High Sensitivity VLBI Sheperd Doeleman MIT Haystack Observatory.

High Sensitivity VLBI

Sheperd Doeleman

MIT Haystack Observatory

BW and Continuum Sensitivity

• Bandwidth much cheaper than steel.

• New technical developments – – Disk based VLBI systems – Digital wideband backends

• VLBA now: 256Mb/s sustainable

Rate BW BW increase Sens increase

2 Gb/s 250 MHz x 8 x 2.8

10 Gb/s 1.25 GHz x 40 x 6.25

64 Gb/s 8 GHz x 250 x 16

•EVLA WIDAR Correlator and software correlators can process 8GHz stations.

cm UVLBI Array• Arecibo, Jodrell Bank (Lovell), Effelsberg,

Westerbork, GBT

• Rms map sensitivities: (2 hours with Arecibo)

1 Gb/s 2 Gb/s 4 Gb/s

L Band Array 2.3 Jy 1.8 Jy 1.3 Jy

C Band Array 3.1 Jy 2.4 Jy 1.7 Jy

Motivation for wideband DBE• VLBI data rates up by only x4 since 1980’s:

– Moore’s Law ~10,000– SNR~(BW)0.5 * (Diam)2

• Modern FPGAs give increased performance at small fractions of Mark4/VLBA cost (sample IF, filter digitally).

• For continuum obs. widebanding now cost effective vs. larger dishes: 512Mb/s to 4Gb/s same as 42m VLBA dishes.

• DBE systems are portable.• Wideband obs. important for key science:

– SgrA*, GRB afterglows, ULIGs, Gravitational Lenses, Pulsar Astrometry, Astronomical Masers

• Industry driven growth path: DSP, storage media, high speed data protocols (10GbE).

DBE prototype

IF1 Sampler1 PFB1

IF2 Sampler2 PFB2

1024 MHzSampleclock

ChannelSelect

VSI2

VSI1

TVG

CLOCK

1PPS1PPSGen

512 MHzclock

ChannelGain

Control(serial port)

Ext 1PPS tick

Sampler Module PFB Module

(500MHz BW)

(500MHz BW)

(2048 Mbps)

(2048 Mbps)

Current modes: 15 channels, each 32MHz, 2-bit = 1920Mb/s 15 channels, each 16MHz, 2-bit = 960Mb/s

H Maser

Prototype DBE

Sampler boards

iBOB

Hardware developed by Berkeley Space Sciences Lab (CASPER)iBOB board and iADC sampler. Haystack/CASPER collaboration onFirmware.

COST: $7.5K (includes purchase ofVirtex2Pro FPGA) for 4Gb/s.

16MHz vs 32MHz Channels8 vs 4 filter taps

16MHz Channel4 taps

32MHz Channel8 taps

Disk Based Recorders

Mark5B: 2Gb/s using VSI interface.Mark5C: 4Gb/s using 10GbENext Gen: 8-16Gb/s using 10GbE (Commercial Off The Shelf – COTS)

Final pieceof the system

Next stop 16Gb/s

230GHz VLBI2x 2Gb/s = 4Gb/s

SMTO-JCMT: 1749+096 - 60uas fringe spacing. Tcoh~60sec

Stellar UVLBI

• Stars exhibit radio activity all over HR diagram - at various stages of stellar evolution.

• Non-thermal radio emission, due to magnetic activity – VLBI scales.

• Magnetic fields are critical in Pre Main Sequence stellar evolution with energetic particles emitting both Xrays and gyromagnetic radio.

• Radio follow up of identified PMS stars from Spitzer surveys.

• Use UVLBI to differentiate between stellar flares, magnetospheres, star-disk interfaces.

• Brown Dwarfs: mysterious mechanisms.

xray/radio correlation

Dwarf LP944-20Violates this relationby 4 orders of magnitude.

Central Gravitational Lens Images• Lens theory predicts ‘odd’ number of images, but almost

all systems have 2 or 4: a mystery.

H1413+117B2114+022

• ‘Missing’ images are faint and close to lensing galaxy: can’t see them in the optical.

• Only one central image has been detected so far, but UVLBI sensitivities should be sufficient to detect ~50%.

• Statistical studies of central regions of galaxies possible.

Central Image

Winn et al 2004 J1632-0033

Free-free absorptionin lensing galaxy, 200pcfrom core.

Density and BH mass• Detection or limits on central image flux density constrains core size and steepness of density profile.

• SMBH at center of lensing galaxy can create a 4th image (central) that can be used to directly estimate mass of BH.

- 30% flux density of first central image.- ~20mas separation

Boyce et al 2006

Planned Observations

• Nov 2007 – 8 asymmetric double lens systems.• Arecibo – GBT at 4Gb/s (<2Jy/beam)

– factor of x10 improvement in limits.

• VLBA at lower bit rate to model bright images and subtract from AR-GBT data.

• Goal: probe cores of z~0.3-1 galaxies on same scales as HST does for local galaxies.

• High resolution and high sensitivity required.– SKA ideal instrument

– AR-anchored VLBI arrays available now (>0.1SKA)

Pulsar Astrometry• New VLBI instrumentation allows record-

time gating of pulsar data: wide bandwidths but conserving of media.

• Astrometry complements pulsar timing:– VLBI observes all pulsar types (not just MSP).– Supernovae core collapse (pulsar proper

motions)– NS-supernovae associations.– Timing vs. VLBI: ties together extragalactic

and solar system ref. frames.