Proton damage testing is currently underway at the Harvard
Cylotron Laboratory. The test device w159c1 is an unreflexed ccd typical
of the ACIS devices. Irradiation is done at room temperature with all inputs
shorted. At least two days are allowed after irradiation for any short
term activation to subside, then subsequent testing is done under normal
calibration conditions. In addition to the standard screening tests a series
of dark current maps (hot pixel finder) are made at warmer temperatures.
The typical exposure/measurement cycle takes one week.
(For low energy proton damage experments go here.)
| Date of Irradiation | Dose | Date of Testing | Image | Analysis |
| Dec 15 1997 | 50 Rads | Dec 17 1997 | -60c | Screening-report |
| Jan 22 1998 | 50 Rads | Jan 28 1998 | -55 c | Screening-report |
| Feb 02 1998 | 50 Rads | Feb 05 1998 | -55 c | Screening-report |
| Feb 12 1998 | 50 Rads | Feb 19 1998 | -55 c | Screening-report |
| Feb 27 1998 | 100 Rads | Mar 05 1998 | -55 c | Screening-report |
| Mar 16 1998 | 100 Rads | Mar 18 1998 | -55c | Screening-report |
Even at the lowest doses an oxide charging effect is seen. While the level of this glow increases with accumulated dose it is never visible below -90 c. This is the same behaviour seen with front side ccds overexposed to xrays at 500 eV and 6 keV. No amount of operation or elevated temperature (+40 c) annealing changes the level of the glow. Superimposed on the glow are a number of hot-pixels. The number and intensity of the hot-pixels decreases as the operating temperature of -120c is approached. The hot-pixels are easily distinguished from cosmic-ray or activation by the mean-bias method and because all hot-pixels are "single events". This particular ccd has some problems with repeated temperature cycles, perhaps because of the flexprint. The output noise has been increasing with time but this has nothing to do with radiation damage. Also a nasty feature toward the center and edge seems to be growing. The only screening result which is showing any significant change is the parallel CTI at 5900 eV. The parallel CTI at 525 eV is not reliable because of variable source intensity.
