10.1	Introduction

Both the FUV XDL and the NUV MAMA detectors are subject to damage or destruction by excessive illumination. An excessive local count rate can permanently reduce the sensitivity of the affected detector region. The most likely causes are a bright spectral emission line (XDL or MAMA) or a bright source observed in direct imaging (MAMA). A global over-illumination of the detector can result in its loss. To protect the detectors onboard software monitors the local and global count rates. It shutters the instrument in case of a local violation and lowers the high voltage if a global count-rate violation is detected. The local rate is checked before an exposure is begun, while the global rate is monitored continuously during the exposure. Under certain circumstances damage can result despite the onboard safety measures, and in any event lowering the high voltage will disrupt the HST schedule and operations. Therefore, all proposed COS observations must meet count-rate screening limits with safety margins to allow for uncertainties. COS is currently the only HST instrument to use UV detectors for target acquisitions; such acquisitions must be screened, as well. It is the responsibility of the observer to screen all proposed targets and fields during Phase II preparation, and of STScI to check and enforce these limits. The COS Bright Object Protection (BOP) policies and procedures are described in this chapter.

Table 10.1: COS Count Rate Screening Limits

Detector	Source type	Type of limit	Limiting count rate (counts/s)
FUV	predictable	global	15,000 per segment
		local	0.67 per pixel1
	irregular	global	6,000 per segment
		local	0.67 per pixel1
NUV	predictable	global	170,000 (imaging) or 30,000 per stripe (spectroscopic)
		local	50 per pixel2 (imaging) or 70 per pixel (spectroscopic)
	irregular	global	68,000 (imaging) or 12,000 per stripe (spectroscopic)
		local	50 per pixel (imaging) or 70 per pixel (spectroscopic)

---

1

This local count rate screening limit is not applicable to the 1055 and 1096 cenwaves of the COS G130M grating, for which, based on recent calibration data, we expect a lower value for segment B, of order 0.2 cts/pix/s. The data to accurately calibrate this value for segment A of 1055 and 1096 have not been obtained at this time. Thus, the calibration data required to derive an accurate value for the count rate limit for both segments of the 1055 and 1096 cenwaves are still being worked on. Users should check the COS Web site for updates on the local count rate screening limits.

2

For imaging acquisitions, a count rate of 360 counts/s in the 9×9-pixel box surrounding the target (as computed by the COS Imaging Acquisition ETC) represents an equivalent safe upper limit.