


Next:Flux
correction via the
Up:ACIS/HRMA
Performance Prediction
Previous:Size
of ACIS-2C pixels
Pileup for PSF concentrated measurements
When more than one X-ray photon interaction with the CCD places charge
in the same or neighboring pixels, the ACIS event processing software has
difficulties reconstructing the energy, grade, or even number of photons.
The effect of these multiple photon charge cloud combinations is called
`pileup'. As discussed in Chapter 4,
the case of uniform illumination pileup removal is reasonably well modelled
for moderate incident flux levels.
At XRCF, and in flight when point sources are being observed, the HRMA
produces a highly non-uniform illumination of the CCD chips in ACIS. This
focussed beam is sharply peaked in the core of the HRMA PSF, with additional
flux in wings of the PSF (especially off-axis), and with corrections resulting
from the pixelization effects of the CCD and internal charge cloud spreading.
In the standard processing of ACIS data the event lists are generated
by summing the charge found in 3x3 pixel regions of the CCD. This approach
assumes that the probability of a 3x3 pixel detection cell containing charge
from more than one photon is insignificant, which is true for most astrophysical
sources. When bright (greater than 100 micro-Crab) sources are observed,
however, photons can pile up on top of each other, i.e. a significant number
of detection cells can be contaminated by charge from nearby photons. This
so-called pile-up effect can lead to several types of error[ASC1997].
-
Spectral distortion: A detection cell containing charge from a secondary
photon nearby will overestimate the energy of the primary photon.
-
Grade migration: The spatial distribution of charge among the nine
pixels in a detection cell (an event's grade) is an important event
property that is used during data analysis. When extra charge is added
to a detection cell, the grade is often changed.
-
Raw event rate underestimation: The raw event rate can be underestimated
because closely spaced photons are detected as a single event.
-
Clean event rate corruption: If an event list is filtered using
the corrupted energy or grade properties, then the resulting number of
``clean'' events will be inaccurate. Some grade filtering is always applied
to reduce the number of events associated with the particle background.
-
Point spread function (PSF) distortion: The severity of pile-up
depends on the surface brightness of the incident photon distribution.
Thus the core of the PSF suffers more pile-up, and loss of events, than
the wings do.
Data to evaluate these pileup effects on ACIS performance were collected
at the X-ray Calibration Facility (XRCF) in 1997 February and April during
Phases F and H of the calibration operations. The primary goals were to
investigate possible trends of pileup with source flux and grade selection,
and to compare in detail the measured pileup fraction with current theoretical
models, in order to assess the general validity of the latter for predicting
pileup for in-flight observations.
In the following sections we present our efforts to understand and correct
or ameliorate pileup effects.



Next:Flux
correction via the
Up:ACIS/HRMA
Performance Prediction
Previous:Size
of ACIS-2C pixels
Please address comments and questions to Dr. John Nousek ( nousek@astro.psu.edu
)