integrated line profiles; (middle) comparison of cumulative line profiles in a linear scale; (bottom) the same as the middle plot above in a log-log plot.

Notice that the whooping mismatch in the red wing of LSF for HEG in Figure 15. This probably results from a slightly misaligned HRMA mirror shell used for HEG, generating more assymetric line profile than anticipated with marx.

Plots of cumulative line fluxes (cumulated from left to right across the line profile) illustrates the same trend. In the middle panel of Figure 15, the difference between MARX and ChaRT grows most dramatically after the bin #320 (which nearly corresponds to the Gaussian center of the line profile) or after the cumulative power = 0.5. The cumulative plots are all normalized. Also notice in the bottom plot of Figure 15 that the wing of LSF in MARX simulation is also underestimated when compared with ChaRT. Figure 16 shows the same for MEG +1 order. While the difference in the core of LSF is small, it is also clear from the bottom panel that the wing of LSF is still underestimated in MARX. However, the magnitude of the difference is quite small; and therefore the detector background would be likely to swamp the difference. In any case, every user should be aware that the current LSF model may not be absolutely perfect.

At this point we are waiting for the HRMA Cal group to provide us the best calibration of PSF to show which model represents true PSF more accurately (From Priv. Comm.: apparently ChaRT gives a very slightly broader PSF wing compared to the flight calibration dataset, though that doesn't mean that marx gives more accurate PSF either. So ChaRT and marx would eventually have to come to an agreement. That's for future release, folks). On the positive note, the comparison of MEG line profiles looks fine (so does MEG -1, not shown here).

Previous: Verification with Real HETG Dataset Next: Summary


This page was last updated Dec 4, 2002 by Bish K. Ishibashi. To comment on it or the material presented here, send email to bish (at) space.mit.edu.
Valid HTML 4.01! Made with JED Viewable With Any Browser