Fiducial light drift

This page shows the drift of fiducial light positions over the course of the mission. Since launch, there has been a drift of about 12 arcsec along both spacecraft Y and Z. This apparent drift in position can be due to a combination of any of the following factors:
  • Shifting ACA alignment
  • Physical shift of the SIM relative to the OBA
  • Bending of the OBA
  • Alignment drift in the FTS
The first three factors are nearly degenerate, and could only be distinguished by determining if the X-ray PSF of sources at the nominal aimpoint has degraded due to the 9 arcsec drift. Only the last factor (FTS alignment drift) produces a change in the boresight calibration (i.e. celestial location offset). Based on the time history of the boresight alignment, the FTS alignment has drifted 2.7 arcsec in the Y-axis and -1.2 arcsec in the Z-axis.

Fid light drifts (Updated 09/26/23)

The data for these plots were generated by extracting Aspect L1 centroid data and fid properties data for all archived observations. The last 4 ksec of data are used, to minimize scatter due to thermal transients. Then a median centroid position is calculated for each fid light. Finally, for each science instrument and fid light, the "drift" is calculated by removing both the mean centroid for that fid light (over observations in early 2003) and the SIM-Z offset (for that observation). In the plots below, different colors and symbols represent the different fid lights for each SI.

Fid light commanded search position versus telemetry (Updated 09/26/23)

The data for this plot represents the difference in Y and Z between the commanded search position for fids and the actual observed location. The red dashed line represents the fid light search box half-width, at which point the likelihood of a failed fid acquisition is high. Within the green dashed region (15 arcsec half-width) the likelihood of failure is low.

ACA CCD cooldown in late 2006

Starting around 2006:320 there is a strong trend in the fid light positions, primarily in the minus-Y direction. Over an interval of about 40 days the alignment has drifted approximately -8 arcsec in Y and -4 arcsec in Z. This coincides with the ACA cooldown from -15 C to -20 C and it is very likely that the additional heat generated by the CCD TEC is responsible for the alignment change.

Functional fit to drift data (pre-2003)

An exponential of the form

Y = Ampl * exp( -[Time - Time_launch] / Tau )

has been fit to all of the data prior to June 2003. The best fit parameters are:

   Ampl (arcsec) Tau (days)
Y-axis 3.7 300
Z-axis 7.6 370

Sharp shift around Jul-2003

Near Jul-2003 there is a sharp shift in alignment of nearly 6 arcsec. This is associated with the swap from IRU-1 to IRU-2, and is discussed in detail in the Alignment drift during IRU swap and ACA cooldown document.
Tom Aldcroft

Last modified:02/18/14