mmVLBI in the Future:
technical/experimental
Shep Doeleman
Haystack Observatory
In Defense of Theory
Do I contradict myself?
Very well then I contradict myself,
(I am large, I contain multitudes.)
--Walt Whitman, "Song of Myself"
High Frequency VLBI Challenges
• Sensitivity: limited by atmospheric
coherence, telescope apertures, bandwidth,
weather (opacity and coherence).
– Past observations limited to brightest
sources.
• Baseline coverage: limited by relatively small
number of high demand sub/mm telescopes.
• LO quality: phase noise in electronics
increases – loss of fringe amplitude.
– First VLBI at new mm sites requires effort.
Challenges cont’d
• Scheduling: mm/submm telescopes difficult
to schedule.
• Array phasing: all antennas must be phased
prior to summing (you only get one chance).
• Hydrogen masers: if atmosphere is good,
maser should not limit coherence time.
• Amplitude calibration is difficult: array
phasing, pointing, atmosphere, gain curves.
• BBCs have limited Bandwidth
• Major improvements can be expected in
some of these areas.
Sensitivity: a solution.
• Old conventional wisdom: collecting area.
• New technical developments – new wisdom
– 1983 VLBI BW: 224Mb/s,
– 2004 VLBA BW: 512Mb/s.
– New disk based VLBI recording systems will
allow multi Gb/s data rates
Mark 5A VLBI Data System
Mk5 cost ~ $15K
700 GB disks expected ~2005 – 24 hours @ 1 Gbps unattended
(comparable to ~19 VLBA tapes)
Sensitivity: a solution.
• Old conventional wisdom: collecting area.
• New technical developments – new wisdom
– 1983 VLBI BW: 224Mb/s,
– 2004 VLBA BW: 512Mb/s.
– New disk based VLBI recording systems will
allow multi Gb/s data rates.
– Leverages COTS technology and industry
demand for affordable, high capacity, high
speed hard disks.
– Units can be used in parallel for N Gb/s.
Tape vs. Disc Price Comparison
7
6
5
Disc Drive
Street Prices
log($/GB)
4
1998 IBM
Disc Projection
3
c
Computer Tape
Media
2
1998 NSIC
Disc Projection
1
LTO Media
Projection
Mark IV/VLBA/K4 Media
0
S2, S3 Media
Disc industry
Projections
-1
-2
1980
1985
1990
1995
Year
2000
2005
2010
mmVLBI at Haystack
• 2001-present: focus on 1-2mm VLBI
• 2002: Carried out successful 129, 147GHZ
VLBI on Pico Veleta, KittPeak12m, SMTO
triangle:
– High resolution: Pico-SMTO fringes – 49mas
– SiO Masers
– Collaborators: MPIfR, IRAM, Onsala,
Metsahovi, Arizona Radio Observatory
mmVLBI In the News
v=1 J=3-2 SiO Masers in VY CMa
• First 129GHz VLBI
image using all phase
and amp information.
•Relative astrometry
small fraction of beam.
30 AU
mmVLBI at Haystack
• 2001-present: focus on 1-2mm VLBI
• 2002: Carried out successful 129, 147GHZ VLBI
on Pico Veleta, KittPeak12m, SMTO triangle:
– High resolution: Pico-SMTO fringes – 49mas
– SiO Masers
• 2003: Carried out 129, 230GHz VLBI using Pico,
KittPeak, SMTO, Plateau de Bure
– Detected Pico-SMTO 1.3mm fringes – 34mas on 3C279
– World Record: equivalent to 5 RSCH of SgrA* MBH
– Used new generation of VLBI recorders.
• Very promising, but future depends on funding.
Evolution of 3mm Capability
Present: 256Mb/s
3-5 years: 1Gb/s
Antenna
Antenna
SEFD
(Jy)
Antenna
GBT
500
ALMA
VLBA
SEFD
(Jy)
5170
5-10yrs: 10Gb/s
PicoVeleta 1700
Apex (Chile) 5000
P. de Bure
1700
CARMA
1700
Onsala
4000
LMT
350
Effelsberg
2000
Nobeyama
3000
VLBA-VLBA = 0.8Jy
PicoV-PdeB = 0.25Jy
PicoV-PdeB = 0.13Jy*
PicoV-VLBA = 0.5Jy
VLBA-VLBA = 0.4Jy
GBT-LMT = 0.032Jy
GBT-LMT = 0.016Jy*
LMT-VLBA = 0.1Jy
SEFD
(Jy)
60
VLBA-VLBA = 0.13Jy
ALMA-LMT = 0.004Jy
ALMA-VLBA = 0.014Jy
Resolution (8000km)=90mas
Tcoh = 20 seconds
Higher recording rates.
Evolution of 1mm Capability
Present: 0.5Gb/s
3-5 years: 4Gb/s
Antenna
Antenna
SEFD
(Jy)
Antenna
JCMT
7800
ALMA
KittPeak
SEFD
(Jy)
22000
5-10yrs: 10Gb/s
PicoVeleta 2500
APEX-Chile 5500
P. de Bure
2500
CARMA
2500
SMT
15000
LMT
500
SMA
2500
CSO
15000
PicoV-PdeB = 0.4Jy
PicoV-KittPeak = 1.15Jy
LMT-CARMA = 0.06Jy
LMT-SMT= 0.15Jy
SEFD
(Jy)
100
ALMA-LMT = 0.008Jy
ALMA-SMT = 0.042Jy
Resolution (8000km)=34mas
Tcoh = 10 seconds
UltraVLBI Proposal
• Submitted to ATI program at NSF
• Outfit largest cm antennas (Effelsberg, Arecibo,
GBT, WSRT, Jodrell bank) with 4Gb/s recording
systems: <1mJy/beam rms.
• Outfit best mm sites with 4Gb/s systems: JCMT,
SMTO (CARMA, LMT, SMA, ALMA, APEX)
• Target science requiring sensitivity: SNR in
Galactic mergers, GRBs, Grav. Lenses, Stellar
VLBI, SgrA*!!
• Waiting to hear …
230GHz VLBI Observations
Proposed VLBI
on JCMT-HHT
baseline.
Just a detection will
provide a firm upper
size limit.
Next Generation VLBI System
• Multiple Digital BBC units (500MHz each).
– Cost: 20K each.
• Multiple MK5b units (1Gb/s each).
– Cost: 16K each.
• H-maser with sy(10sec)<=3e-14.
– Cost: 175K.
• Rack with power supplies, backplane, GPS clock,
– Cost: 20K
• Total: $270K/new station, much less if maser is
already available.
Summary
• Over next decade: BW x40, Element area x25, means
capable 230GHz imaging arrays.
• Phase stability of mm/submm arrays is very important:
CARMA, ALMA, PdeBure.
• Creating low phase noise LO systems for submm
VLBI is necessary.
• VLBI arrays up to 345 and 450GHz are possible.
• ‘Widebanding’ IFs and receivers will be a requirement.
• Scheduling: need turn-key operation from single dish
to VLBI.
• Current proposed work: 4Gb/s VLBI on SgrA* in 3
years.
END