Transcript JWST Reference Pixels and Readmodes

Reference pixels and readout modes:
What we have learned thus far
Don Figer, Bernie Rauscher, Mike Regan
March 13, 2003
JWST Radiation Environment
1
March 13, 2003
Detectors Are Important for
JWST
4
Duration of DRM NIR Observations [yrs]
1.E+02
Sunshield
Signal [e-/sec/pix]
1.E+01
JWST requirement
1.E+00
JWST goal
1.E-01
R=5
1.E-02
Zodiacal Light
1.E-03
Dark current =
0.126 e
- /sec
0.020 e
- /sec
0.003 e
- /sec
3
Spectra
Images
2
1
R=1000
0
1.E-04
0.1
1
2
10
Wavelength [mm]
JWST Radiation Environment
4
6
8
Read noise per exposure [electrons]
2
March 13, 2003
NIR Detector Characteristics
 Dark current
 Read noise
 Linearity
 Latent charge (persistence)
 Quantum efficiency (QE)
 Intra-pixel sensitivity
 Thermal stability
 Radiation immunity
JWST Radiation Environment
3
March 13, 2003
IDTL Test System
Controller Electronics
Dewar
Entrance
Window
Vacuum Hose
He Lines
JWST Radiation Environment
4
March 13, 2003
JWST MIR Detector Requirements
JWST Radiation Environment
5
March 13, 2003
Dark Current

Lowest measured dark current is ~0.005 e-/s/pixel.
JWST Radiation Environment
6
March 13, 2003
IDTL Measurements: Read Noise

Read noise is ~10 e- for Fowler-8. (system read noise is ~2.5 e-)
JWST Radiation Environment
7
March 13, 2003
Reference Pixels

All candidate JWST detectors have reference
pixels

Reference pixels are insensitive to light

In all other ways, designed to mimic a regular
light-sensitive pixel

NIR detector testing at University of
Rochester, University of Hawaii, and in the
IDTL at STScI -> reference pixels work!

Reference pixel subtraction is a standard part
of IDTL data reduction pipeline
Raytheon 2Kx2K
NIR Module
Rockwell 2Kx2K
NIR Module
Raytheon 1024x1024
MIR MUX
JWST Radiation Environment
8
March 13, 2003
Use of Reference Pixels

JWST’s NIR reference pixels are grouped in columns and rows

Most fundamentally


–
reference pixels should be read out in exactly the same manner as any “normal” pixel
–
data from many reference pixels should be averaged to avoid adding noise to data
We have begun to explore how reference pixels should be used. Approaches considered
include the following.
–
row-by-row subtraction
–
maximal averaging (average all reference pixels together and subtract the mean)
–
spatial averaging
–
temporal averaging
Spatial averaging is now a standard part of IDTL calibration pipeline
JWST Radiation Environment
9
March 13, 2003
A Picture of IDTL System Noise

Shorting resistor mounted at SCA location

1/f “tail” causes horizontal banding.

Total noise is =7 e- rms per correlated double sample.
JWST Radiation Environment
10
March 13, 2003
Averaging small numbers
of reference pixels adds noise

Averaged the last 4 columns in each row and performed row-by-row subtraction
After
Before
JWST Radiation Environment
11
March 13, 2003
Spatial Averaging

In spatial averaging, data from many
(~64 rows) of reference pixels are used
to calibrate each row in the image

A Savitzky-Golay smoothing filter is
used to fit a smooth and continuous
reference column

This reference column is subtracted
from each column in the image

Using this technique, we can remove
some 1/f noise power within individual
frames

In practice, this technique works very
well
JWST Radiation Environment
This is a standard
part of the IDTL data
calibration pipeline
12
March 13, 2003
Spatial Averaging: Before & After
Before
JWST Radiation Environment
After
13
March 13, 2003
Temporal Averaging

Dwell on the reference pixel and sample
many times before clocking next pixel

Potentially removes most 1/f

Not tried this in IDTL yet. U. Hawaii
has reported some problems with
reference pixels heating up
JWST Radiation Environment
14
March 13, 2003
Temporal Averaging: Before &
After
Before
JWST Radiation Environment
After
15
March 13, 2003
Summary of Reference Pixel
Calibration Methods

Spatial averaging works well using a
Rockwell HAWAII-1RG detector

Based on conversations with U.
Rochester, we foresee no problems
with SB-304

Temporal Averaging is promising.
More work needed using real
detectors.
JWST Radiation Environment
16
March 13, 2003
Summary

Reference pixels work and are an invaluable part of the data calibration pipeline

We have explored three techniques for using reference pixels
–
–
–
–
row-by-row subtractions,
maximal averaging,
spatial averaging, &
temporal averaging

Averaging at the end of row will not work

Spatial averaging works well and is robust

We have found:
–
–
–
–
–
–

dark current is low (~0.01 e-/s/pixel)
glow is very small
noise goes down as roughly 1/root(N) up to 8 reads (at least)
persistence is observed
JWST requirements seem realizable
saving all the data are necessary to mitigate unforeseen detector effects, such as
the non-linear bias drift after reset ("shading" in NICMOS). Note that ref pixels
do not get rid of all of the effect.
Cosmic ray rejection requires careful handling of reference pixels, output voltage
drifts, and knowledge about previous history (persistence)
JWST Radiation Environment
17
March 13, 2003
Appendix
JWST Radiation Environment
18
March 13, 2003
NIR Detector Effects NICMOS
 Dark current
 Bias drifts
 QE variations
 Amplifier glow
JWST Radiation Environment
19
March 13, 2003
NIR Detector Effects NICMOS
 Persistence
JWST Radiation Environment
20
March 13, 2003
NIR Detector Effects NICMOS
 DC bias level drift
 Ghosts
JWST Radiation Environment
21
March 13, 2003
NIR Detector Effects NICMOS
 Linearity
 Well depth
JWST Radiation Environment
22
March 13, 2003
NIR Detector Effects NICMOS
 QE
 Dark current “bump”
JWST Radiation Environment
23
March 13, 2003