hamogu
commented
9 years ago To find out which values these keywords have when I add them to MARX simulated data, I contacted Cathrine Grant. Here is her response:
I'm not a CIAO expert in any sense, so I'm not sure where any of these parameters would get used, once you get past the initial level 0 processing. Glenn would probably know better than I. They all come out of the parameter block that gives ACIS the parameters for each science run. They are not user selectable parameters, thus they don't get mentioned in the POG.
As far as I am aware, we have never changed the values of OCLKPAIR, ORC_MODE, or SUM_2x2 since launch, except possibly in the checkout period before the doors were opened. The only reason we would change these would likely be associated with some kind of major instrument anomaly, so I would think leaving these at the defaults you have below should be fine.
The below is off the top of my head. I couldn't quickly find any useful documentation, although I'm sure there's something out there.
OCLKPAIR. Every time a row of data is read out of one of the four CCD amplifiers, additional "non-physical" pixels are read out after the real pixels. For OCLKPAIR=8, you get an additional sixteen "blank" pixels per row for each of the four output nodes. These "overclock" pixels have the same bias level and electronic noise as the imaging pixels and are used to track the variation of the bias level throughout a science run and to flag data frames that are anomalously noisy. I think the overclocks are only used during event finding and bias subtraction, both of which are done on-board and again on the ground in level 0 processing.
ORC_MODE. Each ACIS CCD has four output nodes. The standard readout direction has the outer nodes (node 0 and 3) clocked towards the edge of the CCD and the inner nodes (1 and 2) clocked toward the middle. You can, however, choose to change the direction of the clocking and read out different CCD regions from different amplifiers, i.e. clock the CCD such that all of nodes 0 and 1 gets read out in the middle, for example. This would be helpful if there was some kind of hardware problem with an amplifier or some other kind of physical damage that you would trying to avoid clocking through.
SUM_2x2. Pretty much what it sounds like. The pixels are clocked such that the charge in a 2x2 pixel region is combined into one summed "pixel". It's never been calibrated and isn't particularly useful for ACIS for a variety of reasons. I think Astro-H SXI will be run using summed pixels.
FEP_CCD. ACIS has six Front End Processors (FEP) that each can talk to one of the ten CCDs. The parameters blocks contain the assignments of which FEP will talk to which CCD during a science run. In the case below the assignments are: FEP 0 = CCD 4 (S0) FEP 1 = CCD 7 (S3) FEP 2 = CCD 5 (S1) FEP 3 = CCD 6 (S2) FEP 4 = CCD 8 (S4) FEP 5 = CCD 9 (S5) Royce Buehler has a ruleset he uses to determine the FEP/CCD pairing for each parameter block and the FEP/CCD mapping can change from observation to observation. The one here is a pretty standard one for 6-chip grating observations. I'm not sure if MARX or CIAO would ever care about what FEP was being used, since all the calibration is CCD-based.
Hopefully this makes sense and helps!
Catherine
Chandra repro-4 added new header keywords to evt2 files. Before, these parameters where read from the pbk and asol files, now access to these files is no longer required for some processing tools. On the downside, these tools will now break with an error if the header keywords are not present (even though I am not convinced that they actually need them in some cases). I noticed this in the scripts that make the grating responses (mkgarf, mktgresp, fullgarf). MARX should add those keywords with reasonable value to the fits files it generates.
Here is more info on those changes, pulled from the CIAO docs:
For ACIS and HRC, three new keywords were added to capture the mean SIM offsets during an observation: DY_AVG, DZ_AVG, and DTH_AVG. These numbers are needed by tools like dmcoords, mkacisrmf, and mkwarf to correctly map detector coordinates back to the location on the instruments. Previously each tool used the information in the asolfile parameter to compute the mean value, every time the tool was run. With the mean values now simply encoded as header keywords, the tools can run more efficiently.
For ACIS, the information in the Level 0 parameter block file, pbk0, needed to compute the dead area calibration was also added to the event file headers. The new keywords include: SUM_2X2, FEP_CCD, ORCMODE, and OCLKPAIR. This information together with some existing keywords, is used by ARF and Exposure map tools when creating response.
http://cxc.harvard.edu/ciao/watchout.html
http://cxc.harvard.edu/ciao/ahelp/r4_header_update.html