Atacama Large Millimeter Array Update

January 30 2006 ALMA NA Cost/Management Review 1

Atacama Large Millimeter Array

Update

Slides Unabashedly Stolen byAl Wootten

NA ALMA Project Scientist

FromALMA NA Cost/Management Review

January 30 – February 1 2006


The ALMA Partnership

• ALMA is a global partnership in astronomy to deliver a truly transformational instrument

– North America (US, Canada; Taiwan in process)– Europe (via ESO with Spain)– Japan (now including Taiwan)

• Key Science goals include– Image protoplanetary disks, to study their physical, chemical, and magnetic-field

structures, and to detect tidal gaps created by planets undergoing formation in the disks;

– image starburst galaxies as early as z = 10; – image normal galaxies like the Milky Way out to z = 3

• Located on the Chajnantor plain of the Chilean Andes 16500’ above sea level

• The way ALMA is being built is via a 50:50 partnership between NA & Europe and a closely coordinated but separate effort from Japan

• ALMA will be Operated as a single Observatory with scientific access via regional centers

– North American ALMA Science Center (NAASC) is here


What is ALMA?

• Up to 64 12m antennas– Plus the Compact Array of 4 x 12m and 12 x 7m antennas from Japan

• Baselines from 15m to 15km• 5000m site in Atacama desert• Receivers: low-noise, wide-band (8GHz), dual-polarisation, SSB• Digital correlator, >=8192 spectral channels, 4 Stokes• Sensitive, precision imaging between 30 and 950 GHz

– 350 GHz continuum sensitivity: about 1.4mJy in one second– Angular resolution will reach ~40 mas at 100 GHz (5mas at 900GHz)– First light system has 6 bands: 100, 230, 345 and 650GHz – Japan will provide 140, 460 and 900GHz

• 10-100 times more sensitive and 10-100 times better angular resolution compared to current mm/submm telescopes


Where is ALMA?

El llano de Chajnantor

ALMA


Chajnantor

AOS TB

Toco

Chajnantor

Negro

MacónHonar

Road

43km=27 miles

Chascón

OSF


AOS TB Center of Array

Pampa La Bola

V. LicancaburCº Chajnantor

Chajnantor

Cº ChascónCº Toco


OSF Facilities ALMA and Contractors Camps

ALMA Camp

Contractors Lay-down area

Contractors Camp


OSF Facilities ALMA and Contractors Camps

Contractors Dormitories

Contractors recreation

room

Contractors offices

ALMA camp

Water tanks

Contractors kitchen and dining room


Dormitories at ALMA Camp

Recent Camp Development


Recent Camp Development

Contractors Camp dining room


APEX - The Atacama Pathfinder Experiment

A Vertex RSI Antenna Operating at Chajnantor

APEX - The Atacama Pathfinder Experiment

A Vertex RSI Antenna Operating at Chajnantor

R.GüstenBonn 21.10.05


Access Road

AOS Technical Building

AOS Facilities

85 % complete


AOS Technical Building


AOS TB Construction (1)General view, January 2006


AOS TB Construction (2)


Vertex SEF grading


ALMA Status

• ALMA has just undergone a major rebaselining and subsequent review• The review declared the technology readiness of ALMA very high and

judged that most technical risk has been eliminated• Five years ago ALMA was a "must do" scientifically but with high

technical risk pushing the state of the art • We now have:

– prototype antennas that meet ALMA’s demanding requirements– receivers with near quantum-limited performance, unprecedented bandwidth

and no mechanical tuning– the first quadrant of the correlator completed below cost and with enhanced

performance– The baseline includes appropriate contingency for remaining technical risks

(e.g. photonic local oscillator, highest frequency cold multipliers)


Front End Key Specifications(and Preliminary Results)

ALMABand

Frequency Range

Receiver noise temperatureMixing

schemeReceiver

technologyResponsibleTRx over 80% of

the RF bandTRx at any RF

frequency

1 31.3 – 45 GHz 17 K 28 K USB HEMT Not assigned

2 67 – 90 GHz 30 K 50 K LSB HEMT Not assigned

3 84 – 116 GHz 37 K (35K) 62 K (50K) 2SB SIS HIA

4 125 – 169 GHz 51 K 85 K 2SB SIS NAOJ

5 163 - 211 GHz 65 K 108 K 2SB SIS 6 units EU ?

6 211 – 275 GHz 83 K (40K) 138 K (60K) 2SB SIS NRAO

7 275 – 373 GHz* 147 K (80K) 221 K (90K) 2SB SIS IRAM

8 385 – 500 GHz 98 K 147 K DSB SIS NAOJ

9 602 – 720 GHz 175 K (120K) 263 K (150K) DSB SIS SRON

10 787 – 950 GHz 230 K 345 K DSB SIS NAOJ ?

•Dual, linear polarization channels:•Increased sensitivity•Measurement of 4 Stokes parameters

•183 GHz water vapor radiometer:•Used for atmospheric path length correction

* - between 370 – 373 GHz Trx is less then 300 K


ObservationPreparation

Scheduling

Data ReductionPipeline

Archive

Executive

ALMA Common Software

PrincipalInvestigator

1. Create observing project

2. Store observingproject

3. Get project

definition

4. Dispatch scheduling block id

6. Start data reduction

8. Notify PI

7.1. Get raw data & meta-data

7.2. Store science results

9. Get projectdata

ArchiveResearcher

TelescopeOperator

f. Get science data

d. Notifyof

Special

Condition

e. Start

StopConfigure

c. Alter Schedule / Override action

Control System

Correlator

Calibration Pipeline

Quick Look Pipeline

5. Execute scheduling block

5.2 Setup correlator

5.3. Store

raw data

5.4. Store

meta-data

5.6. Store calibration results

5.7. Store quick-look results

Primary functional paths Additional functions ALMA software subsystem external agent

Real-time

a. Monitor

points

b. Monitor

points

5.5b. Access raw data & meta-data

h. Store admin data

g. breakpointresponse

5.5a. Access raw data & meta-data

5.1. Get SB

Software Architecture


Pre-production ALMAWater Vapor RadiometerOperating in an SMA Antenna on Mauna Kea(January 19, 2006)

Photo courtesy ofMagne Hagstrom &Ross Williamson

Relay mirrors


System Integration Activities: Prototype Integration

Electronics are first integrated as a system and characterized in the lab at AOC, Socorro.


Canada

• Canada is part of the North American ALMA project• As part of this they are members of the North American Partnership

in Radio Astronomy– This gives them the “right to compete” for time on all NRAO facilities

including ALMA

• They are delivering on of the receiver bands (Band 3) plus cash and software effort to an agreed Value of $20M FY2000– They are also committed to providing 7.25% of the ALMA Operations

costs

• Canada will cover all cost overruns associated with their work– As such they were not part of the ALMA rebaselining exercise

• Canadian ALMA work is covered by an MOU which empowers the NA ALMA PM and the relevant NA IPT leads to direct their work


Japan

• Japanese contribution to ALMA – Enhanced ALMA

• Atacama Compact Array (ACA) System – Twelve 7-m antennas + four 12-m antennas

Higher photometric accuracy– ACA Correlator

high sensitivity, simultaneous realization of wide frequency coverage and high spectral resolution

• New frequency bands– Band 4 (125-163GHz), Band 8 (385-500GHz), and Band 10 (787-

950GHz) [R&D]– Emphasis on submillimeter wavelengths

• Contributions to infrastructure & operations


ALMA-J plans

• Reexamine funding/Value agreements between projects• Complete agreement with ALMA-J – June 2006• Respond to RFQ – summer 2006• Late 2006 – 3rd Executive, E-ALMA


12-m array

ACA

Enhanced ALMA


Reviews, Reviews and More Reviews


AOS 6 Ant Array

Evaluation Complete

1.09 Science Summary Schedule20072006 20092008 20112010 2012

41 2 3 41 2 3 41 2 3 41 2 3 41 2 3 41 2 3 41 2 3

(Data from IPS as of 2006Jan13)

ATF Testing Support

OS

F/A

OS

Commissioning Antenna Array – Finish dates

16th 32nd 50th

Science Verification

AT

F

`

Jan ’10 Early Science

Mar ’09 Early Science Decision Point

Call for Proposals / Early Science Preparation

Sept ’12 Start of Full Science

8th

OSF Integration – Start dates

1st 16th 32nd 50th3rd2nd

SE

&I

Re

fere

nc

eATF Testing

8th

June ’06 ATF First Fringes

SC

IEN

CE

SU

MM

AR

Y

Site Characterization

Science Support OSF

Time Now


J1148+5251: an EoR paradigm with ALMA CO J=6-5

Wrong declination! But…High sensitivity

12hr 1 0.2mJyWide bandwidth

3mm, 2 x 4 GHz IFDefault ‘continuum’ modeTop: USB, 94.8 GHz

CO 6-5HCN 8-7HCO+ 8-7H2CO lines

Lower: LSB, 86.8 GHzHNC 7-6H2CO linesC18O 6-5H2O 658GHz maser?

Secure redshiftsMolecular astrophysicsALMA could observe CO-luminous galaxies (e.g. M51) at z~6.


ALMA into the EoRSpectral simulation of J1148+5251

Detect dust emission in 1sec (5) at 250 GHz

Detect multiple lines, molecules per band => detailed astrochemistry

Image dust and gas at sub-kpc resolution – gas dynamics! CO map at 0”.15 resolution in 1.5 hours

HCN

HCO+

CO

CCH

N. B. Atomic line diagnostics[C II] emission in 60sec (10σ) at 256 GHz[O I] 63 µm at 641 GHz[O I] 145 µm at 277 GHz[O III] 88 µm at 457 GHz[N II] 122 µm at 332 GHz[N II] 205 µm at 197 GHzHD 112 µm at 361 GHz


Bandwidth CompressionNearly a whole band scan in one spectrum

Schilke et al. (2000)LSB USB


Antenna Designs in ALMA

• Three antenna designs currently in hand:– Two will be operated in PSI interferometer in near future:

• Vertex (APEX close copy operational at Chajnantor, destiny of this prototype uncertain).

• AEC (Basis of AEM design, destiny uncertain).

– MElCo prototype disassembled for retrofit to design similar to 3 MElCo production antennas

• Four others expected– Production Vertex design (25-32 antennas)– Production AEM design (25-32 antennas)– Production MElCo 12m antennas (3 antennas)– Production MElCo 7m antennas (12 antennas)

• For present purposes, only consider production Vertex and AEM designs– As these are evolving, must assume they will be identical to the

prototype antennas


Antennas

• Demanding ALMA antenna specifications:– Surface accuracy (25 µm)– Absolute and offset pointing accuracy (2 arcsec

absolute, 0.6 arcsec offset)– Fast switching (1.5 deg sky in 1.5 sec) – Path length (15 µm non-repeatable, 20 µm

repeatable)

• To validate these specifications: two prototype antennas built & evaluated at ATF (VLA)


AEC Prototype Antenna


Vertex Prototype Antenna


VertexRSI and AEC Prototype Antennas

Property VertexRSI AEC

Base/Yoke/Cabin Insulated Steel Steel/Steel/CFRP

BUS Al honeycomb with CFRP plating, 24 sectors, open back, covered with removable GFRP sunshades

Solid CFRP plates, 16 sectors, closed-back sectors glued and bolted together

Receiver Cabin Cynlindrical Invar; thermally stabilized steel

CFRP; direct-connection cabin to BUS

Base 3-point support; bolt connection with foundation

6-point support; flanged attachments

Drive Gear and pinion Direct-drive with linear motors

Brakes Integrated on servo motor Hydraulic disk

Encoders Absolute (BEI) Incremental (Heidenhain)

Panels 264 panels, 8 rings, machined Al, open-back, 8 adjusters (3 lateral/5 axial) per panel

120 panels, 5 rings, Al honeycomb with replicated Ni skins. Rh coated, 5 adjusters per panel

Apex/Quadripod CFRP structure, “+” configuration

CFRP structure, “x” configuration

Focus Mechanism Hexapod (5 DOF) 3-axis mechanism

Total Mass ~108 tonnes ~80 tonnes

Mass Dist. (El/Az) 50%/50% 35%/65%


Science Implications

• Prototypes accepted from manufacturers• Final technical evaluations complete• Both antennas meet the specifications• What happens with two different antenna "designs"

– common mode errors don’t cancel– But differences may help– cost (construction, commissioning, operation)– other ?

• Consider:– Surface differences– Pointing– Pathlength– Mosaicking and polarization


Science Implications:The Antenna Surfaces

Both telescopes easily meet specifications (<25 µm); both are excellent antennas.

Source: AEG Results


Prototype Pointing Results

Source: AEG Results

Spec: 2” all-sky; 0.6” offset pointing under primary operating conditions


Fast Switching

Specification: 1.5 degrees in 1.5 seconds, settling time under 3 seconds.


Path Length Stability

• Spec: 15/20 µm repeatable/nonrepeatable

*Δt = 3, 10, 30 minutes; **Wind-induced, Δt = 15 minutes



• Pointing– Both antennas meet specifications, but the character of pointing

differs– in compact configuration

• WIND: wind "shadowing“ may have some effect• SUN: sunrise may have some effect• GRAVITY: both designs are essentially rigid

– in other configurations• WIND: differs over the site as will the antenna response• SUN & GRAVITY remain constant over the site

• Fast Switching– Both antennas meet specifications

• Awaiting redesign of AEC quadripod

– If not, effect would be to decrease throughput/efficiency



• Phase / pathlength / focus– as pointing, but a more subtle effect.

– Axis non-intersection may be the dominant effect on pathlength (baseline) prediction, and has no common mode error

– Other mechanical deformations would benefit from identical antennas• Gravitational sag, thermal deformation, perhaps other environmental items

• Phase effects due to fiber length– Fiber run to antenna is dominant in effective length change (but if

monitored and corrected, no common mode)

• Polarization matching and primary beam shape– determined by quadripod leg design (shadowing of quadripod legs, but

exact shape plays a minor role too)

– Lesser effect from the differing arrangement of panels and therefore character of scattering from the edges


Fiber Length

• The effective length of the fiber is dominated by the run up the antenna (see ALMA Memo 443).

• Differences in the two designs include– Length of fiber run– Degree of thermal shielding

• Such variations are monitored and compensated.


Pathlength Effects

• Temperature:– Surface RMS changes with ambient temperature from holography:

• * VertexRSI: ~0.6-0.7 micron/K.• * AEC: ~0.8 micron/K.• Both deformations had a high degree of structure (like BUS segment print-

through for VertexRSI, large-scale 45-degree plus inner-ring print-through for AEC); probably in the noise at highest frequencies, where frequent calibration will be done in any event.

– Focal length change due to ambient temperature changes:• * VertexRSI:

– 34 micron/C from holography– 36 micron/C from radiometry

• * AEC:– 14 micron/C from holography– 20 micron/C from radiometry

• All within specification and unlikely to impact science (focus tracked; surface changes small)


Quadripods

• The optical path from the sky off the reflector to the subreflector intercepts the quadripod. In both designs, the solid angle subtended by the quadripod is minimized and the point of attachment to the antenna is as close as possible to the edge of the reflector to minimize shadowing.

• The shadowing profile is less than 1% of the antenna diameter.– Owing to careful minimization of the quadripod profile, the sidelobes will be

small and distant from the primary beam.– Beam profiles were calculated from the shadowing profiles (next slide).

• Quadripod shadowing is known for the Vertex design (ALMA Antenna Group Report #40), estimated for the AEC design by Lucas.

• Reflections are minimized by profiling of the inward edge of the quadripod legs.

• Different lateral motion of the subreflectors with elevation in a homologous antenna could effect cross-polarization; amenable to calculation.

• Shadowing is measured using holography and is the same for both antenna designs within a few tenths of a per cent.– Integrated power <1% of that in the main beam, hence sidelobe power will be

more than 40 dB below that of the main beam.


Quadripod-dependent Questions

Vertex AEC Cross

Three sorts of interferometric baselines provide three sorts of beams:Vertex-Vertex, AEC-AEC, and Vertex-AEC. For the most sensitive imaging,these must all be measured and tracked. The most sensitive images includemosaics and polarization images.


Effects of Quadripod Differences

• “If one ignores the effects of the sidelobes, it is better to have antennas with different configurations; if you are going to correct for it then it is easier if they are all the same.” –James Lamb

• Case One—no correction– The effect of the different sidelobes is small– Since the sidelobes differ, a source won’t be in both at once and

the effect on an image is diminished– Interferometric data provide a strong discriminant for sources

near the main beam owing to fringe rotation/delay offset

• Case Two—correction applied– Worst case is an interfering source in a sidelobe. But with two

designs it cannot be in a sidelobe of all antennas at once. One will want to correct for the different antenna patterns


Summary

• If quadrupod layout is identical, advantage of a single design exist, but is rather limited

25 excellent antennas + 25 good antennas is better than 50 good antennas

50 (or 64) excellent antennas is even better• Each prototype met specifications and qualifies as an excellent antenna• Conclusion: The effect of having two designs for the 12m

antennas in ALMA is small. Any imaging effect can be dealt with for the most sensitive images which might need additional care.

• Cost probably has a greater effect– 2 designs– 2 software interfaces– 2 Assembly, integration, verification, commissioning and science verification– 3 beams to track in the most sensitive applications

www.alma.infoThe Atacama Large Millimeter Array (ALMA) is an international astronomy facility. ALMA is a partnership between Europe, North America and Japan, in cooperation with the Republic of Chile. ALMA is funded in North America by the U.S. National Science Foundation (NSF) in cooperation with the National Research Council of Canada (NRC), in Europe by the European Southern Observatory (ESO) and Spain. ALMA construction and operations are led on behalf of North America by the National Radio Astronomy Observatory (NRAO), which is managed by Associated Universities, Inc. (AUI), on behalf of Europe by ESO, and on behalf of Japan by the National Astronomical Observatory of Japan.

Atacama Large Millimeter Array Update

Documents

Transcript of Atacama Large Millimeter Array Update