The Streak Camera Development Programat the Laboratory for Laser Energetics

P. A. Jaanimagi, R. Boni, D. ButlerS. Ghosh, W. R. Donaldson, and R. L. KeckLaboratory for Laser EnergeticsUniversity of Rochester

26th International Congresson High Speed

Photography and PhotonicsAlexandria, VA

20–24 September 2004

E13229

The ROSS camera is a comprehensive diagnostic systemwith auto-focusing and self-calibration capability

Summary

• System performance is limited by the streak tube electron optics

– currently implemented with P510, P820, and PJX tubes

• All functions can be accessed and monitored remotely

• The streak camera can be photometrically calibrated

• 8 prototype cameras have operated on OMEGA for 5 years,accumulating over 500,000 streaks, with better than 99.9% reliability

E13230

Outline

• Motivate why we are doing it

• What we are trying to accomplish

• Describe ROSS

• Calibrations

TC2998a

The OMEGA laser is designed to achieve high uniformitywith flexible pulse-shaping capability

• 60 beams• 1%–2% irradiation nonuniformity• Flexible pulse shaping• Short shot cycle (1 h)

E10902

P510 ten-channel streak cameras provide pulseshape and power measurements on OMEGA

69686766656463626160

–1 0 1 2 3Time (ns)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

UV

po

wer

(T

W)

22707

fidu

61

fidu

62

63

64

65

66

67

68

69

60

335 J

327 J

283 J

319 J

317 J

318 J

308 J

306 J

302 J

311 J

E13266

OMEGA experiments demand quantitive, high-precision (SNR > 100), large dynamic range (>1000)high-bandwidth (10 GHz) measurements

• Quantitative– nonlinearity <1%– scientific-grade CCD recording without intensifiers

• 1% measurements– long term stability– calibrate and then calibrate some more

• Integral of streak record ∝ energy

• Stand-alone system, full remote operation

• Streak camera bandwidth and dynamic rangeare still better than oscilloscopes and diodes

– streak tube limited system

E13261

The most important streak tube specificationis its current handling capability

• Imax gives us the number of photoelectrons per time and spatial-resolution element and thereby the peak SNR

• Photoelectrons follow Poisson statistics ⇒ SNR ∝ Signal1/2

• Single photoelectron events are recorded well above the noise floor

• A high current tube must have

– a high accelerating field at the photocathode

– a large area photocathode

– good photoelectron throughput (fraction that reaches the screen)

• Typically, 100–200 spatial and 100–200 temporal resolution elements

• Generally, large current is commensurate with good time resolution

E13257

The ROSS camera shown without its optical/calibration module has all its fiber-optic andelectronic services through the back panel

E13267

The ROSS camera is a flexible, modular, stand-alone,electro-opto-mechanical system

• Camera dimensions are 7-in. wide × 21-in. long × 12-in. highwith optics module 10-in. wide × 26-in. long × 12-in. high

• Streak tube is potted in a closed-end mu-metal shell

– accepts P510, P820, and PJX streak tubes

• Modular electronics are packaged into mu-metal boxes

• Closed-loop voltage control (0.02%) with 24-bit ADC/16-bit DAC

• 4 remotely selectable streak speeds

– 30 ms hold-off for retrace

• 1-ms-resolution timer for shutter and CCD exposure control

E13258

The input imaging system is an Offner triplet withmotorized controls for the dual object planes andthe secondary mirror

Primarymirror

Streak tubewindow

Field-flatteninglens

Homogenizerarrays

Fiber-opticinputs

Secondarymirrorwith Iris

Reticlearray

Flip-inmirror

Air path input

E9815a

Solid glass homogenizer bars – rectangular 4-mirrorkaleidoscopes – are used to form uniform illuminationprofiles that are relayed to the photocathode

Multichannel Input Flat-Field Input

Signal froma launcher

Multiple fiberdelivery

Homogenizerbar

Channel outputimaged onto thephotocathode

Illuminationinput

Multiple fiberdelivery

Homogenizer bar

Output imagedonto thephotocathode1/2”

E13265

The calculated Offner triplet MTF is better thanthe streak tubes’ electron-optics MTF

Streak tube Photocathode Primary MTF @ F/2.5 MTF @ F/5.0

P510 22.0 mm R = 140 mm 79% @ 10 lp/mm 91% @ 10 lp/mm60% @ 15 lp/mm 84% @ 15 lp/mm

P820 10.0 mm R = 140 mm 77% @ 20 lp/mm 91% @ 20 lp/mm52% @ 40 lp/mm 81% @ 40 lp/mm

PJX-std. 60.0 mm R = 160 mm 63% @ 5 lp/mm 91% @ 5 lp/mm28% @ 8 lp/mm 79% @ 8 lp/mm

PJX-inv. 6.0 mm R = 160 mm 79% @ 40 lp/mm 84% @ 40 lp/mm69% @ 60 lp/mm 75% @ 60 lp/mm

Nu

mb

er o

f o

ccu

rren

ces

Var

ian

ce (×

104 )

105

104

103

102

101

100

6

4

2

00 2 4 6–40 –20 0 20 40

Mean (ADU)(× 104)Pixel value (ADU)

Subtract 2 dark fields

Read noise is Gaussian

Subtract 2 light fields

Variance � Mean forPoisson statistics

E12920

One must first calibrate the CCDcamera recording system

Spectral Instruments 800 seriesE2V 42-40 BI, 2 k × 2 k, 13.5 µm pixelsReadout 200 kHz, σ = 4.77 electronsDark current 0.025 e–/pix/s at –40° C

Gain1 = 0.98 e–/ADU

Gain2 = 2.35 e–/ADU

σ = 6.75 ADU

Full well

E13263

The imaging quality of a fiber-coupled CCD cameramay be compromised by a poor fiber stub

Sig

nal

(A

DU

)

100

10–1

10–2

10–3

10–4

10–5

10–6–200 0 200

Pixels

E2V 42-40 CCD13.5 µm pixels

ESF

LSF

PSF

E13259

The Spectral Instruments CCD camera with a 1:1 fiberstub (Incom BPLSE-6) can generate a sharp ESF

Sig

nal

(A

DU

)

106

105

104

103

102

101

100360 380 400 420

X (pixels)

150 µm

104

E2V 42-40 CCD13.5 µm pixels

E12924

The pulse height distribution for recording singlephotoelectron events is fit to D1(S) = S × exp [–(S-µ)2/2σ2]

Nu

mb

er o

f o

ccu

rren

ces

105

104

103

102

101

100–200 0 200 400 600

Signal (CCD electrons)

UV on Al photocathode

Gain = 150±1 CCD electrons/15 keV photoelectron

SNR1 = 2.4F2 = 1.170 electron

events � N0

1 electronevents

D1 � N0

2 electronevents from overlap

D1 � D1 � N0

E13268

The ROSS camera has a comprehensiveself-calibration capability

• Flat-field data is generated with internal sources and slow (1- to10-s) ramps

– CW-LED or white light with interference filter– DC-DC converter driven by DAC with variable time

and amplitude steps

• Geometric distortion data is also generated with slow ramps– reticle over flat-field homogenizer– modulate the LED duty cycle

• Sweep-speed time-calibration with on-board comb generator– 4 frequencies: 100 MHz to 10 GHz

• Auto-focus the input optics and the electron optics

E13264

Slow ramp, geometric distortion data shows theannulus of best focus for the P510 streak tube

Time →

Space

Annulus is theintersection ofa curved imageplane with aflat screen

E13260

The temporal resolution of the P510 tubeis optimized with an auto-focus routine

16

14

12

10

8

6

LS

F F

WH

M (

pix

els)

–14.48 –14.44 –14.40 –14.36 –14.32

Focus voltage (kV)

e

d c

b

a

E13262

The P820 has three lenses, allowing temporal andspatial focus to be controlled independently

Temporal lens(controls FWHM)

Spatial lens(vertical contrast)

E13229

The ROSS camera is a comprehensive diagnostic systemwith auto-focusing and self-calibration capability

Summary/Conclusions

• System performance is limited by the streak tube electron optics

– currently implemented with P510, P820, and PJX tubes

• All functions can be accessed and monitored remotely

• The streak camera can be photometrically calibrated

• 8 prototype cameras have operated on OMEGA for 5 years,accumulating over 500,000 streaks, with better than 99.9% reliability