cryogenic measurements of cmb polarimeters 073009.ppt filepolarimeter development at jpl reduced the...

Cryogenic Measurements of CMBCryogenic Measurements of CMB Polarimeters

P. Kangaslahti, K. Cleary, T. Gaier, D. Harding, S. Xenos, A. Campbell, M. Soria, L. Newburgh*, R. Williamson*, B H b * R D li * A Mill * A K k **B. Hooberman*, R. Dumoulin*, A. Miller*, A. Kusaka**,A. Brizius**, C. Bischoff**, D. Kapner**, I. Buder** and 

B. Winstein**

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA

*Columbia University New York NYColumbia University, New York, NY

** University of Chicago, Chicago, IL

Copyright 2009 California Institute of Technology. Government sponsorship acknowledged.

Detecting polarization of Cosmic Microwave Background (CMB) with Q, U Imaging ExperimenT (QUIET)g g p

• QUIET experiment phase 1: 91-element array at 90 GHzp p yand a 19-element array at 40 GHz operate for one year

• 1m side-fed Cassegrain telescope mounted on theCosmic Background Imager (CBI) platform in Chajnantor,g g ( ) p j ,Chile (altitude of 5080 m)

• QUIET is an integrated approach to characterizing theCMB polarization power spectra using 20% frequencyCMB polarization power spectra using 20% frequencybandwidth arrays of 40 GHz and 90 GHz polarimeters

• QUIET arrays offer the most sensitive detectortechnology for ground-based CMB observations attechnology for ground based CMB observations at100 GHz or below

6/24/2009 2Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Detecting polarization of Cosmic Microwave Background (CMB) with Q, U Imaging ExperimenT (QUIET)g g p

• Established techniques for controlling polarizationq g psystematic error

• Each module in the array measures simultaneously bothQ and U, the linear Stokes parameters (we also includedQ , p (a few modules for I parameter measurement)

• Array is a self-contained unit incorporating feed optics,polarizing elements, lock-in modulation (4kHz), tunablepolarizing elements, lock in modulation (4kHz), tunabledetector biasing, and analog to digital conversion

6/24/2009 3Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Polarimeter development at JPL reduced the size of the receiver for QUIET

QUIET 90GHz Polarimeter• Suitable for automated volume productionvolume production

• Fits behind the feed horn 

CapMap 90GHz Polarimeter

6/24/2009 4Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Block diagram of the polarimeterLens

Feedhorn

OMT (L-R circular)

h, v

The OMT splits the radiation h l f h MCM

LNA

LNA

Mod

ule

h+iv h-ivto the two legs of the MCMIndium Phosphide MMIC LNAs provide the low noise amplification for coherent LNA

Phase switch

LNAed M

ultic

hip amplification for coherent 

detectionPolarization hybrids enable the detection of Stokes LNA

180 deg hybridPower splitterBandpass filter

Inte

grat

ethe detection of Stokes parameters Q and U

Diode detectors

v90 deg hybridh

h v h+vQ detectors

6/24/2009 5Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

h-v h+v

U detectors

Planar multichip module polarimetersDetectors BP filters

QUIET 90GHz Polarimeter WG input

LNA

Phase  Polarization 

6/24/2009 6Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Switch hybridsQUIET 40GHz Polarimeter

Planar multichip module polarimeters

Planar modules work well in automated assembly

PolarimeterEpoxies

MMICs, substrates

Epoxies

6/24/2009 7Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Photo by NxGen Electronics

Testing of the 40 GHz polarimeters

Bi PCB

Polarimeter

Bias PCB

Magic T

Coupler for

g

Coupler forCW signal

Temperature controlled 

l d

6/24/2009 8Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

WG load

Testing of the 40 GHz polarimeters

Cryogenic measurements:

• Noise temperature of less than

1.0E

-02

V]

h output

Noise temperature of less than 25 K on best modules, 25K ‐ 30 K typicalF b d idth i 8 GH 04

1.0E

-03

Vol

tage

[ V

• Frequency bandwidth is 8 GHz

• Detector isolation > 15 dB

1.0E

-0

35 37 39 41 43 45 47 49

Frequency [GHz]

02 v output

E-0

31.

0E-0

Vol

tage

[V]

v output

1.0E

-04

1.0E

35 37 39 41 43 45 47 49

Det

ecte

d V

6/24/2009 9Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

1 35 37 39 41 43 45 47 49

Frequency [GHz]

Testing of the 90 GHz polarimeters

• Aim:• Aim:– Establish baseline performance of modules– Room‐temperature & cryogenic (20 K)P ti• Practice:– Check basic operation of each module

• Polarity of diodes correct?h d?• Devices shorted?

• Phase switches turn on?– May require several iterations– ‘Manually’ find ‘good’ bias values, to optimise

• Balance• Isolation

– Measure swept response– Measure system temperature 

• Also tests to try to improve systemic problems

6/24/2009 10Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

• And testing new module designs

Testing of the 90 GHz polarimetersFrequency sweep test set‐upq y p p

Sweep Generator

LabView PCLabView PC

BiasPower Supplies

DiodeMultimetersGPIB

6/24/2009 11Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Testing of the 90 GHz polarimetersNoise test set‐up

Noise obtained in a QUIET polarimeter module, at 30K ambient

6/24/2009 12Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Testing of the 90 GHz polarimeters

Histogram of the receiver temperatures of leg A (left) and leg B (right)for 35 production modules, derived from a linear fit to load temperatures in the range 27-52 K. The median TRX for leg A is 55 K and for leg B is 54 K The receiver noise temperature is the average of the noise

6/24/2009 13Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

54 K. The receiver noise temperature is the average of the noise temperatures of the two legs.

Testing of the 90 GHz polarimeters

Gain fluctuations• Common-mode gain fluctuations

subtracted out• Phase modulation, differencing and

weighting• Knee frequency 100-300 mHz

6/24/2009 14Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Joey Richards & Mike Seiffert (JPL)

Noise Testing of the 40 GHz polarimeter array

Cryogenic measurements:

• noise testing setup atnoise testing setup at Columbia University

• two zotefoam loads LN2 (77K) / LO2 (90K)

6/24/2009 15Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Noise Testing of the 40 GHz polarimeter array

Frequency sweeps

6/24/2009 16Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Polarised testing of the 90 GHz Array 

Testing set up at University of Chicago

Reflector rotates above the cryostatEnables automated optimisation of the modulesThe signal is provided by aThe signal is provided by a rotating 1% polarising wire grid in front of a 77K target Amplitude of the signal isAmplitude of the signal is roughly ~1K

6/24/2009 17Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Polarised 90 GHz Array testingPlot of a time series of data of all 91 receivers on the 90 GHz QUIET arrayThe bottom row in black are I sensitive receivers.

QUIET 90 GHz array assembledQUIET 90 GHz array assembled

6/24/2009 18Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS

Instrument integration and operation

QUIET 40 GHz array integration and on‐site with ground screen assembled

6/24/2009Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS19

CMB observations – initial results

6/24/2009Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS20

QUIET capabilities

6/24/2009Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS21

Slide: C.R. Lawrence (JPL)

Conclusions

• Successfully completed build and test of over 100• Successfully completed build and test of over 100polarimeters

• Automated assembly and array optimization was applied(current production rate about 20 units/month)(current production rate about 20 units/month)

• Module noise temperatures were about 20% higher thanamplifiers tested individuallyT i (20K) i t t d• Two cryogenic (20K) arrays integrated

• QUIET has completed phase I Q‐band observations andstarting phase I W‐band observations

6/24/2009Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS22

Acknowledgement

Thi h i d t t th J t P l i L b tThis research was carried out at the Jet Propulsion Laboratory,California Institute of Technology, under a contract with the NationalAeronautics and Space Administration

We would like to acknowledge the work done by all members of theQUIET collaboration (5 countries, 12 institutions, ~32 people)http://quiet.uchicago.edu/team

The QUIET project was funded by NSF

6/24/2009Low Noise Figure Measurements at 

Cryogenic and Room Temperatures WS23