Presentation is loading. Please wait.

Presentation is loading. Please wait.

Photon Systems Breakout Lehman Review July 11, 2007

Similar presentations


Presentation on theme: "Photon Systems Breakout Lehman Review July 11, 2007"— Presentation transcript:

1 Photon Systems Breakout Lehman Review July 11, 2007
XTOD Diagnostics Photon Systems Breakout Lehman Review July 11, 2007

2 FEE (Front-End-Enclosure)
LCLS Layout FEE (Front-End-Enclosure) Diagnostics SOMS HOMS Linac Undulator Hall X-Ray Tunnel FEH (Far Experimental Hall) NEH (Near Experimental Hall)

3 XTOD Commissioning Diagnostics and Offset Mirrors in the Front End Enclosure (FEE)
Wall penetration K Spectrometer Indirect Imager Slit Solid Attenuators Collimators Wall penetration Thermal Sensor Fixed Mask Gas Attenuator Gas Detector Beam Direction Gas Detector Direct Imager (Scintillator) FEL Offset Mirror Systems

4 XTOD Optics and Diagnostics in FEE
5 mm diameter collimators Gas Detector Soft X-Ray Offset mirror system Hard x-ray Monochrometer (K Spectrometer) Direct Imager Solid Attenuator NFOV Slit e- Indirect Imager Total Energy Thermal Detector Hard X-Ray Offset mirror system Gas Attenuator WFOV Gas Detector Start of Experimental Hutches Muon Shield

5 Attenuator on order N2 boil-off (surface) Flow restrictor
Gas detector Flow restrictor 4.5 meter long, high pressure N2 section Differential pumping sections separated by 3 mm apertures 3 mm diameter holes in B4C disks allow 880 mm (FWHM), 827 eV FEL to pass unobstructed Solid attenuators N2 Gas inlet Green line carries exhaust to surface Gas detector

6 Gas detectors share differential pumping with the Gas Attenuator
PMT Electronics Photo Multiplier Tube Bandpass Filter Al liner 3 mm apertures along beam path Differential Pumping Section Magnet Coils Cylindrical Vessel Beam / Gas Interaction Region (~0.1 – 2 Torr N2) Magnet Power Supply and Controller APD Gas Feed And Pressure Control LCLS X rays cause N2 molecules to fluoresce in the near UV

7 Gas detector SSRL prototype
Port for pumping Photo Multiplier Tube Magnet Coils Be window Gas Feed And Pressure Control Avalanche Photodiode Port for LED illuminator and florescence samples

8 Prototype Gas Detector insert for measuring x ray induced photoemission of candidate wall materials

9 Gas Detector signal vs. magnetic field at various pressures
Simulated Measured

10 Results of SSRL Gas Experiments: Comparison Model and Experiments
(2X correction of calculated signal) We suspect that discrepancy at lower pressures is due to photo electrons (and secondaries?) hitting the chamber ends

11 Measured luminescence of solids at 8 keV
Al UV signal closely represented by 9:00, B on (red): Al is the best

12 Time dependence of gas detector signal from the 8keV fundamental
UV signal within X rays scattered into walls ~ 1 ns X rays scattered into detector window ? Photoelectrons hitting walls 0 – 18 ns Photoelectrons hitting end caps 0 – 15 ns Secondaries hitting walls and end caps 0 – 200 ns (?) Energy of photoelectrons deposited into N2 0 – 45 ns

13 Indirect imager finds spontaneous core
Raw soft spontaneous After reflection Princeton Instruments back illuminated CCD camera 25 x 25 mm chip, 20 um pixel size ML mirror 0.1% reflectivity, 1% bandwidth Status Indirect Imager: PRD in progress Vacuum chamber

14 Detector monochrometer measures intensity at a single point
Channel-cut Si Monochrometer will be used to measure relative K of two undulator segments Monochrometer Two undulator spontaneous spectrum. Falloff of high energy tail is independent of aperture Detector Detector monochrometer measures intensity at a single point Linac E variation and measurement Use linac E variation and measurement to obtain other points along curve Two undulator spontaneous high energy falloff has highest slope when DK/K=0. Status K Spectrometer: PRD in progress

15 Total Energy (Thermal) Sensor provides calibrated measurement of FEL pulse energy
Measures FEL energy deposition through temperature rise Cu heat sink Thermistors Nd0.66Sr0.33MnO3 (On back of substrate) Sensor Temperature Rise 1 2 3 4 5 0.05 0.1 0.15 0.2 0.25 0.3 Time [ms] Tc200 Tc150 Tc100 FEL pulse [K] 0.5 mm Si substrate t = 0 t = 0.1 ms t = 0.25 ms Thermal diffusion of FEL energy Status Thermal Sensor: PRD done SCR done PDR done Prototype done FDR in progress

16 Thermal sensor prototype
Pulse tube cooler Optics for measuring, focusing, and steering laser beam 532 nm laser Sensor cryostat

17 Thermal sensor signal at 1mJ
2 volt bias 0 volt bias V1 V1 Vbias Rsensor R1 V1 V2 R3 R4 V2 V2

18 Channel 1 pulse at 400 uJ with 2V bias
Subtract value at -100us Sample voltages at particular times, convert to DRcmr then to DT

19 Total energy prototype measured and predicted signal
Finite difference prediction DT, °K DT, °K Measured data Absorbed Laser Energy, mJ Absorbed Laser Energy, mJ At 100 msec At 3 msec

20 Thermal sensor, preliminary design
Thermal Detector Calibration Laser Laser Energy Meter

21 Scintillators on movable shaft Photodiode for K measurement
Direct Imager 30 fps CCD Camera Scintillators on movable shaft ND filter wheel High resolution lens 4 Scintillators: Thin, stops 8 keV FEL find soft x-ray FEL Very thin, stops 826 eV FEL Thick, stops spontaneous spontaneous studies + one other for tests diamond Stops FEL Transmits spontaneous Photodiode for K measurement 30 fps CCD Camera Low resolution lens

22 Direct Imager, preliminary design
Single shot measurement of f(x,y), x, y ,u Camera Status Direct Imager: PRD done SCR done Prototype done PDR done in Final Design Scintillators

23 Scintillator emission spectrum
Camera QE Scintillator emission spectrum

24 Soft X-Ray Spontaneous signal in WFOV Direct Imager
Absorbed in 5 um YAG, Maximum ~ 20,000 photoelectrons/pixel Camera: Photometrics 512B Objective: Navitar Platinum 50 Power: NA:

25 Scintillator signals in FEL equivalents
Needed x-ray attenuation Needed x-ray attenuation YAG Range Needed visible attenuation CCD Range Need x-ray attenuation of > 100 and visible attenuation of > 10

26 Prototype Direct Imager testing
Flat field testing of camera only Enclosure CCD Integrating Sphere Camera Pulsed UV laser testing with YAG WFOV Optic UV beam splitter N2 Laser YAG Insertable photodiode Lens ND filter Photodiode

27 Photon Transfer Curve Our fit: Manufacturer:

28 Cascade 512B on-chip gain measurements
On-Chip: 0 to 4000, Exposure Time: 5 to 190 Lamps: 1, 2 Measured Gain = Slope g(j)/Slope g(0) On-Chip Gain Measured Gain Slope 45.09 500 62.44 1000 77.59 1500 106.06 2000 168.20 2500 307.83 3000 848.24 3500 1.00 1.38 1.72 2.35 3.73 6.83 18.81 60.39 February 12, 2007 Run 12

29 Direct Imager image of N2 laser excited YAG scintillator at 20 Hz
YAG excited by N2 laser Boundary of 10 mm dia. YAG disk

30 FEL Offset Mirror Systems have “Pop-in” imagers for alignment

31 Bidirectional Popup Viewer allows viewing of the x-ray beam and the collimator

32 We are studying expected signal levels in the Pop-in cameras
Low Energy, All undulator modules 100% Spontaneous Propagated through fixed mask, pipes 12 Boron Carbide windows are open Slit is open No attenuation, no gas detector Photons absorbed in 1 mm YAG, Full Well: 200,000 # photoelectrons ~ 5660 per pixel Misses Mirror 1 Reflected off Mirror 1 SOMS Run005

33 Progress continues on XTOD diagnostics:
Summary Progress continues on XTOD diagnostics: Procurement - Slit, Fixed Mask, Attenuator PDR – Direct Imager, SOMS, Thermal Detector SCR – K-Spectrometer PRD – HOMS, Indirect Imager


Download ppt "Photon Systems Breakout Lehman Review July 11, 2007"

Similar presentations


Ads by Google