Image quality for WL: engineering comments

Image quality for WL: engineering comments

Image quality for WL: engineering comments Outline 1 [pp. 2-5]. Recap of discussion Neil/Paul/Dave on unobscured trade pros and cons [action on me: turn my notes into ppt] 2. [pp. 6-16] Ellipticity comparison of J-Omega v. 4c3 [Lehan working this but initial draft is here] 3. [pp. 17-22] Discussion of pointing & guiding architecture for WL [Kruk] 1 Telecon w/ Schechter and Gehrels Goal clarify the concerns on the unobscured aperture telescope alternative to Omega Context is enabling WL observations that meet the need for exquisite stability Requirements on ellipticity: Drift in ellipticity as a function of time need to be stable during an observation Change in ellipticity across the field Rms ellipticity static across the field would be ideal (ie stable in time and with field angle) [also ideally, only psf chromatic variation is diffraction scaling w/ ]] Design includes a slower PM vs. J 2 Design considerations for WL imaging Consensus is to avoid refractive cameras for WL imaging Short term AI: How many psf calibration stars can we expect in a WL

exposure [SDT] answer from quick look by Rhodes is >900 We can expect more bright stars than CCD observations, e.g. COSMOS, because of s/w to avoid saturated H2RG pixels Short term AI: document variation across the field in static intrinsic ellipticity, compare Omega to candidate uTMA design Below, pp, 6-15 PS: Hubble ellipticity varies across the field 0.1 this is certainly too much. 3 HST performance v. WFIRST Discussion of HST thermal and jitter performance vs. WFIRST expectations HST has 15 degree C axial gradient changes, unacceptable focus variability compared to WL stability requirements; HST jitter and drift are low (4 mas) and it may be challenging to be sure we will get nearly this low on a lighter, cheaper observatory. No question it can be done with enough $. [see pointing/guiding presentation below] Thermal instability of HST largely due to its low orbit and operational constraints, e.g. Earth-pointing during portions of orbit when targets out of CVZ (continuous viewing zone) go behind the earth. Also more modern construction techniques that all were demonstrated on Chandra should be used on WFIRST. Chandra thermal stability of 0.2 degree (gradient stability) is ~ 2 orders of magnitude better than the 15 degree gradient instability observed on HST. Detailed pitch on HST performance v. WFIRST expectations is available 4 Jitter considerations

Jitter may be constant across field but given that our field is much larger than others, this would need to be shown through modeling PS agrees that the imaging performance of the uTMA is a strong argument for its use (e.g. the EE50 comparison Hirata showed at the SDT3 telecon). Another consideration is the additional ellipticity uncertainty we have seen introduced by PSFs with spider diffraction. Longer term action items: SDT needs to help flow down the WL stability requirements towards engineering stability requirements Project needs to continue to update predicted stability, integrated modeling required. Project should share charts on TMA heritage with SDT [in backup of project presentation on uTMA trade space & design 4c3] 5 PSF ellipticity: a comparison of an obscured and unobscured point design for the SDT weak lensing subgroup J. P. Lehan May 6, 2011 Overview Compare obscured design to unobscured Obscured: JDEM Omega Unobscured: Option 4c3 (focal imager as similar to JDEM Omega as practical) Use direct pupil integration so we can chose image plane sampling Pupil sampling: 512x512 Image sampling: 512x512 (1.75 um spacing)

Field sampling: 3x3 [only middle point is inside perimeter, so a quick, conservative look] 0.23 arc-sec gaussian galaxy (full width 1/e max size) 7 Ellipticity metric definitions For a circular image e1=0.5, e2=0 8 Omega simulation details Spiders and cold-stop mask (Mentzell Sim 4-2011) Nominal focus (F/#) Uses nominal detector position and orientation Accounts for focal plane obliquity (14.254) 9 Option 4c3 simulation details

No spiders or cold mask Accounts for exit pupil shape Nominal focus (F/#) Uses nominal detector position and orientation Accounts for focal plane obliquity (10.924) 10 Variations with field e1 Obscured x/y -0.357 0.0 0.357 0.234 0.0 0.513353 0.509292 0.512823 0.509570 0.511992 0.509404

-0.234 0.511566 0.510974 0.510296 e1 ave = 0.5110.0015 Unobscured x/y 0.2315 0.0 -0.2315 -0.357 -1.08e-3 -3.31e-3 -4.90e-3 0.0 -5.30e-3 -2.90e-3 -2.85e-3 0.0 0.505299 0.506246 0.504514

0.459 0.501826 0.500314 0.501205 e1 ave = 0.5049.0031 e2 Obscured x/y 0.234 0.0 -0.234 -0.459 0.508732 0.507902 0.507704 Unobscured 0.357 -9.70e-3 -4.33e-3 1.39e-3 e2 ave = (-3.66 3.06)x10-3 x/y 0.2315

0.0 -0.2315 -0.459 -1.06e-4 -9.60e-5 1.34e-4 0.0 -8.60e-4 2.91e-4 6.18e-4 0.459 -5.50e-4 2.62e-4 8.87e-4 e2 ave = (-0.422 6.624)x10-4 Field in object space degrees11 summary Ellipticity of 4c3 design residuals is closer to ideal than that from J design residualsdesign design residualsresiduals excess design residualsin design residualsmetric design residualsfor design residuals4c3 design residualsfrom design residualsideal design residualsis design residualsroughly design residualshalf design residualsof design residuals that design residualsfor design residualsJ True design residualsusing design residualse,e1,e2 design residualsmetric design residualsor design residualsinvariant design residualsmetric design residuals(in design residualsbackup) 12 Extra Material follows

13 Lehan metric Motivation: SNAP metrics assume a preferred orientation in space (x and y). True for array but not nature. One number metric for ellipticity Pxx, Pyy, Px+y, Px-y all geometrically-equivalent ~ 1-(RMS deviation from RMS average 2nd moment) = 1 for perfectly circular PSF Pij is RMS spatial average 2nd moment Px+y 1 ( Pxx Pij ) 2 ( Pyy Pij ) 2 ( Px y Pij ) 2 ( Px y Pij ) 2 4 Pij 14 Omega variations with field Pxx x/y -0.357 0.0 0.357

0.234 0.546231 0.551042 0.558245 0.0 0.535538 0.542301 0.550308 -0.234 0.543251 0.548048 0.555605 Pxx, etc. moments have units of arc-sec^2 unitless Pyy x/y

-0.357 0.0 0.357 0.234 0.517816 0.523485 0.532094 0.0 0.515997 0.521932 0.529989 -.234 0.518687 0.524508 0.533185 Field in object space degrees

Pxy x/y -0.357 0.0 0.357 0.234 -1.15e-3 -5.69e-3 -1.06e-2 0.0 -3.48e-3 3.08e-3 -4.68e-3 -0.234 -5.22e-3 -3.06e-3 1.51e-3

x/y -0.357 0.0 0.357 0.234 0.991012 0.990524 0.989408 0.0 0.993101 0.992907 0.992751 -0.234 0.991206 0.992267

0.992896 ave = 0.99180.0013 15 4c3 variations with field Pxx x/y -0.459 0.0 0.459 0.2315 0.229438 0.228884 0.22884 0.0 0.228737 0.229544 0.22766 -0.2315

0.228498 0.228167 0.22827 Pyy Pxx, etc. moments have units of arc-sec^2 unitless Field in object space degrees x/y -0.459 0.0 0.459 0.2315 0.221562 0.224083 0.227175 0.0

0.221620 0.223880 0.227381 -.2315 0.221563 0.224083 0.227172 Pxy x/y -0.459 0.0 0.459 0.2315 -4.8e-4 -3.90e-4 -2.50e-4

0.0 -4.3e-5 1.32e-4 1.19e-4 -0.2315 6.04e-5 2.80e-4 4.04e-4 x/y -0.459 0.0 0.459 0.2315 0.994940 0.997249 0.999385

0.0 0.994586 0.997198 0.999911 -0.2315 0.994888 0.997216 0.999362 4c3 ave = 0.99720.0021 Omega ave = 0.99180.0013 16 Pointing Control and Knowledge Jeff Kruk 17 Pointing Knowledge - 1 Nominal S/C performance requirements:

Control: p/y: 25 mas rms/axis, roll: 1 arcsec Jitter: p/y: 40 mas rms/axis, roll: 1.6 arcsec (TBR) Knowledge: p/y: 4 mas rms/axis, roll: 300 mas(TBR) Attitude Sensor suite: FGS w/in payload Two star trackers ~perpendicular to boresight 2 arcsec accuracy Gyro: Kearfott SIRU AWN: 1mas/Hz, ARW: 36mas/Hr 18 Pointing Knowledge - FGS Outrigger SCAs on Imager focal plane Supplemented by separate guider channel for slitless spectroscopy Plate scale: 180 mas/pixel FOV per SCA: 6.12 arcmin on a side Performance at 10Hz: Noise Equivalent Angle at AB=15.5: 5-10 mas depending on filter Noise Equivalent Angle at AB=16.0: 7-18 mas depending on filter (when tracking 4 stars can track more if necessary) For accurate revisits to a field, pre-select guide stars on the ground to ensure that the same stars are used for each revisit. 19 FGS cont. Guide star density at NGP: Probability of finding N stars brighter than AB=15.5 AB=15.5

1 2 3 4 1 SCA 0.93 0.74 0.50 0.28 2 SCA 0.99 0.97 0.91 0.80 Probability of finding N stars brighter than AB=16.0: AB=16.0 1

2 3 4 1 SCA 0.97 0.88 0.72 0.50 2 SCA 0.99 0.99 0.98 0.94 AB=16.0 gives adequate performance at 10Hz. 20 Telemetry downlink It is standard practice to downlink samples of sensor data; question is

the sampling rate. Probably not worth downlinking full gyro rate, for example. Not necessarily better than the FGS data if flexible modes in the instrument are important Can downlink full 10Hz FGS GS position data Can downlink Kalman filter output at its full rate, which indirectly provides the net results of the high-rate gyro data. What knowledge is required? 21 Present Status Have begun modeling integrated S/C, payload, ACS. FEM of Omega payload and S/C incorporated into simulator Includes both fixed and articulated solar arrays, fuel slosh model At early stages in tuning control law for slew-settle studies May need to iterate on star-tracker, rate gyro selection. 22

Recently Viewed Presentations

  • Web Content Management - National Grid plc

    Web Content Management - National Grid plc

    To meet the conditions set out within Regulation EC No 715/2009, a number of European Network Codes have been developed. Once passed into EU law, the European Codes supersede Great Britain's (GB) law and therefore the Uniform Network Code (UNC)....
  • Diabetes Mellitus - Vincent's

    Diabetes Mellitus - Vincent's

    Insulin Pump. The MiniMed Paradigm insulin pump (A) delivers insulin into a cannula(B) that sits under the skin. Continuous glucose monitoring occurs through a tiny sensor (C) inserted under the skin. Sensor data are sent continuously to the insulin pump...
  • Shakespeare's Native English - Lifelong Learning Mississauga

    Shakespeare's Native English - Lifelong Learning Mississauga

    Shakespeare's English. Because the linguistic climate of Shakespeare's time was, in David Crystal's words, "one of the most lexically inventive periods in the history of the language, "word invention was simply a matter of course for early modern English poets...
  • Graph Operations - University of Pennsylvania

    Graph Operations - University of Pennsylvania

    Graphs and Hypergraphs * * * * * * * * Graph definitions A directed graph consists of zero or more nodes and zero or more edges An edge connects an origin node to a destination node The origin and...
  • Agreement Between Groups of Raters

    Agreement Between Groups of Raters

    Why Agreement? Agreement: Closeness of rating patterns by different raters. Application: Between old and new diagnostic tests. Group of assessors testing reliability of scales
  • MIKROELEKTRONIKA Dioda Semikonduktor D3  Teknik Komputer Universitas Gunadarma

    MIKROELEKTRONIKA Dioda Semikonduktor D3 Teknik Komputer Universitas Gunadarma

    Lapisan pengosongan (The depletion layer) Potensial Barier Jenis Dioda Model Dioda Model adalah representasi dari suatu komponen atau rangkaian yang memiliki satu atau lebih Sifat atau karakteristik. Tiga model diode : 1. Model Diode Ideal 2. Model Diode Praktek 3....
  • Standard Setting in High-Tech Industries

    Standard Setting in High-Tech Industries

    Buying a House in Hyde Park. Search on "purchase house Hyde Park Chicago" ... on terms other than those contained in the traditional form of listing agreement known as an Exclusive Right to Sell .
  • The Role Of ASIP In Programmable Platforms 1

    The Role Of ASIP In Programmable Platforms 1

    I can do it in 1!!! ? Initial Permutation Expansion Permutation S Boxes P Permutation Final Permutation Key Generation State Machine I'll show you how fast it can be Initial Permutation Expansion Permutation S Boxes P Permutation Final Permutation Key...