ESP Status. Calibration,
Degradation, Validation
Leonid Didkovsky, Darrell Judge,
and Seth Wieman
USC Space Sciences Center
ESP Status (1)
• Fully operational, correctable, and predictable. No
degradation, except for the Al filters is detected. We do:
• Regular tests of ESP variables (changes of dark count-rates,
gain, visible light, filter degradation) based on daily and
weekly calibrations.
• Prompt data reduction and submitting EVE internal Reports,
e.g. related to the Filter Scans, FOV maps, Cruciform maps,
etc. These reports allow us updating ESP Level1 program
for calculation of absolute irradiances with better accuracy,
to calculate locations of the flares, etc.
• Regular workshops to discuss current and future works (the
last one was on Aug 23-24) @ LASP, including technical
issues and science with ESP.
Oct 6, 2010
LASP SWx Workshop
2
ESP Status (2)
• ESPR calibration @SURF BL-2 in Sep was
successful. The results will be ready upon receiving
SURF fluxes. A fast look shows very small changes
compared to the previous ESPR calibration.
• ESP QD data were used as near-real time proxies
for GOES XRS-A,B with small latency in Level-0CS.
• Level-0CS updated to get ESP irradiance-like data
(all channels) with small latency.
• ESP Ch9 (30.4 nm) channel is validated as a proxy
for SEM 30.4 nm channel.
• We improve ESP Level1 Program. Current work is
to implement filter degradation ‘live’ updates from
daily/weekly calibrations.
Oct 6, 2010
LASP SWx Workshop
3
How Filter Degradation is
Determined
• We use ESP daily and weekly calibrations to
calculate ratios between effective counts
measured with different filters.
• Time difference between two filter
measurements is as small as one minute to
provide smallest signal fluctuations related to
possible changes of the solar EUV flux.
• We work on a new algorithm, which can provide
accurate measurements of the ratios for the
calibration performed during rapid changes of
the solar irradiance, e.g. after a solar flare*.
* Time for each calibration is fixed, thus, we may catch some flare(s).
Oct 6, 2010
LASP SWx Workshop
4
What ESP Tests During
Daily/Weekly Calibrations
• ESP provides absolute solar irradiance in different (5) spectral
bands centered at 3.6, 19.0, 25.4, 29.9, and 36.4 nm.
• As an advanced version of SOHO/SEM, ESP electronics has more
test options than SEM electronics, such as
• Test / correction for the electronics gain, including changes of the
electronics related to the TID;
• Test / correction for the thermal / shunt_resistance changes of the
detectors;
• Test / correction for the visible light / filter_pinholes;
• A filter wheel with three science aluminum filters, one prime (F3) and
two spare (F2 and F4). This option allows us to measure a filter
degradation as a ratio of effective counts between the tested and
reference filters
Oct 6, 2010
LASP SWx Workshop
5
ESP Calibrations
ESP has a Filter Wheel with 5 filters: Dark, three Al
filters (F2, F3, F4) and one Fused Silica Filter (F5).
• Before DOY = 249 (Sep 5):
Filter positions for daily: 3, 4, 5, 1, 3: Ratio 4/3
Filter positions for weekly: 3, 1, 2, 3: Ratio 2/3
• After/on DOY = 249:
Filter positions for daily: 3, 4, 1, 3: Ratio 4/3
Filter positions for weekly: 3, 4, 5, 1, 2, 3: Ratios
4/3 and 2/3.
Oct 6, 2010
LASP SWx Workshop
6
ESP Filter 3 (prim sci) Degradation
ESP F4/F3 Ratios
1.4
1.35
y = -0.0000026x2 + 0.0030624x + 0.7109408
2
R = 0.9995215
1.3
Ch9
Ch8
Ch2
QD
Poly. (Ch9)
Poly. (QD)
Ratio
1.25
1.2
1.15
1.1
1.05
1
0.95
120
140
160
180
200
220
240
260
280
Time 2010, DOY
Degradation of ESP Al filter is due to deposition of hydrocarbon, which reduces the
transmission of the filter. The change of transmission is wavelength dependent.
Predicted Ch9 ratio for DOY=365 is 1.45 0.03
Oct 6, 2010
LASP SWx Workshop
7
A Comparison of a Henke-Model and Data
ESP F3 Degradation With a CH Deposition
1.05
1
Transmission
0.95
Henke CH_30A
ESP_DOY146
Henke CH_40A
ESP_DOY165
Henke CH_60A
ESP_DOY207
0.9
0.85
0.8
0.75
0.7
0.65
0
5
10
15
20
25
30
35
40
Wavelength, nm
Oct 6, 2010
LASP SWx Workshop
8
Thickness of F3 CH Layer: a Model
Based on Measurements
Thickness of ESP F3 layer of CH
2
y = -0.000957x + 0.838783x - 72.532824
Thickness, A
100
80
60
ESP F3 CH layer
Poly. (ESP F3 CH layer)
40
20
0
100
150
200
250
300
350
Time, DOY
Predicted thickness of CH layer for DOY=365 is about 107 A
Oct 6, 2010
LASP SWx Workshop
9
Another Evidence of F3 Degradation
ESP F2/F3 Ratio
1.4
1.35
1.3
2
y = -0.0000013x + 0.0024398x + 0.7603834
R2 = 0.9993904
Ch9
Ch8
Ch2
QD
Poly. (Ch9)
Poly. (QD)
Ratio
1.25
1.2
1.15
1.1
1.05
1
0.95
120
140
160
180
200
220
240
260
280
Time 2010, DOY
Data from weekly calibrations (using F2) show F2/F3 ratios similar to
F4/F3 ratios. However, they are smaller than F4/F3 ratios.
Oct 6, 2010
LASP SWx Workshop
10
An Evidence of F2 Small Degradation
y = -2E-05x + 0.9919
ESP F2/F4 Ratio
y = -3E-05x + 0.9912
y = -5E-05x + 0.9948
Ratio
0.995
Ch9
Ch8
Ch2
Linear (Ch8)
Linear (Ch9)
Linear (Ch2)
0.985
0.7%
0.975
120
140
160
180
200
220
240
260
280
Time 2010, DOY
These slopes for the first-order ratios show some degradation of ‘reference’ F2, which
was exposed to the Sun for 1 min/week (compared to 6 min/week for F4)
Oct 6, 2010
LASP SWx Workshop
11
ESP Validation: SEM / ESP (30.4 nm) Comparison
That Revealed a Filter Degradation Issue
SEM / ESP Ratio
2
y = 0.000000190x + 0.001909746x + 0.779555980
1.40
1.35
The change of
SEM/ESP ratio
1.30
Ratio
1.25
SEM/ESP
Poly. (SEM/ESP)
1.20
1.15
1.10
1.05
1.00
120
130
140
150
160
170
180
190
200
210
220
230
240
2010, DOY
Oct 6, 2010
LASP SWx Workshop
250
260
270
was discussed
at the EVE/AIA
Workshop at
Stanford and at
COSPAR (2010).
We pointed out
that one of
explanations of
this change is
ESP degradation
12
ESP Validation: Ch9 as SEM Proxy
SEM and ESP
8.0E-04
Irradiance, W/m^2
7.5E-04
7.0E-04
SEM
ESP
36.258
ESPcorr 4ratio
ESPcorr 4degrad
6.5E-04
6.0E-04
5.5E-04
5.0E-04
120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270
2010, DOY
Black: ESP corrected using a fit to the SEM/ESP ratio; Red: corrected using
a fit to the filter degradation curve.
ESP Validation: Ch9 Provides Accurate ( 2.5%) SEM Proxy!
ESP/SEM Irradiance Ratio Change
Ratio Difference, %
y = 0.001x + 3.7146
9
7
5
ESP/SEM ratio change
3
Linear (ESP/SEM ratio
change)
1
-1
-3
120
140
160
180
200
220
240
260
Time 2010, DOY
Practically horizontal linear fit (no trend) shows that both SEM and ESP
filter degradations were calculated correctly and accurately. STD
deviations around the trend line are about 2.5%.
Oct 6, 2010
LASP SWx Workshop
14
Acknowledgments
• This work was supported by the University
of Colorado award 153-5979
Oct 6, 2010
LASP SWx Workshop
15
Backup Slide: GOES Proxies
XRS − A(0.05 − 0.4nm) = 10[
−18.94 +3.606•log10 (ESP0−7nm )]
XRS − B(0.1− 0.8nm) = 10[
−11.86+2.259•log10 (ESP0−7nm )]
Oct 6, 2010
LASP SWx Workshop
16