Transcript Review

IRTF Adaptive Optics System
Review
Overview
• IRTF is building a 36 element curvature based AO system
• Purpose of this Review - Is to identify any remaining design,
installation or operational problems and to expose the IRTF staff to
the system details
• Present Status – Design 90% complete, construction 70%
Overview
AO System Block Diagram
AO Relay
AO Wavefront Sensor(WFS)
Target Science
•
Planetary Science
Jupiter and it’s satellites (using the satellite to guide)
Saturn and it’s satellites (using the satellite to guide)
Neptune (using the disk to guide)
Uranus (using the disk to guide)
Mars (with to-be designed wide field WFS)
•
Mission support
Cassini(saturn), Galileo (Jupiter), Mars Global Surveyor,
Mars 2001
Non-planetary Science projects
•Searches for companions to nearby stars (BDs and planets)
•Astrometry of companions for second epoch confirmation
•AO Spectroscopy for spectral typing R~1,000
•Large telescope follow up for Radial Velocities R~10,000
•YSO disks and companions
•Imaging of disks and companions
•Companion spectroscopy
•Imaging (JHKL) of young star clusters
•high resolution imaging for luminosity and mass functions,
disk lifetime studies
•Seyfert galaxies
•AO spectroscopy of nuclear regions
•Quasar hosts
•JHK imaging of underlying host galaxy
Expected Image Quality Performance
Filter
Wavelength
(microns)
Lambda/D
Expected
Strehl
J
H
1.25
1.6
0.086
0.112
50%
60%
K
L
M
2.2
3.8
4.6
0.151
0.259
0.315
70%
85%
95%
Estimated Performance (continued)
Estimated Emissivity
Addition of 6 mirrors – 5 Silver(1%), one Aluminum(2.2%)
Wavelength
Telescope
Emissivity
Sky
Emissivity
Telescope+
Sky
AO
Emissivity
Telescope+
Sky+AO
L(3.77)
7%
2.5%
9.5%
7.2%
16.7%
M(4.58)
7%
10%
17%
7.2%
24.2%
Estimated Performance(continued)
Sensitivity Limits
•Full AO correction, average to good night, within 30 degrees
of zenith is expected to 12-13th visual magnitude
•Partial AO correction out to 15th magnitude with a gradual
decrease in performance.
•Tip/tilt only correction to 16th
Top Level Requirements
•A design that is optimized for planetary observations
•Operation with NSFCAM(two plate scales) and SPEX
•Optics that allow easily removing the AO optics from the beam
•Support of differential track guide/science objects
•A wavefront sensor field of view that allows tracking directly on Neptune and
Uranus.
•A wavefront sensor that bolts on the instrument and is fed via the science
instrument cold dichroic
•Operation optimized for J, H, K wavelengths with low enough emissivity to
work at L and M.
•Pass a corrected science field of 80x80 square
Desirable Features
• Easy changing between platescales on NSFCAM
• WFS optics that allow IR wavefront sensors in the
future.
• Guide camera w/visible Photometry capability
Optical Design
IRTF AO Relay Optical Layout
Optical Design
Wavefront Sensor
Optical Performance
Relay performance ~96% Strehl over the 80 arcsecond field
WFS performance ~95% on axis to ~80% in the corners at
visible wavelengths
Relay optics specified at Lambda/20 in the visible to reduce
scatter to below the atmospheric scatter.
AO Components
• Dan to add
AO SYSTEM
NASA IRTF
MECHANICAL
SECTION
Vern Stahlberger
Mechanical Tasks:
1) RELAY
2) WAVE FRONT SENSOR
3) APD RACK MOUNT
4) SPOOL
5) AO AND SpeX (Interface)
6) AO AND NSFCAM (Interface)
1) RELAY
I: DESIGN DRIVERSA) Tolerances: Position Optical Elements to 0.001 inch
Tilt Tolerance: 0.016 degrees (~ 54’)
B) Must be able to assemble Relay with MIM in place
C) Kinematic Mounts for Optical Elements
A) Tolerances: HOW TO ACCOMPLISH!
General Rules: 1)Design for absolute positioning (no adjmts)
2) Minimize interfaces => (more complex prts)
• Limit Structure Deflection: FEA done;
shows < 0.0001 inch max. deflection
• Positional accuracy: Precision machined optical
Mounts with doweled interfaces to Relay.Detail
drawings are built with tolerances to meet these
requirements. Sub-sections of Relay precision
machined and doweled.
• Tilt tolerance of 0.016 degree (~ one minute)
The tolerances on the optics mounts are well
B) Ability to assembly above MIM ,Relay is 8ft+long OA
RELAY SUB-SECTIONS
C) Kinematic Mounts for Optical elements (Flat2)
II: Construction:
1) Relay made up from 3 subsections;
all bolted assemblies.
2) Made from Plates Alu Sup K100
(used for Stability and Flatness)
3) Weight: ~ 260 lbs
4) Overall Dimensions: 101 x 21 x 12 inches
III: Relay Optical Subassemblies
1) Flat1&3Mount (6 “ translation, in-out of beam)
2) Flat2 Mount (Fixed)
3) Fiber Optics Guide (2” translation, in-out of beam)
4) OAP1 (Fixed)
5) DM (Fixed)
6) OAP2 (Fixed)
Optical Elements supported by Relay Structure
DM
Flat2
OAP
TO DEWAR
OAP
Flat1
Flat3
Relay Optical Subassemblies
DM
DM
mount
Fold2 Mount
OAP
Fold1&3Mount
OAP
Interfaces: doweled/bolted
DM-Subsection of Relay
Relay: Fold1&Fold3 Assembly
Relay assembled except some optics mounts
2) WAVE FRONT SENSOR
I: Design DriversGeneral Rules: Absolute positioning
Minimize interfaces
A) Positioning Tolerances for Optics: 0.001 inch
B) Optics Bench: Flatness over surface <.0025
C) Assemble WFS to both NSFCAM and SpeX
D) Kinematics Mounts for Optical Elements
E) Light Tight Enclosure (1 bend)
F) Cable Feed Through ( Fiber +)
A) Positioning Tolerance:how to accomplish
• Precision machined Optics Mounts: GD&T
is used for specifying important dimensions
and geometric relationships on individual detail
drawings.
• Each Optics Mount is individually doweled to
the Optics Bench. Bench is toleranced to meet
specs
Example of kinematic mounting
Assembly Drawing for Steering Mirror:
Example of Kinematic Mount
B) Optics Bench: (Features)
• Rigidity: Primary concern (to prevent relative
platform motion)
• Guaranteed flatness by Vendor < 0.0025 inch
• Lightweight Structure: 47 lbs
• Max. Static Deflection 0.001 inch/40lbs
load applied at center.
• Dimensions: 44 1/2 x 15 x 2 inch
Custom Optics Bench with Ray-trace
Lenslette array
Membrane Mirror
Dichroic
Newport Optics Bench Design:
TRUSSED CORE DESIGN:Extra Steel Member through center of
cell significantly stiffens cell with little increase in weight.
C) Assemble to both NSFCAM and SpeX
• WFS will bolt to Vacuum Jacket of either
NSFCAM or SpeX
• NSFCAM will have two focal positions:
WFS will be shifted by 2.75 inches
when changing plate scales.
Note: Apd Mount will not move when
changing Plate-scales on NSFCAM.
Tie brackets are used to bolt Apd Mount and
WFS together only when changing
instruments
D) Kinematic Mount for Optical Elements
E) Light tight Enclosure:Access covers
WFS Optical Subassemblies
3) APD MOUNT
I: Design DriversA) Cool Apd’s + Thermal Managment
B) Need Access to Apd’s for Service/Replacement
C) Need Wire feedthroughs for Fiber and Coax’s
A) Cool Apd’s
Heat dissipation per Apd = 2.6 Watts
For 36 Apd’s ~ 100 W
Set: In calculator- (next slide..)
Apd Steady State working Temp. = 25 degrees C
Fluid Stream Velocity = 1 m/s
Fluid Free Stream Temp. (Dome) = 5 degrees C
Fluid (Air) density=1.252 kg/m^3
Calculated: Heat Transfer Rate to Air: Q
with the purchased heat sinks
For 2 large heat-sinks =171.6 W +
for 2 small heat sinks =85.8 W
Total Q=257.4 W (Q is proportional to Area of heat sink)
Heat Sinks: Two sizes
Heat Transfer Rate to Air: Large sink
However….
Air Properties at Sea level
Air Properties at 5000 m
Density at 4000m
Recalc.Heat Transfer Rate to Air: Fluid density at 50%
from previous calculation
27% Reduction
We can expect:
Heat transfer Rate to Air Q: on
Summit
2 * 62.3
+
2* 31.1
= 187 W
B) Access to Apd’s (Service/Replacement)
Basic Heat Exchanger Core with mounted APD’s
Heat exchanger core with G-10 insulation
Thermal Insulation from Rack;
Use of Sil-pads under Apd’s
Sil-Pads
Sil-Pads were developed to eliminate
the need for thermal grease. They consist on an elastomeric binder compounded with a thermally conductive
filler coated on a carrier.
A typical filler material used is
alumina.
On application, the Sil-Pads are
meant to flow under pressure.
C) Wire feedthrough for Fiber/Coax Cables
Green part is neoprene, slits not shown (Coax shown)
4) SPOOL: STARTING CONCEPT
SPOOL: NEEDS FEA
5) AO AND SPEX
Modifications to SpeX:
•Vacuum Jacket: no changes
•Calibration Box: will need
3 tapped holes
SpeX Vacuum Jacket and Cal-box with Ao
Spex
CalBox
Tap 3 holes
Vacuum
Jacket
WFS
Apd Rack Mount
6) AO AND NSFCAM
1:1 PLATE-SCALE SHOWN
Modifications to NSFCAM:
• Design is shown for NSFCAM Upgrade:
• Interface Box needs to be redone
• Bottom of instrument needs an extension
•Note: The AO System can be fitted to
•NSFCAM as it exists presently. Holes
•will need to be drilled into the VJ.
NSFCAM: 1:1 PLATE-SCALE
NSFCAM: 3:1 PLATE-SCALE
THK Mechanism to Change Plate-scales
Note: Telescope is at Zenith
Rails bolt
to NSFCAM
Interface
AO ON NSFCAM w/WFS +APD Mount
AO AND NSFCAM: 3:1 PLATESCALE
Shim
for
Adjustm.
Procedure to change Plate Scales for NSFCAM:
Notes: a) When using NSFCAM, the quick release
ball lock pin should always be in place!
b) The Apd Rack Mount will NOT move with the
WFS box when changing plate scales, except
possibly be moved out on the drawer slides for access.
1) Telescope is at Zenith
2) Loosen 8 Captive Screws (1/4-20)
(sliding the Apd Rack mount out will make it easier
to get at the bolts and pivot brackets)
3) Move the WFS box AWAY from the VJ of NSFCAM.
4) Rotate the pivot brackets in or out depending on Plate scale to use
5) Make sure the field adjusted spacers are in place
6) Bring WFS box in slowly, the tapered pins will help to guide it
into proper position.
7) Tighten 8 Captive Screws (1/4-20)
Status:Mechanical (% complete)
• Relay: 90% (OAP Optics Mounts not done)
• WFS: 90% (some detailing left,Fab started)
• Apd Rack Mount: 100% complete
• Ao to SpeX interface: 75% (detailing only)
• Ao Spool: 10% (FEA required)
• Ao to NSFCAM interface: 75% (detailing only)
• Offset Guider/On-axis camera: not started
• Handler on Telescope (10%)
AO Electronics
Peter Onaka
Major Subsystems
• Wavefront Sensor Assy = Deformable piezo mirror, optomechanical subsystems + lenselette fiber feed
• APD Mount Assy = Fiber in/Counts out, EG&G
Avalanche Photodiodes, cooled chassis
• AO Crate = APD counters + mirror high voltage
amplifiers
• AOPC = AO calculation CPU (Real Time Linux PC)
• Motion Control Box = opto-mechanisms in WFS
• Vendor support electronics = tiptilt drive, X,Y,Z stage
IRTF NETWORK
TELESCOPE
CONTROL ROOM
AOPC
PCPS
TERMINAL SERVER
ETHERNET
AO CRATE
MOTION CONTROL
PCI-DIO-32HS
NAT. INST.
+12V
RACK PS
X Y Z STAGE
-12V
SH6868-D1
+5V
AOPC
TX
MULTI-FUNCTION
BOARD
RX
FIBER OPTIC 62.5
J1
10 METERS
RX
FIBER OPTIC 62.5
MEMBRANE MIRROR
APERATURE CONTROL
AO CRATE
MULTI-FUNCTION
BOARD/INTERFACE
TX
PICK OFFMIRROR
TIP/TILT
N D FILTER
MASTER CLOCK
+12V
OUT OPTIMA
IN
AUDIO AMP
25W X2
FIBER CALIBRATION STAGE
SINE WAVE PHASE CONTROL
GATE GENERATOR
BULKHEAD
24 CHNL COUNTER #1
BULKHEAD
P2
P2 -C1
BNC X36
24 CHNL COUNTER #2
P2 P2 -C2
TO MECHANISMS
12 CH HV AMP #1
P2
SOLA-SDN
5-24-100
24V 5A
12 CH HV AMP #2
ULTRAVOLT
+ /- 500V PS
#1/2 24-NP125W
IN
BNC X3
P2 P2 -HVA2
12 CH HV AMP #3
PI TIP/TILT AMP
19" RACK
P2 -HVA1
P2 P2 -HVA3
OUT
LEMO X3
S P
HVA-DMCABLE 20 FT.
PI-TIP/TILT CABLE 20 FT.
S
LEMO X3
W AVEFRONT FRONT SENSOR
MEMBRANE
APCONTROL
MEMBRANE
MIRROR
LENSLET
BNC X36
APD'S
P
ADAPTIVE OPTICS
RELAY
66-PIN
DM
ND FILTER
X,Y,Z STAGE
3MFIBER X36
AC/DC
5V 50A
SUPPLY
FIBER CALIBRATOR
PICK-OFFMIRROR
SCIENCE INSTRUMENT
APD-COUNTER CABLE
IRTF ADAPTIVE OPTICS BLOCK DIAGRAM
Wavefront Sensor Assy Electronics
•
Membrane Mirror: speaker driver
•
Lenselette: 36 SC terminated fiber
cables
•
•
•
XYZ stage: Ball Aerospace
Filter Wheel: vendor supplied
Inputs: Photons from telescope, filter
wheel control signals, X/Y/Z stage
control, membrane mirror drive
Outputs: 36 fiber cables
•
WAVEFRONT FRONT SENSOR
MEMBRANE
APCONTROL
MEMBRANE
MIRROR
LENSLET
3M FIBER X36
ND FILTER
X ,Y ,Z S T A GE
APD mount
•
•
•
•
Cooled housing for 36 EG&G Avalanche
Photo-Diode modules
Inputs : 36 fiber cables from lenselette array,
power for APDs and fans.
Outputs: 36 APD outputs, temp sensors,
thermal cutoff switches.
Power Dissipation ~100W nominal
BNC X36
APD'S
AC/DC
5V 50A
SUPPLY
3M FIBER X36
AO Crate
•
•
•
•
7U Eurocard Chassis (not a VME
computer system)
Control boards
Power supplies
Amplifier for membrane mirror
AO CRATE
+12V
RACK PS
-12V
+5V
J1
RX
TX
AO CRATE
MULTI-FUNCTION
BOARD/INTERFACE
TIP/TILT
+12V
OUT OPTIMA
IN
AUDIO AMP
25W X2
MASTER CLOCK
SINE WAVE PHASE CONTROL
GATE GENERATOR
24 CHNL COUNTER #1
P2 P2 -C1
24 CHNL COUNTER #2
P2 P2 -C2
12 CH HV AMP #1
SOLA-SDN
5-24-100
24V 5A
ULT RA VOLT
+ /- 500V P S
#1/2 24-NP 125W
12 CH HV AMP #2
12 CH HV AMP #3
P2 P2 -HVA1
P2 P2 -HVA2
P2 P2 -HVA3
PI-TIP/TILT CABLE 20 FT.
S P
AO Crate Boards
•
“remote” Multifunction board
–
–
–
•
24 Channel Counter boards x 2
–
–
•
Fiber interface
Sinephase Generator = system clock
Backplane bus interface
Inputs: APD signals, backplane bus
Outputs: backplane bus
12 Channel High Voltage Amps x 3
–
–
Inputs: backplane bus
Outputs: HV drive to Deformable
Mirror
AO Crate I/O
•
•
I/O
Inputs
–
–
–
•
36 Coax cables from
APDs
Fiber from PC MFB
AC power X 2
Outputs
–
–
–
HV output cable to DM
3 channels of analog
drive to PI piezo
amplifiers
Output cable to
membrane mirror
AO Crate Power Supplies
•
Power supplies
–
–
•
+/- 500V Ultravolt High Voltage Power
Supply.
Sola +24V supply for Ultravolt
Amplifier for membrane mirror
–
Basic audio car amp
AOPC
AOPC
•
Real Time Linux PC
–
•
•
Read in APD counts/phase
National Instruments DIO fast parallel
PCI interface board
“local” Multifunction board
PCPS
PCI-DIO-32HS
NAT. INST.
SH6868-D1
AOPC
TX
MULTI-FUNCTION
BOARD
RX
Motion Control Box
Form factor TBD
• Ball Aerospace control electronics
• Filter Wheel control
• Physik Instrumente Tiptilt piezo
amplifiers (may be separate assy)
• Other mechanisms
MOTION CONTROL
X Y Z STAGE
MEMBRANE MIRROR
APERATURE CONTROL
PICK OFF MIRROR
N D FILTER
FIBER CALIBRATION STAGE
BULKHEAD
Software
Tony Denault
Don’t blame me I am new
1. Software Plan A
Zyoptics will give us a copy of their AO
software.. Compile & Run!
•
•
•
•
Received a port of Hokupa’a from VXWorks/RTLinux (40 % completed).
Just enough to sample sensor data & check timing. (Does not output to
D.Mirror).
All of the higher level function also commented out.
Major issues not addressed:
– LP_SHMEM – pointers problem.
– Semaphore – resource locking/protections.
•
Zyoptics software was configured to their hardware setup – slightly difference
from IRTF hardware.
2. Plan B – “Oh no….. what’s plan B?”
Zyoptics software can be used as raw material for IRTF
system.
Implement similar structure in RTLinux.
to
Reverse engineer & import code from zyoptics system
IRTF system.
Implement additional IRTF requirements.
3. The Guider (Porky)
Detailed Design to be done later. Currently we just have a
rough outline of hardware requirements.
On/offset Acquisition
/Guide
Camera System
Guider Layout
Schedule to Completion
Changes Required for AO Installation
• Relocation of Electronics
• Replacement of On and Off-axis Guiders
• Replacement of NSFCAM’s mount
Changes Desired for AO Installation
• Control of Dome Thermal Environment
• Removal of Electronics Heat
• Control of Astigmatism with the Mirror Bender
Questions