MTA CTI Trending (Jan 2006)

MTA group is monitoring AICS charge transfer inefficiency (CTI) continuously from the beginning of the mission. In this memo, we summarize what we see on MTA CTI page.

CTI of front side CCDs are very sensitive to a CCD focal plane temperature and a particle background. However, these effects are easily removed. (Grant et al (2005). See also Removing the CTI/Background Anti-Correlation by Grant, and Does The CTI Correction Factor Changes With Time? for related topics ). After removing these effects, the trend of the CTI degradation is a very smooth linear increase. The back side CCDs are, on the other hand, not sensitive to these two factors, and the changes of CTIs are also smaller compared to those of the front side CCDs. For the last 5 years, the rate of CTI change is very steady and gradual of ~ a few 10e-6 CTI/yr. Even the Oct 2003 CME did not affect the rate of the change (See Did ACIS CTI change after the Oct 2003 CME? by Grant).

The following table and the plots show CTI trends after corrected temperature dependence of CTI, and particle background (Mn K-alpha). For comparison, CTIs with long exposure ( >7000 sec) and a focal plane temperature lower than or equal to -119.7 C are also plotted (both of them are particle background corrected, except for backside CTI). For I-array CTI computation, CCDs i-0, i-1, i-2, and i-3 are included. For S-array CTI, only s-0, s-2, s-3, and s-4 (front side CCDs) are included.

Since the temperature corrections uses the long exposure data to estimate correction factors, the similarity of the slope values between these two data sets is expected. Since the error of the slopes are about 0.07e-06 for all, the slopes computed on these two data sets are basically identical.

	Focal Temp<-119.7C Time >7000sec	Focal Temp Corrected Time >2000sec
I-array CCD:	2.750e-06	2.929e-06
S-array CCD:	3.132e-06	3.074e-06
Backside (CCD5):	1.029e-06	1.043e-06

Note: unit is CTI/yr, and errors are around 0.07e-06 for all the slopes. The values are those of Jan 9, 2006.

Occasional outliers on the plots below are due to CTI computational problems, not a real physical effect. For example, a point on Day of Mission (DOM) = 1154.7 (09/18/2002) has one anomalous CTI reading from CCD 0 node 2 (4.182), probably due to missed line fitting on the peak of the Mn K-alpha. All other nodes are normal for this data. Most outliers have shorter integrations time, which makes finding a peak more difficult.

plot 1: I-array CCD CTI

plot 2: S-array CCD CTI

plot 3: Backside CCD (CCD 5) CTI

The following table shows CTI change per year and average CTI of December 2005 of quad 0 of each CCD (based on Mn K-alpha). Except CCDs 5 and 7, all CTIs are corrected for the focal temperature and particle background. The CTI and the rate of the change are much smaller for the back side CCDs, and among the front side CCDs, CCD 4 has much higher CTI and the higher CTI change rates.

For more detail analyses of CTI data, please refer to a paper by Grant et al (2005). There are also arrays of analyses and papers related CTIs posted by Grant (including the paper above) on http://space.mit.edu/~cgrant/cti/cti120.html.