[1] The DES Collaboration, The Dark Energy Survey whitepaper; submitted to the Dark Energy Task Force June 2005, astro-ph/0510346. See booth at this meeting.

[2] S.E. Holland, D.E. Groom, N.P. Palaio, R. J. Stover, and M. Wei, "Fully Depleted, Back-Illuminated Charge-Coupled Devices Fabricated on High-Resisitivity Silicon," IEEE Trans. Electron Dev. 50 (3), 225--238 (January 2003) LBNL-49992.ttp://www.noao.edu/ets/monsoon/

Dark Energy Camera CCD and Imager Production and Testing

Dark Energy Camera (DECam) is a 520 Mpix, 3 square-deg FOV imager being built for the Blanco 4m Telescope at CTIO. This facility instrument will be used for the "Dark Energy Survey“ [1] of the southern galactic cap. DECam has chosen 250 um thick CCDs with good QE in the near IR for the focal plane. We present the progress in production and testing of the detectors and focal plane assembly. We compare them with the technical requirements.

CCDsDeveloped by LBNL [2].• 250 microns thick, fully depleted • 2 Readout channels/device • 2k x 4k 15µm pixels. Size 31.5 mm x 63 mm• High QE in the near infra-red compared to standard astronomical CCDs.

Summary

Production CCD packaging and testing is proceeding well. We are about 2/3 finished. So far we have 63+ devices that satisfy the DES requirements. We are now building up a supply of spare detectors.

Some characteristics of the science grade devices are shown in detail.

We use a prototype camera to perform integration studies and operation of the CCDs, electronics, and cooling system.

Construction of the production imager is nearly done.

We will test DECam using a “Telescope Simulator”.

DES wafer with four 2048x4096, one 2048x2048, and eight 1024x512 CCDs.

CCD Test Setup at Fermilab

At Fermilab we have a CCD testing facility with five testing stations; For the full focal plane we need to have 622kx4k devices (+ spares), for which we expect to test about 250 CCDs. We also need to produce 12 (+ spares)2kx2k packages for Guide and Focus needs. As of December 25, 2009 we have 63 science grade 2kx4k packagesthat have passed all test criteria; enough to build DECam. We have 6 science grade 2kx2k packages.

We use standard CCD measurement techniques such as photon transfer curve, clock voltage scans for optimal CTI performance, and 55FE x-rays to measure charge diffusion. Filters and a monochrometer control the wavelength of the light for quantum efficiency measurements. All our instruments are under GPIB control and monitors, which allows a complete automation of the CCD testing. Characterization of a device corresponds to ~700 images and approximately 14 hours. We test all the technical specifications.

6’’ Integrating sphere

Monochrometer

Filter wheel

Halogen lamp

Readout cable connected to Monsoon controller.

LN2 Fill Pipe

To Vacuum Pump

The PCB in the dewar connector has a preamplifier that enables us to readout with low noise (<10e RMS) and @ 4 µsec/pixel. The Monsoon system will used to readout the survey instrument.

Shutter

The “Multi-CCD Test Vessel” is a full-sized prototype ofthe camera that allows us to perform systems integrationtests of the electronics, cooling, camera controls andCCDs as we design and build our equipment.

Examples: focal plane flatness, heat load, CCD cross-talk & readout noise.

So far we have found 63+ that pass all requirements and are working on building up a supply of spares. Here we show some of their characteristics.

Cosmetics. The plot shows the fraction of hot pixels (excess dark current), dark or dead pixels (20% less response), pixels with traps deeper than 700 electrons, unstable pixels, or pixels that are > 1% non-linear. The mean is 0.13% per device. The best 60 have a mean of 0.076%.

Full Well. The Dark Energy Survey Calibrations group is already compiling and studying standard stars for photometric calibration. The full well constrains the maximum star brightness to griz magnitude > ~16 and Y-band magnitude > ~14.

Quantum Efficiency vs. λ. We measure the relative QE every 10 nm. The different science grade CCDs are indistinguishable. We measure the absolute QE to about 3% using calibrated photodiodes.We also study QE vs. T(oK) as we warm up each CCD.The plot is absolute QE for three different CCDs.

Readout Noise. One CCD at a time in the test “Cube” using Monsoon electronics @ readout speed 250 kpixels/s (right). The mean is 8.6 electrons.

CCD Flatness Measurement

(µm)

Flatness Specifications (at T=172K):• Height difference between adjacent 1 cm2 subsurfaces < 10 µm (overall shape)

• Within a 1 cm2 subsurface there are no devations > 3 µm (small bumps).•This is the result (topographical map) from a typical device.

(µm)

T=294K T=172K

(cm) (cm)

Imager & CCD Packaging and Testing TeamH. T. Diehl1, T. Abbott2, B. Angstadt1, M. Bonatti2, L. Buckley-Geer1, J. Campa3, H. Cease1, G. Derylo1, J. Estrada1, B. Flaugher1, G. Gillespie1, J. Hao1, S. Holland4, M. Jonas1, W. Kolbe4, J. Krider1, D. Kubik1, S. Kuhlmann5, K. Kuk1, K.W. Merritt1, N. Palaio4, N. Roe4, V. Scarpine1, R. Schmidt2, L. Scott1, T. Shaw1, K. Shultz1, H. Spinka5, W. Stuermer1, J. Tweed1 and W. Wester1. 1Fermi National Accelerator Laboratory, 2Cerro Tololo Inter-American Observatory, 3Ctr. de Investigaciones Energeticas, Medioambientales y Tecnologicas, Madrid, Spain, 5Lawrence Berkeley National Laboratory, 6Argonne National Laboratory.

“Science Grade” Detectors

Integration Platform 2 – “Telescope Simulator”

DECam systems will be commissioned at Fermilab prior to shipping the camera to the telescope. The telescope simulator will allow us to test all systems in any physical orientation.

“Pedestal” (Focal Plane) Package

Anti-Reflective Coating

Requirements for DECam CCDsPixel array 2048 4096 pixelsPixel size 15 µm x 15 µm # Outputs 2QE(g,r,i,z) 60%, 75%, 60%, 65%

QE Instability <0.3% in 12-18 hrsQE Uniformity in

focal plane <5% in 12-18 hrs

Full well capacity >130,000 e-

Dark current <~25 e-/hr/pixelPersistence Erase mechanismRead noise < 15 e- @ 250kpix/s

Charge Transfer Inefficiency <10-5

Charge diffusion 1D σ < 7.5 µm

Cosmetic Requirements

<# Bad pixels><0.5%

None worse than 2.5%Linearity 1%

Package Flatness 10 microns. See Below

Prototype Imager & Test Dewar

Production Imager Status

Au-platedInvar-36 foot

Shorting plug insertedinto AirBorne connector

Alignment pinwith threaded hole

Aluminum nitride spacer

Aluminum nitridereadout board

Back-side-illuminated 2kx4k

CCD

Wirebonds

Threaded holefor module insertion

to focal plane

We measure the focal plane flatness through the fused-silica window while the focal plane is at operatingtemperature. The requirement is that entire focal planebe within a 60 micron envelope. Above we show theresult: better than a 45 micron envelope with σ = 9.2 µm.

S3-70

S3-81S3-61

S3-102S3-153S3-76

S3-43S3-51S3-44

S3-47S3-172S3-126S3-04

S3-77 S3-85S3-82S3-67

S3-28? S3-94 S3-79 S3-39

S3-35S3-18S3-16

S3-73S3-62S3-07

Data minus BFPS3-80

µm

mm

mm

Back of Production Imager (above). New aluminumcold plate (right). The part of the cold plate betweenCCDs will be painted black Aeroglaze to reducereflected light.

Turbo Pump(ion pump is mostlycut away in this view)

VacuumGauge

Insulated Electronics Crates

PositivePressureRelief

Feedthrough(one of two)

Vessel Shell (stainless steel)

LN2 Piping

CCDs andthe Focal PlaneSupport (Cold) Plate

Vacuum Interface BoardsFeedthrough electrical signals

Front Window (fused silica)is last optical element

Optical referencesurface

Alignment Screws

0

2

4

6

8

10

12

14

16

18

0.01 0.02 0.05 0.1 0.2 0.5 1 2 5

% Bad Pixels

1.45%

0

5

10

15

20

25

30

5 6 7 8 9 10 11 12 13 14 15 16 17

Readout Noise (electrons)

Production Imager Design

0

5

10

15

20

25

30

35

145000 155000 165000 175000 185000 195000 205000 215000 225000

Full Well (electrons)