CALBLOCK-PV-018: NISP Nonlinearity¶
Observation configuration file¶
Scheduling¶
Once, during PV phase. The photo-MACC calibration (using high LED fluxes) must be taken after the spectro-MACC calibration.
In case of failure, repeat as soon as possible.
Time-critical¶
No.
Dependencies¶
NI-CU commissioned.
Objective¶
Determine the nonlinearity correction for NISP, and update the model if necessary. On ground we have tested five different flux levels of 16, 30, 70, 110, 150 e-/s, obtaining spectro-MACC ramps and saving each of the 15 group images. The ramps for the three lowest fluxes were repeated 30 times, the two highest ramps are repeated 10 times to obtain sufficient S/N. The same must be done for photo-MACC mode, with higher flux settings.
The nonlinearity model is validated by a separate visit to the self-cal field, obtaining images with two different exposure times.
Observations¶
1. Nonlinearity measurement (spectro-MACC):
Observations must be done in engineering RAW mode, taking two detectors at a time (one per DPU). This gives us access to the individual group images. All detectors must finish a given fluence setting before moving on to the next higher fluence. We use LED C, which is the most stable and was also used on ground. A LED flux stabilization period is not required, as the LED stabilization time is much shorter than the detector burn-in time, and the first ramp is not used for the analysis.
for LED C flux in [16, 30, 70] e-/s:
for detector in [1,...,16] (two at a time):
Take 30 spectro-MACC ramps in enginering RAW mode
for LED C flux in [110, 150] e-/s:
for detector in [1,...,16] (two at a time):
Take 10 spectro-MACC ramps in enginering RAW mode
2. Nonlinearity measurement (photo-MACC):
Since a photo-MACC ramp is about 5 times shorter than a spectro-RAMP, we use 5 times higher LED fluxes.
for LED C flux in [80, 150, 350] e-/s:
for detector in [1,...,16] (two at a time):
Take 30 photo-MACC ramps in enginering RAW mode
for LED C flux in [550, 750] e-/s:
for detector in [1,...,16] (two at a time):
Take 10 photo-MACC ramps in enginering RAW mode
The different levels of LED fluxes can be scheduled independently of each other, provided all detectors are completed. There are no darks between the individual ramps, to maintain the trapping / detrapping in a somewhat steady state.
3. Nonlinearity validation:
To validate the nonlinearity model (and CALPRODUCT-NI-003: NISP Brighter-fatter Effect), we obtain images in all filters of the Self-cal field, with photo-MACC(4,16,4) and non-standard photo-MACC(15,16,11) ramps. We use different filters to probe for any residual sampling effects. The dither offsets should be contained in a 2 arcmin wide box, such that no new FGS catalog is required.
for filter in [Y,J,H]:
Repeat 4x:
Random dither offset in a 3' wide box
Take 1 photo-MACC(4,16,4) image
Random dither offset in a 3' wide box
Take 1 photo-MACC(15,16,11) image
Fulfilling¶
Targets and pointing constraints¶
Pointing: Straylight suppression as per MOCDA-SYS-003.
The nonlinearity validation requires on-sky observations.
# ID |
R.A. |
Dec |
SELFCAL-FAINT |
268.51 |
65.09 |
Instrument configuration¶
Engineering Raw, for the nonlinearity measurement.
Science for the validation.
Instrument:
NI-CU on (see table below), NI-FWA closed, NI-GWA RSG000 for the nonlinearity measurement.
Uses non-standard photo-MACC(15,16,11) setting for the validation part.
# LED ‘C’ Flux [e-/s] |
I (mA) |
PWM [%] |
8 |
1.0 |
5.4 |
12 |
1.0 |
11.5 |
16 |
5.0 |
6.3 |
20 |
5.0 |
7.7 |
30 |
5.0 |
11.0 |
40 |
5.0 |
14.3 |
50 |
5.0 |
17.6 |
70 |
5.0 |
24.2 |
80 |
5.0 |
27.5 |
110 |
5.0 |
37.4 |
150 |
10.0 |
27.9 |
175 |
10.0 |
32.0 |
200 |
10.0 |
36.1 |
350 |
30.0 |
20.7 |
450 |
30.0 |
26.8 |
550 |
30.0 |
32.8 |
750 |
30.0 |
44.8 |
950 |
60.0 |
29.4 |
Methodology / Uplink procedure¶
Entity: NISP
Commanding Method: NISP: Observation; must command LED C. Uses both standard photo-MACC(4,16,4) as well as non-standard photo-MACC(15,16,11) setting.
Dither Sequence: None for the nonlinearity measurements. Random dithers within a 3’ box for the validation observations.
Inter-operability constraints¶
VS-CU off
In-orbit duration¶
In-orbit Duration: 173.9 h (7.3 days)
8*110*573 s + 170*70s = 143.4 h for spectro-MACC, allowing for 170 dither slews
8*110*111 s + 33*70s = 27.7 h for photo-MACC, allowing for 33 dither slews
3*4*(70s+111s+70s+573s) = 2.8 h for the nonlinearity validation.
10 ramps each for the two bright LED settings, and 30 ramps for the faint three faint settings, i.e. 110 ramps in total. Two detectors can be done at the same time, hence this must be repeated 8 times to get all 16 detectors.
The acquisition time for the Engineering Raw ramps is perhaps 20s shorter than the nominal acquisition times adopted here, since all group images simply have to be passed on to the MMU without processing. This could potentially save up to 10 h, but is ignored in the above estimate.
Data volume¶
Total: 283164 MB
NL ramps: 110*(NPR + NSR) = 280500 MB
Validation: 3*4*(NP + NS) = 2664 MB
Processing input¶
Ground Processing Input Data: BPM
Data Description: 33440 group images (single detector). There might be additional dark images accompanying these data.
Processing output¶
Linearity correction for each pixel. Comparison to ground characterization.
Required accuracy, constraints¶
Must be able to compare the in-flight nonlinearity correction to the ground characterization.
Processing tools¶
L1 nonlinearity pipeline, offline analysis likely
Validation¶
Update the nonlinearity coefficients if significantly different from the on-ground characterization.
Recovery¶
Repeat.
Quality control parameters¶
DataModel: https://gitlab.euclid-sgs.uk/
Contact person¶
NISP-P calibration scientist