Cosmology Large Angular Scale Surveyor - T. Marriage for the CLASS Collaboration U. Michigan - Cosmology After Planck - Sep 24, 2013
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Cosmology Large Angular Scale Surveyor T. Marriage for the CLASS Collaboration U. Michigan — Cosmology After Planck — Sep 24, 2013 Wednesday, September 25, 13
CLASS Collaborators NASA GSFC JHU NIST D. Chuss A. Ali H-M. Cho K. Denis J. Appel K. Irwin A. Kogut C. Bennett (PI) G. Hilton N. Miller J. Eimer C. Reintsema H. Moseley T. Essinger-Hileman K. Rostem D. Gothe CfA-SAO E. Wollack K. Harrington L. Zeng C. Huang PUC de Chile UBC J. Karakla R. Dünner M. Amiri D. Larson M. Halpern Columbia T. Marriage G. Hinshaw D. Araujo D. Watts Northwestern G. Jones Z. Xu G. Novak M. Limon A. Miller Wednesday, September 25, 13
CLASS targets CMB B-modes at large angles. 1) Recombination bump packs a lot of signal. 2) Avoids lensing B-modes. Also E-mode Reionization Constraints 100 CAPMAP BOOMERanG CBI l(l+1)ClBB/2 [ K2] 10 Reionization DASI Bump QUaD 1 WMAP QUIET 0.1 BICEP r=0.7 e ns i ng L 0.01 r=0.2 Grav. 0.001 r=0.01 0.0001 CLASS Noise Level (20 uK-arcmin over 50% of total sky) 10 100 1000 Multipole Moment l Wednesday, September 25, 13
CLASS targets CMB B-modes at large angles. 1) Recombination bump packs a lot of signal. 2) Avoids lensing B-modes. Katayama & Komatsu 2012 Also E-mode Reionization Constraints significant S/N in first 10 ell 100 CAPMAP BOOMERanG CBI l(l+1)ClBB/2 [ K2] 10 Reionization DASI Bump QUaD 1 WMAP QUIET 0.1 BICEP r=0.7 e ns i ng L 0.01 r=0.2 Grav. 0.001 r=0.01 0.0001 CLASS Noise Level (20 uK-arcmin over 50% of total sky) 10 100 1000 Multipole Moment l Wednesday, September 25, 13
CLASS targets CMB B-modes at large angles. 1) Recombination bump packs a lot of signal. 2) Avoids lensing B-modes. Katayama & Komatsu 2012 Also E-mode Reionization Constraints significant S/N in first 10 ell 100 CAPMAP CBI A unique range of angular scales! BOOMERanG l(l+1)ClBB/2 [ K2] 10 Reionization (in a field 1 largely Bump targeting DASI the recombination QUaD peak) CLASS is the QUIET WMAP leader in this regime. 0.1 BICEP r=0.7 e ns i ng L 0.01 r=0.2 Grav. 0.001 r=0.01 0.0001 CLASS Noise Level (20 uK-arcmin over 50% of total sky) 10 100 1000 Multipole Moment l Wednesday, September 25, 13
PIPER Bands 1.0 Transmission 100 CLASS is an 150 GHz Band 40 GHz Band 90 GHz Band array of 4 telescopesBOOM Atmospheric 0.6 l(l+1)ClBB/2 [ K2] 10 operating at DASI Reionization Bump 0.2 three frequencies 1 WMAP 0.1 that straddle the BICEP 100 r=0.7 synch foreground 0.01 r=0.2 minimum. Temperature ( K) E modes dust 1 Antenna 0.001 B modes Additional r=0.01 foreground 0.01 0.0001 constraints from PIPER (200 GHz, 270 GHz) and 10 100 1000 Planck (217,10353 GHz) 10 Frequency (GHz) Multipole M FigureCLASS Frequency 1: (Left:) The frequency Detectors bands in which Resolution will observe this instrument Surveyforeground spectral the Galactic 40 GHz minimum and 72 to minimize atmospheric e 1.5o shownDesign is from the Atacama with 1 mm precipitable water vapor. Synchro 90 GHz 15 1200 40’ are based on WMAP 5-year data . This instrument complements the hig Parameters 150 GHz 120 24’ of other instruments, such as PIPER, SPIDER, and EBEX so the combin Wednesday, September 25, 13
To detect large-angle modes, CLASS needs a wide survey. The Atacama is the best site for large sky coverage. Polarbear 23o ACT/ABS CLASS 65% of the Extragalac9c Sky CLASS ALMA APEX NANTEN Site in Atacama Desert is not far from the equator: most of sky is surveyed at zenith angle 45 deg. Wednesday, September 25, 13
To detect large-angle modes, CLASS needs a wide survey. CLASS Survey Boundary Galactic Mask 23o CLASS 65% of the Extragalac9c Sky Extragalactic Survey Region in Black Site in Atacama Desert is not far from the equator: most of sky is surveyed at zenith angle 45 deg. Wednesday, September 25, 13
To detect large-angle modes, CLASS needs a wide survey. Multiple observing angles through sky and deck rotation 23o CLASS 65% of the Extragalac9c Sky in the JHU highbay Site in Atacama Desert is not far from the equator: most of sky is surveyed at zenith angle 45 deg. Wednesday, September 25, 13
The CLASS Way 1. Systematics control with front end modulation. 2. Sensitivity with high efficiency optics and TES bolometers cooled to 150 mK. 3. Galactic foreground cleaning with multi- frequency telescope array. Wednesday, September 25, 13
Continuous Operation The CLASS Way with 50 μW at 100 mK 1. Systematics control with front end modulation. 2. Sensitivity with high efficiency optics and TES bolometers cooled to 150 mK. 3. Galactic foreground cleaning with multi- One of the four CLASS receivers frequency telescope (PT+DR Cooler) undergoing array. tilt test. Wednesday, September 25, 13
The CLASS Way 1. Systematics control with front end modulation. 2. Sensitivity with high efficiency optics and TES bolometers cooled to 150 mK. 3. Galactic foreground cleaning with multi- 40 GHz Focal Plane Assembly. frequency telescope array. Wednesday, September 25, 13
The CLASS Way #1: Systematics control with front end modulation Wednesday, September 25, 13
CLASS uses modulation to measure large scales. 2. The Variable-delay Polarization Modulator (VPM) Input Port Output Port A Variable-Delay Polarization Modulator (VPM) is the front-end optical element. θ Modulates signal at ~5 Hz to separate signal from the I-to-Q leakage of g atmosphere and other instrument- Wire Grid related drift. d 2θ Mirror Moves Mirror Fig.dP/dU 1. The VPM consists of a polarizing grid placed in front of and para dP/dQ a planar mirror. The polarization parallel to the grid wires is reflected b dP/dQ grid. The orthogonal linear polarization passes through the grid and is refl off the mirror. The two components Q are :1st recombined harmonic at the output port a relative phase delay that is dependent upon the grid-mirror separatio The wire grid spacing or pitch is indicated by g. Q The VPM consists of a polarizing grid positioned parallel to and in front of V: 2nd harmonic change in grid-mirror separation corresponds to a change in introduced phase linear orthogonal polarizations. The device is show in Figure 1. Choosing coor that Stokes Q gives the difference between the polarization states parallel and p Wednesday, September 25, 13 to the VPM wires, the polarization transfer function can be expressed as
CLASS uses modulation to measure large scales. CMB Simulation A Variable-Delay Polarization Modulator Miller et al (Prep) (VPM) is the front-end optical element. Modulates signal at 5-10 Hz to separate and signal the results from the are discussed in (unpolarized) section 7. atmosphere and other instrument- related larization Modulator drift. (VPM) Recovery with Modulation Q Input Port Output Port and simple map-making θ i f t Dr g Wire Grid Signal d 2θ Mirror *Both the atmosphere and gain time streams have 1/f2power spectra. The t atmosphere has an amplitude of 0.05 K at 0.1 Hz and the gain fluctuation has an amplitude of 0.5% at 0.005 Hz. Atmosphere + Differential Gain* tsWednesday, of a polarizing September 25, 13 grid placed in front of and parallel to
CLASS uses modulation to measure large scales. CMB Simulation A Variable-Delay Polarization Modulator Miller et al (Prep) (VPM) is the front-end optical element. Modulates signal at 5-10 Hz to separate and signal the results from the are discussed in (unpolarized) section 7. atmosphere and other instrument- related larization Modulator drift. (VPM) Recovery without Modulation Q Input Port Output Port and simple map-making θ i f t Dr g Wire Grid Signal d 2θ Mirror *Both the atmosphere and gain time streams have 1/f2power spectra. The t atmosphere has an amplitude of 0.05 K at 0.1 Hz and the gain fluctuation has an amplitude of 0.5% at 0.005 Hz. Atmosphere + Differential Gain* tsWednesday, of a polarizing September 25, 13 grid placed in front of and parallel to
CLASS uses modulation to measure large scales. CMB Simulation Miller et al (Prep) 200μ pitch Recovery without Modulation 50 cm and simple map-making Prototype *Both the atmosphere and gain time streams have 1/f2 power spectra. The atmosphere has an amplitude of 0.05 K at 0.1 Hz and the gain fluctuation has an amplitude of 0.5% at 0.005 Hz. Atmosphere + Differential Gain* Wednesday, September 25, 13
Preliminary Simulation Results EE theory, r=0.01 Effects Included: EE systematic BB systematic *Atmosphere EE theory *VPM temperature drift + differential emissivity BB theory r=0.01 *Detector pair differential gain fluctuation *VPM-Detector misalignment (0.5 deg) 10% BB systematic due to VPM-Detector misalignment likely further reduced by adjusting angle so EB=0 Miller et al. (in prep) Wednesday, September 25, 13
The CLASS Way #2: Sensitivity with high efficiency optics and tweaked-up detectors cooled to 150 mK Wednesday, September 25, 13
CLASS Detectors : Design Designed and made by Horns and Planar OMT produce simple single-moded beams. High-efficiency and design repeatability is achieved through use of monocrystalline silicon dielectric. Intrinsic OMT design achieves broad 50% fractional bandwidth, which may be divided for multi-frequency operation. On-chip transmission line filtering, shielding and niobium gap provide well defined bandpass and stringent blue leak control. 40 GHz Detector Integrated backshort + Magic Tee transmission line shielding Probe Antenna Band filter detector chip Wednesday, September 25, 13
CLASS Detectors : Excellent Design; Excellent Tests TESs with tuned transitions and Detectors coupled to thermal conductances thermal source show 90% efficiency and no out of band leakage. Wollack et al. (in prep) Wednesday, September 25, 13
CLASS Detectors : Sensitivity (or how to detect B-modes with fewer than 10,000 detectors) Three Designs* 20 0μ N K- ET 150 GHz √ s = high efficiency Phonons 27 150 mK 0 μK 150 GHz -√ 14 modest efficiency s 0 500 mK μK 90 GHz -√ high efficiency 150 mK s Photons *examples for argument; not exact; for instance need to add amplifier noise Significant advances in SQUID amplifier noise over previous experiences. Wednesday, September 25, 13
CLASS Detectors : Sensitivity (or how to detect B-modes with fewer than 10,000 detectors) Three Designs* 20 0μ N K- ET Bicep II has 150 GHz √ s = nearly achieved high efficiency Phonons 27 this 150 mK 0 μK 150 GHz -√ 14 CLASS modest efficiency s 0 500 mK μK goal 90 GHz -√ high efficiency 150 mK s Photons *examples for argument; not exact; for instance need to add amplifier noise Significant advances in SQUID amplifier noise over previous experiences. Wednesday, September 25, 13
The CLASS Way #3: Galactic foreground cleaning with multi- frequency telescope array Wednesday, September 25, 13
Template-based Likelihood for r, s, and Foregrounds (Efstathiou et al. 2009; Katayama & Komatsu 2011) Full likelihood is computationally feasible because we are probing large angles=fewer data with larger signal per-alm. Approach infeasible at smaller angles. Built-in handling of E-B mixing etc. Wednesday, September 25, 13
Galactic Coords 40 GHz Template cleaned map (Noise=11 μK-arcmin) 90 GHz Residuals from CMB 270 input map at 5nK level GHz (PIPER) 15 μK-arcmin noise Watts, Larson et al (in prep) Wednesday, September 25, 13
Exploring Constraints with Sky Cuts and Foregrounds (Pixel-based likelihood as in Katayama & Komatsu 2011) Note Non-Gaussian dust rinput = 0.02 dashed lines = likelihood input values using large angular scales can yield a synchrotron detection with tail to high r. tensor-to-scalar ratio Watts, Larson et al (in prep) Preliminary!!! More work to be done. Wednesday, September 25, 13
Exploring Constraints with Sky Cuts and Foregrounds (Pixel-based likelihood as in Katayama & Komatsu 2011) Analytic calculation (K&K ’11) Note Non-Gaussian dust rinput = 0.02 dashed lines = likelihood input values using large angular scales can yield a synchrotron detection with tail to high r. tensor-to-scalar ratio Watts, Larson et al (in prep) Preliminary!!! More work to be done. Wednesday, September 25, 13
Outside View on r (Upper limits) Current experiment predictions (trend ruled not just by shear sensitivity but systematics etc) Wednesday, September 25, 13
Stay tuned! Deploying telescopes 2014-2015. Mounts Cryostats Atacama Site Preparation Polarbear ACT/ABS CLASS Optics Focal Planes VPMs 1.5 m Wednesday, September 25, 13
You can also read