John Krist, Robert Effinger, Brian Kern, Milan Mandic,

James McGuire, Dwight Moody, Patrick Morrissey,

Ilya Poberezhskiy, A.J. Riggs, Navtej Saini, Erkin Sidick,

Hong Tang, John Trauger

© 2018 California Institute of Technology, Government sponsorship acknowledged

WFIRST Coronagraph

Flight Performance Modeling

The de isio to i ple e t WFIRST ill ot e fi alized u til NASA s o pletio of the Natio al E iro e tal Poli y A t (NEPA) process. This document is being made available for information purposes only.

Observing Scenario Modeling

Goal: Simulate a realistic WFIRST coronagraph (CGI) observing sequence STOP modeling provides thermally-induced wavefront

changes

Dynamic modeling provides pointing and wavefront jitter due to reaction wheel vibrations (provided by Goddard)

Optical modeling propagates wavefront through complete representation of realistically aberrated system to produce time series of speckle fields

IFS model (provided by Goddard)

Detector noise model

2

Observing Scenario 6 Definition

3

24 hours on 61 UMa to settle thermal model

8 hours on η UMa (B3V, V=1.86) for EFC

speckle time series will start in last 2 hours of this span due to timing of jitter model

Observation sequence (repeating)

8 hours on science target, 47 UMa (G1V, V=5.04)

• 2 hours each at rolls of -13°, +13°, -13°, +13°

2 hours on reference star, η UMa

10 minute slews/rolls

Observing Scenario 6

4

Roll 1 Roll 2 Roll 1 Roll 2

Target Star

(47 UMa: G1V, V=5.0)

Reference Star

(η UMa: B3V, V=1.9)

2 h 2 h 2 h 2 h 2 h

Repeat 13 times

OS6 intended for integral field spectrograph observations. Direct imaging

observations would take only 2 – 3 iterations instead of 13.

CGI Image Generation

5

Calibration

& Post-

Processing

Scene

Timing

Orbit

Thermal Structural OpticsConvert

Wa

vefr

on

ts

Thermal

Desktop

NASTRAN SIGFIT Code V

LOWFS

Matlab

IMG Det

IFS Det

CG

Front

End

HLC

SPC

Imager

IFS

FSM

FOC

DM

PROPER

PROPER IDL &

PythonPython (IFS)

CMDS

Telescope

Exit Pupil

wfe

Conventional STOP Model

Ob

serv

ing

Sce

na

rio

Jitter

STOP = Structural/Thermal/Optical Performance

What s Old & New in Simulations

• Static optical aberrations & polarization

• Thermally-induced wavefront variations & LOWFS/C corrections

• Thermally-induced pupil shear and DM variations

• Variations in pointing & wavefront jitter due to changes in reaction wheel speeds over time

• Application of optical Model Uncertainty Factors (MUFs)

• Detector modeling (for HLC image stacks)

6

STOP Results: Focus vs Time

Observing Scenario 6

7

Short cadence & thermal stepsize

0 20 40 60 80 100 120

-300

-200

-100

0

100

200

300

Z4

(p

m R

MS

)

η UMa 47 UMa 10 min timestep & 10 s thermal timestep

Hours

STOP model time to compute = 1 week

3 4 5 6 7 8 9

10-10

10-9

10-8

10-7

Polarization: With and Without MUFs

8

λ / D

Co

ntr

ast

Log

10(c

on

tra

st)

Log

10(c

on

tra

st)

Mean Azimuthal Contrast No MUFs

With MUFs

With MUFs

No MUFs

OS6 Speckle Field Time Series

Observing Scenario 6: HLC

9

η

Red = η UMa Blue = 47 UMa 0 h 132 h

No Optical MUFs With Optical MUFs

Includes: static aberrations (surface errors & polarization), high and low order

wavefront control, thermally-induced wavefront aberration & pupil position

changes, deformable mirror thermal drift, pointing & wavefront jitter, stellar

diameters & colors

Log

10(c

on

tra

st)

Field incident on detector shown. Detector effects not included

10% bandpass

λc = 575 nm

Movie: View in

slideshow mode

Contrast vs Time: with Optical MUFs

Observing Scenario 6: HLC

10

0 50 100 150

2´10-9

4´10-9

6´10-9

8´10-9

Hours

Mea

n C

ont

rast

r = 3 – 4 λ/D

r = 4 – 8 λ/D

OS6 Speckle Field Time Series

Observing Scenario 6: SPC

11

Field incident on detector shown. Detector effects not included

This and HLC OS6 simulated time series data available at

wfirst.ipac.caltech.edu

η

Red = η UMa Blue = 47 UMa 0 h 132 h

Log

10(c

on

tra

st)

No Optical MUFs With Optical MUFs

18% bandpass

λc = 760 nm

Movie: View in

slideshow mode

EMCCD Model

Observing Scenario 6: With Optical MUFs

12

Reference

star

(V = 1.9)

Target

star

(V = 5.0)

Target Star

Minutes total

exposure time

(1st 2 hours)

120 sec

Co-added images

26 hours (ref)

48 hours (target)

Qui k dirty Co-add + CR rejection

Ref = 3 sec/frame

Target = 30 sec/frame

Movie: View in

slideshow mode

Note: Detector modeling software does not currently

handle cosmic rays in short (3 sec) exposures properly

EMCCD Model

Observing Scenario 6: No Optical MUFs

13

120 sec

Qui k dirty Co-add + CR rejection

Reference

star

(V = 1.9)

Target

star

(V = 5.0)

Target Star

Minutes total

exposure time

(1st 3 hours)

Ref = 3 sec/frame

Target = 30 sec/frame

Co-added images

26 hours (ref)

48 hours (target)

Movie: View in

slideshow mode

Note: Detector modeling software does not currently

handle cosmic rays in short (3 sec) exposures properly

HLC OS6 Example Post-Processing

Observing Scenario 6: With EMCCD

14

Target Star Roll 1

- Reference Star

Target Star Roll 2

- Reference Star

Angular Differential

Image (Target only)

No optical

MUFs

With optical

MUFs

1 x 10-8

3.5 λ/D

3 x 10-9

4.5 λ/D

15

Azi

mu

thal

RM

S C

on

trast

λ / D

Azi

mu

thal

RM

S C

on

trast

λ / D

No Optical MUFs With Optical MUFs

Before subtraction

RDI

ADI

HLC OS6 Example Post-Processing

Observing Scenario 6: With EMCCD

4x 7x

Phase B Modeling Activities

• Implement Phase B thermal/structural model, including detailed CGI model

• Apply new MUFs that may be imposed

• Incorporate new knowledge of DM behavior

• Run more observing scenarios, including investigating the effects of internal CGI heat sources (electronics) and DM response to thermal changes

• OS6 simulations available from wfirst.ipac.caltech.edu

16