KMOS: Lamp flat fields
|
HC PLOTS |
slitlet position |
|
slitlet width |
|
number of IFUs |
|
FF lamp flux |
|
QC1 database (advanced users):
browse |
plot
|
Flat-field exposures are measured during daytime for the five wavelength
bands IZ, YJ, H, HK, and K and for six different rotator angles. A
standard calibration sequence uses rotator angles of -60 (i.e. 300), 0, 60, 120,
180, and 240 degrees.
Each IFU comprises 14 slices. Eight IFUs form a sub group which is mapped on
a single slit. The light is then dispersed through one of the three
spectrographs and recorded on one detector array (out of three). There are
three independent optical channels, each having eight IFUs.
|
Flat-field frame. Central area of a lamp flat-field for the H band. Each IFU slitlet corresponds to one stripe with 14 pixels width. 14 slitlets form one IFU.
|
Flat-field slitlets
QC1_parameters
FITS key |
QC1 database: table, name |
definition |
class* |
HC_plot** |
more docu |
none | kmos_flat..slit_edge_x_0 | x position of left edge of leftmost slitlet [pix], 0 degree rotator angle | HC | | [docuSys coming] |
QC.SLIT.MEAN | kmos_flat..slit_wid_mean_0 | mean width of slitlets [pix], 0 degree rotator angle | HC | | [docuSys coming] |
none | kmos_flat..ifu_active_0 | number of active IFUs per detector, 0 degree rotator angle | HC | | [docuSys coming] |
QC.BADPIX.NCONTS | kmos_flat..badpix_0 | number of bad pixels, 0 degree rotator angle | HC | | [docuSys coming] |
INS.TEMP7.VAL | kmos_flat..temp_spec1 | temperature of spectrograph 1 | ENG | | [docuSys coming] |
*Class: KPI - instrument performance; HC - instrument health; CAL - calibration quality; ENG - engineering parameter
**There might be more than one. |
Trending
The average width of each slitlet and their position on the detector is
monitored for each spectral band. The QC parameters at 0 degree rotator angle
(slit_wid_mean_0, slit_edge_x_0) are plotted versus time and scored. The
dependency from rotator angle is shown in separate plots (not scored).
The number of bad pixels is trended for all bands in a single plot. The
parameter badpix_0 is plotted versus time. Since the number of bad pixels is
dominated by the pixels between the slitlets, this parameter is chiefly useful
to detect IFUs that are not operational. Any missing IFU increases the number
of bad pixels by about 400000.
The trending of bad pixels is complemented by the number of active IFUs.
These are the IFUs that could be identified and processed during
pipeline-processing of flat fields. Data from all spectroscopic bands are
combined in one plot. The QC1 parameter for 0 degree rotator angle is trended
versus time and scored. The dependency from rotator angle is shown in a
separate plot (not scored).
Scoring&thresholds Flat-field slitlets
Changes in the slitlet positions should not be larger than +/- 1.5 pixels in
normal operations. After a warm-up of the instrument, jumps in the positions
can occur and adjustments of the thresholds are typically necessary.
The nominal width of the slitlets is 14 pixels. In practice, it is a bit
lower, typically around 13.5 pixels with some variation from detector to
detector and between the five wavelength bands.
There
is a dependency on the focus of the respective sprectrograph and changes have
been observed sometimes after interventions.
Scoring thresholds for the slitlet width have been set empirically to +/- 3 sigma
around the median (determined from data taken between 15 August and 30 November 2018).
When all IFUs are operational, upper and lower thresholds for the IFU
number are identically set to 8. If an IFU is put out of operations then the
lower threshold must be updated accordingly.
The number of bad pixels is not scored.
History
There have been several periods when KMOS was not operated with the full
number of IFUs, especially from 2013 until early-2015.
Algorithm Flat-field slitlets
slit_edge_x_<n>: Average x position of left edge of leftmost slitlet on each detector, for rotator angle 0.
slit_wid_mean_<n>: average widths of slitlets in flat-field exposure.
ifu_active_<n>: The number IFUs per spectroscopic channel/detector which have been identified
during pipeline-processing of flat fields.
badpix_<n>: The number of bad pixels includes those pixels that are
located between the slitlet images on the detector.
with <n> = 0, 1, ..., 6. <n> = 0 denotes the value for a rotator
angle of 0; <n> = 1, ..., 6 denote the values for the first, second,
etc. rotator angle position in the calibration sequence. These typically refer
to rotator angle values of -60, 0, 60, 120, 180, and 240 degrees.
Flat-field lamp performance
KMOS flats are usually taken at six different rotator angles. These fill
the QC parameters flat_eff_1, flat_eff_2, ..., flat_eff_6. The actual angles
can be found in the parameters rot_angle_1, rot_angle_2, ..., rot_angle_6. The
parameter flat_eff_0 is only filled if a measurement at 0 degree rotator angle
is present (which is usually the case for daytime calibrations).
QC1_parameters
FITS key |
QC1 database: table, name |
definition |
class* |
HC_plot** |
more docu |
QC.FLAT.EFF | kmos_flat..flat_eff_0 | brightness of lamp [e-/s], 0 degree rotator angle | CAL | | [docuSys coming] |
*Class: KPI - instrument performance; HC - instrument health; CAL - calibration quality; ENG - engineering parameter
**There might be more than one. |
Trending
The performance of the tungsten flat-field lamp is monitored for all three
channels in the YJ band. Three plots show flat_eff_0 versus time. Since the
flat efficiency does not strongly depend on rotator angle, only flat_eff_0 is
scored. Three other plots show flat_eff_1, ..., flat_eff_6 versus rotator
angle.
Scoring&thresholds Flat-field lamp performance
Thresholds have been set empirically.
History
Adjustments of the scoring thresholds have been necessary a few times after
interventions.
Algorithm Flat-field lamp performance
Pipeline calculated average flat-field flux, converted to e-/sec.
|