EBEX The E and B EXperiment Will Grainger Columbia University Moriond 2008

Collaboration APC – Paris Radek Stompor Brown University Andrei Korotkov John Macaluso Greg Tucker Yuri Vinokurov CalTech Tomotake Matsumura Cardiff Peter Ade Enzo Pascale Columbia University Amber Miller Britt Reichborn- Kjennerud Will Grainger Michele Limon Harvard Matias Zaldarriaga IAS-Orsay Nicolas Ponthieu Imperial College Andrew Jaffe Lawrence Berkeley National Lab Julian Borrill McGill University Francois Aubin Eric Bisonnette Matt Dobbs Kevin MacDermid Oxford Brad Johnson SISSA-Trieste Carlo Baccigalupi Sam Leach Federico Stivoli University of California/Berkeley Adrian Lee Xiaofan Meng Huan Tran University of California/San Diego Tom Renbarger University of Minnesota/Twin Cities Asad Aboobaker Shaul Hanany (PI) Clay Hogen-Chin Hannes Hubmayr Terry Jones Jeff Klein Michael Milligan Dan Polsgrove Ilan Sagiv Kyle Zilic Weizmann Institute of Science Lorne Levinson

EBEX in a Nutshell CMB Polarization Experiment Long duration, balloon borne Use 1476 bolometric TES 3 Frequency bands: 150, 250, 410 GHz Resolution: 8’ at all frequencies Polarimetry with half wave plate BLAST (+ BOOM, MAXIMA) balloon technologies

Science Goals Detect (or set upper bound) in inflationary B- mode –T/S < 0.02 at 2σ (excluding systematic and foreground subtraction uncertainty) Detect lensing B-mode –5% error on amplitude of lensing power spectrum Measure E-E power spectrum Determine properties of polarized dust EBEX, 14 days

Dust Determination and Subtraction Simulate CMB B, dust, noise Reconstruct dust + CMB maps (using the parametric separation technique) Less than 1/3  increase in error on recovered CMB over binned cosmic variance and instrument noise due to foreground subtraction for l=20 to 900. Reconstruction of dust spectral index within 5% Blue = INPUT dust model Red = INPUT CMB + instrument noise + sample variance Black dust = data + errors of reconstruction Black CMB = variance of 10 simulations No systematic uncertainties

Design 250 cm

Cryostat and Optics Polarimetric systematics: Half Wave Plate Efficiency: Detection of two orthogonal states Stop + Reflecting Gregorian Dragone telescope Control of sidelobes: Cold aperture stop

Focal Planes 738 element array Single TES Strehl>0.85 at 250 GHz 3 mm Total of 1476 detectors Maintained at 0.27 K 3 frequency bands/focal plane G = 10 pWatt/K NEP = 1.1e-17 (150 GHz) NEQ = 136 μK*rt(sec) (150 GHz) msec, Meng, Lee, UCB 2.1 mm cm

SQUID arrays (NIST) Digital Frequency Domain Multiplexing (McGill) Detector Readouts LDB: 495 Watt for x12; 406 Watt for x16 FPGA Synthesizes Comb; Controls SQUIDs; Demodulates

5 stack achromatic HWP 0.98 efficiency for 120< ν < 420 Ghz 6 Hz rotation < 10% attenuation from 3 msec time constant Driven by motor outside cryostat via Kevlar belt Supported on superconducting magnetic bearing Half Wave Plate Polarimetry

EBEX Scan Scan is: Constant elevation for 4 repeats, one Q,U per 1/3 beam, (0.7 deg/sec). Step in elevation, and repeat; 102 times. Repeat that 6 hour block on same patch of sky for 14 days. Multiple visitations per pixel from various angles (i.e. crosslinking) on various timescales. Relatively uniform coverage Up to 10^8 samples/beam 17 deg p-p / 0.7 deg/sec x steps6 hours All 150 GHz detectors, 14 Day

Gondola + Pointing Cable Suspension (a-la BLAST) Pointing System (BLAST, MAXIMA, Boom) Gondola integrated at Columbia U. Pointing tests ongoing

EBEX Summary + Schedule 14 day flight 420 deg 2 ~24,000 8’ pixels Low dust contrast (4  K rms) 796, 398, 282 TES detectors at 150, 250, 410 GHz 0.7  K/8’ pixel - Q/U; 0.5  K/8’ pixel – T Currently integrating detectors into cryostat in UMN Pointing sensors onto gondola in CU North American flight: Autumn 2008 Long Duration (Antarctic) flight: Austral Summer 2009

Nothing to see here…

Optics Polarimetric systematics: Half Wave Plate Efficiency: Detection of two orthogonal states Reflecting Gregorian Dragone telescope Control of sidelobes: Cold aperture stop wide range of ls probed. Stop +

Design 250 cm Blue – Synchrotron Pink – Dust Minimize synchrotron by going to high frequency, then only one foreground to deal with.