---+++ Intro

This page describes the calibration data for the April 2009 APEX-SZ Observing run. This set of data is important to all the April 2009 data set, including the high redshift clusters XMMJ1230 and XMMJ2235.

In analysing the high redshift clusters, I've come across some difficulties in terms of calibration- mainly because we do not have any good beam maps or calibrations from Mars, which was our standard tool for previous runs. This page describes these challenges, how I'm addressing them, and the cross-checks we're derived to test our results.

Calibration

This set of data requires special attention for three reasons.

1. It was taken after significant work on the cryostat focal plane, so it's possible individual detector beams have changed.
2. It also includes two new wedges which do not have any previous beam measurements.
3. There are no high-quality Mars beam measurements--> We observed mars only once, but the data was low quality (because of XXX), and un-usable as a calibration. As such, we need to derive our beam maps and calibration from other sources.
The beam measurements, calibration proceedures, and tests for beam saturation are discussed below.

Calibrators

RCW38

• Has been used in the past. We know the temperature of RCW38 from observations of it during previous runs when we had good mars calibrations (i.e. our knowledge of RCW38 is bootstrapped off mars).
• RCW38 is too big to be used for beam maps, so we need beam profile fits and pixel offset positions from other data, such as the saturn jupiter scans.
• Assuming a good knowledge of our beams from is in hand, we know the accuracy of the RCW38 calibration is good to 5% from our previous work when we had both mars and RCW38 observations during the same day in previous runs.
  

Saturn/Jupiter

• As point sources, they can give both beam measurements and absolute calibrations.
• Unfortunately, the signal from saturn/jupiter may have saturated some detectors, especially the new wedges which have lower G.
• Also need to verify that the alignment of Saturn's rings is ok, and doesn't mess up the absolute cal.

Skydips

• The large atmospheric signal seen by the whole array during a skydip can be used to get a relative calibration between detectors, but is useless for a day-to-day absolute calibration.

---

The Calibration data can be broken into two groups. Beam Measurements (Beam offset and FWHM) and Absolute calibration (or beam amplitude)
Beam Measurements

• The beam measurements are not expected to vary from day to day, so we need at least one good beam measurement for the April 2009 run.
  
• Assuming the Saturn and Jupiter Maps are good (they are not saturating the detectors), we can make individual beam templates from the average beam values.
  

here is an example of beam maps from April 2009, taken from observations of XX.

Notice the high elipticity and wide profiles for some of the new detectors near the edge of the new edges (lower right). This may be their actual profile, but it could also be that these detectors saturated on the point sources. (see below for saturation cross-check)

Note: This template still needs to be compared to the individual beam maps to ensure that bad days are not being included.

Cross-Checks:

• Compare the beams for all the old wedges to the beams from previous runs. We expect the differences should be relatively small, assuming no significant changes to the optics from opening/closing the dewar. Will confirm that saturn/jupiter derived beam maps are just as good as mars.
  
• We don't have previous beam maps for the A03 and A02 wedges. Our saturn/jupiter maps will be flawed if the detectors are saturated. We can test for saturation by comparing their absolute calibration off of Saturn and Jupiter with the calibrations from RCW38 and relative skydip calibrations (which should not be saturating the bolometer - these calibrations need the beam maps to be good). If they are consistent, then we have some confidence that the beams are not saturated. If they are inconsistent, we're kind of stuck, with no good means of obtaining beam maps for the new, saturated detectors.

Absolute Calibration

• We can now calibrate our data using RCW38 (which was calibrated off of APEX-SZ Mars observations during previous runs). The 'bootstrapping' method has matched the Mars calibration with a 5% error.
  
• Skydips will also give an independant measure of the relative beam amplitudes for comparison (NOTE: Both the RCW38 and the skydips will assume the same beam sizes).
  

Assuming Saturn and Jupiter don't saturate the beams, we can also use their beam maps as an independent check on that day's calibration.
see: calibration.pdf

---

Tests

BeamTemplate - Compare the April 2009 beam template to August 2007 Mars measured Beams

---

*Sept 16*

Figure1. %Difference between RCW38 Calibration and Saturn Calibration Data

Figure2. Calibration Data for bad day.

Something has gone wrong with the rcw38 calibration data for this scan. This causes the map for this day to have very low map noise (in units of Krj), so these maps get too much weight in the final map coaddition.
Figure3. Typical Error Map

Figure4. Error Map for Scan calibrated with Scan 13695

Figure5. Scan 13695 Beams

The beams with very low adc_k factors shown in figure 2 are highlighted in green. This explains why a small area of the weights map gets too much weight.
Figure6. Recent XMMJ1230 Map

Note. The directions of the axes have been reversed between this map and the error map above (fig. 4).
Figure7. Coadded Map excluding Scans Calibrated with 13695

-- JamesKennedy - 02 Sep 2009

---

This topic: APEX_SZ > WebHome > APR09_Calibration Topic revision: r8 - 2009-09-17 - JamesKennedy

---