One CCD starts dropping every second frame part way through the observation. In addition in OBSID 18278, two events were reported multiple times on consecutive (undropped) frames and no events were reported from rows 1016-1022. The ACIS flight SW team believes that this behavior can be explained by a corruption in the radiation-tolerant memory of the FEP pixel thresholder. The one instance of this anomaly affected the S1 CCD, but if the explanation is truly a corruption in the FEP pixel thresholder memory, any FEP/CCD could be affected in a future anomaly. If the anomaly were to happen again the behavior of dropping every other frame should be observed, but the events that are reported multiple times and the region of the CCD with no event reported would be different depending on the exact nature of the memory corruption.
Unknown. This is the only instance in flight so far.
Most likely the anomaly will occur out of COM and we will be notified by V&V that one of the CCDs in an observation is missing every other exposure after some point in the observation.
These symptoms may be noticed:
We need to:
sot_yellow_alert
and/or brief the 9am teleconIn the somewhat unlikely event that this anomaly is noticed during an observation, and there is time either during the current comm or a following one (still on the same obsid), we have an SOP (DEA FEP Diagnostics SOP) written to intervene if the observation with the anomaly is still in progress to dump diagnostic data.
This procedure suspends the observation while the diagnostics are run, and then resumes all chip/FEP combinations except the problematic one. If the target is not on the affected chip, it may be advantageous to the user to skip the diagnostic procedure and finish the observation uninterrupted.
If it is decided to run the diagnostics procedure, do this:
sot_red_alert
and convene a telecon giving the plan, and the need for quick action to get the SOP done before the end of the science run.