E C-GEn. Overview

Transcription

1 E C-GEn Overview Brad Sawatzky for C-GEN collaboration (Slide Credits to: Arrington, Kohl, Semenov, Tireman, et al.) 1

2 Major Responsibilities Target JLab Dipole magnets JLab SHMS JLab Shield Hut / Stands JLab Electronics JLab Simulations / Shielding design U. of Regina (A. Semenov) N. Michigan (W. Tireman) JLab (Pavel?) NPOL Hampton, JLab, SUNO (M. Elaasar), N. Michigan (W. Tireman) La Tech (S. Wells) W & M (C. Perdrisat) NSU (V. Punjabi) Vetos (28 1x10 cm2) N. Carolina A&T (A. Ahmidouch) de bars ( 52 1x10 cm2) Analyzers ( 28 10x10 cm2) E bars (108 10x10 cm2) Analyzer Software 2

3 Overview 3

4 Recoil Polarimetry Technique 4

5 Precession Magnet Requirements where, g GE/GM Field serves two functions precess the neutron spin to maximize detected asym (low or high field OK, but 'medium' == no good) suppress charged backgrounds from target (need high-field) Optimal B dl : 4.3 T m We're shooting for 4.0 T m 5

6 Polarimeter: PAC37 version 6

7 Improving the acceptance 7

8 Updated neutron polarimeter 8

9 Optimal use of analyzing power 9

10 Good timing for clean QE event selection 10

11 Charybdis + BNL 48D48 Will the two magnets fit? BNL 48D48 Yes! B dl of 2.0 T 2 ka Aperture 122 cm high 47 cm wide (w/ shims: 31.8 cm) 122 cm along z-axis Pole center: = 28, R = 4.64m Back side / Shield wall: R = 5.5 m Charybdis B dl of 2.15 T 590 A (E93-038) Aperture 21 cm high (adj) 56 cm wide 1.22 m along z-axis Pole center: = 28, R = 2.5m 11

12 Charybdis + BNL 48D48 R=5.5 m R=4.64 m 50 cm gap (arb.) between coil packs R=2.49 m R=1.66 m Field clamp R=1.5m (not shown) 12

13 Space for Plan B Polarimeter Rough outer boundary of Polarimeter detectors is ~3m x 3m Shield house for readout electronics needed as well 13

14 Statistical Impact of Moving NPOL Initial proposal had NPOL at R = 5m Can not achieve desired B dl with iron-core magnet(s) in ~3m space between NPOL and target. Updated dual-magnet layout requires NPOL at R=7m Note: 5m/7m = 0.71 (5m/7m)^2 = 0.51 No ba t too d! Q MCEEP 5m (Hz) MCEEP 7m (Hz) Ratio (Rate@7/Rate@5) /SQRT(Ratio)~Stat Uncert Ratio is better than naïve solid angle scaling due to impact of QE cross section convolution and SHMS acceptance. 14

15 Kinematics, beam time allocation PAC 37 Request: 60 days PAC 41 Allocation: 50 days 15

16 DAQ / Readout Plans 16

17 Trigger Options Read out everything for all SHMS triggers (simple option) SHMS trigger will be formed by 'conventional' NIM FPGA trig available as an option Proposal suggests max SHMS trigger rate < 1kHz (verify!) Data rate and storage requirements are small by present standards SHMS DAQ can handle 2 3 khz readout rates with no effort 4 5 khz is achievable with minimal work Will want NPOL singles trigger too, but life is a little simpler if we do not require hardware NPOL+SHMS coincidence trigger Trigger-level NPOL+SHMS coin. is still possible though NPOL trigger would be formed in F250 firmware» removes need for a lot of NIM hardware need to watch lead time on firmware development/debugging 17

18 Channel Count for Plan B 216 scintillator bars 108 (10x10 cm2) E counters (top + bottom) 52 (1x10 cm2) de vetos (top + bottom) 28 (10x10 cm2) analyzer bars very high rates in forward analyzers 28 (1x10 cm2) analyzer vetos very high rates, need good timing & short pulse widths All bars double-ended readout 432 PMTs High res timing: R+L (offline) coincidence on analyzer bars + vetos (56 bars, 112 ch) BG suppression position information (100 ps hardware timing resolution assumed) Coarser timing (~ 1ns) on de/e bars Energy/ADC: (160 bars, 320 ch) de/e on upper/lower bars, walk-correction on TDC timing NPOL readout electronics likely located in the Hall not enough patch cables to run upstairs 18

19 NPOL Readout Options: TDC Proposal Requirements: CFDs on mean-time scatterer detectors/analyzers (100 ps res.) 56 ch (2 x 28 bars) CFDs on analyzer vetos (100 ps res.) 56 ch (2 x 28 bars) FADCs for top+bot E detectors (~ 1 ns res.) 216 ch (2 x 108 bars) FADCs for top+bot de detectors (??) (~ 1 ns res.) 104 ch (2 x 52 bars) CAEN v1190 VME TDCs will be used in SHMS, HMS (replacing FB1877, 1875s), plenty of channels available Resolution: 100 ps/bin Need splitters + discriminators for 112 ch Can we afford passive split, or is active required (presume we need active?) HMS: active splitters: CFDs: located 15x Lecroy 3420 CAMAC CFDs (16 ch/ea) 240 ch Fast Amps: Kent State (???) ~120 ch 90:10 split; 90% sent to TDC, 10% amplified 10x and sent to ADC 96 ch (16 ch x 6 rack units) supposed to be on-site, but have not located them yet... FADCs can provide timing to 1 ns or better on E/dE bars (JLab) 320 ch 19

20 NPOL Readout Options: ADC Combined channel count: 2x 28x 432 channels VXS crates F250 modules JLab F250 Flash-ADCs Will be present in both HMS and SHMS (replacing legacy devices) FPGA-based / programmable: several data output options available total charge (QDC equivalent) full digitized pulse profile (testing/diagnostic use) timing data (4 ns internal sample clock, interpolation shown to provide ~1ns res. or better) pipelined device: no ADC delay lines required! able to produce a neutron-arm trigger in firmware (if desired) can replace a large fraction of frontend NIM logic Can strip HMS for 208 channels (13 modules + 1 VXS crate) Need: 15 additional FADC + 1 VXS crate JLab Physics Div. has purchased FPGA chips for future F250 production run best option, should/will happen, but production run not scheduled yet(?) (~15 spares/v1 modules also available now, but should not be necessary) 20

21 Simulations 21

22 Original Simulation FLUKA / MCEEP / GENGEN 22

23 Estimates from Proposal Old polarimeter design, NPOL closer to target than in present setup Incident NPOL flux estimated using DINREG 23

24 New GEANT4 Simulation Will Tireman (Northern Michigan) and Danial Wilbern have been taking the lead on this. GEANT4 analyzer goals: Cross check FLUKA/MCEEP estimates Study backgrounds Develop shielding Optimize / Improve on FoM estimate Explore new polarimeter configurations to take advantage of the charge-exchange channel 24

25 Simulation Setup Hall Shell includes roof Can be cut out at any point Currently tracks killed that hit the floor, wall, roof Kill tracks that get to the end of the beamline 25 25

26 Experimental Setup Full shield house Constructed with steel and concrete blocks Pulled data on scattering chamber and Horizontal Bender from other Hall C sims and Brad Sawatzky 26 26

27 Downstream Beamline Model with detail provided by Hall C Preliminary beamline from a basic Hall C presentation Best guess on a bunch of it Flanges, gate valve, support block, punched hole in Charybdis field clamp 27 27

28 View of Magnets and Lead Curtain Downstream beamline positioning relative to Dipoles Horizontal Bender Lead Curtain Scattering Chamber with 40 cm LD2 Target 28 28

29 Backside of the polarimeter with the field clamp of Dipole 2 Plan B Layout Shield Hut not shown Relative position of detectors has been estimated 29 29

30 Polarimeter only Exact position of detectors is approximate Center of first layer is at 7.0 meters from Target center All other distances are relative, pulled from proposal/presentations Need full engineering 30 30

31 Overhead view 100 electrons on target 31 31

32 Current Status Running on Jlab batch farm with 4.4 GeV electrons Using 3 magnet setngs: 1, 2, and 4 T m integrate fields Lead thickness of 0, 5 and 10 cm Finished running the 4 T m farm jobs Writing ROOT scripts to assemble the results 32 32

33 VERY Preliminary Results Total Events at 28.0o E = 4.4 GeV 40 cm LD2 Target Bdl = 4 T*m 10 cm Lead Curtain 33 33

34 DINREG plot from Original Proposal Qualitatively similar (note the auto-scaled y-axes on the G4 slide) Significant differences in low-energy gamma yields and in neutron energy spectrum curious... G4 work is very preliminary, need time to dig into it. It is worth noting that Pavel D. disavowed using the 6 GeV DINREG code for higher energy work. He tells me a G4 model has (tentatively) been replaced by a future FLUKA model, however neither is in a useful form yet. 34

35 Simulations (side by side) GEANT4 DINREG 35

36 Future Plans for Simulation Finish up runs with the other magnet setngs and lead thickness Analyze rate questions Investigate different shielding options Setup/run for a charge exchange analysis (Freshman currently working on ROOT script) Modify the polarimeter with a highly segmented front array as per Michael Kohl s suggestion 36 36

37 Near Term To-Do List (partial...) GEANT4 analyzer development Cross check FLUKA/MCEEP estimates Study backgrounds Develop shielding Optimize / Improve on FoM estimate Settle on a optimal Polarimeter design Charge-Exchange study Complete and submit MRI Evaluate Charge-Exchange contribution, study polarimeter modifications Existing polarimeter can already detect forward protons, offsetting veto vs. analyzer bars improves y-coord resolution. Semi-hermetic polarimeter will already allow us to extract ChEx proton from E/QE. FoM boost is an open question... Requires simulation Several modifications have been discussed to enhance ChEx contribution to FoM Requires simulation 37

38 Misc. Backup 38

39 Target Notes Target requirements in proposal: (~ 500 W, should be fine) 15 cm LH2 cell at 80 A Should select one of the new cells: 10 cm or 20 cm avail. (~1500 W, non-trivial!) 40 cm LD2 cell at 80 A high power target, non-trivial cryo requirements LD2 needs on the edge, but just doable with ESR-I's 15K supply Silviu's new cell model can likely be adapted to 40 cm cell (~ $15k + development time) probably a long lead item Hall C Standard 10, 20 cm cells (under development) Goal: < 1% density loss for 20 cm LH2 100 A w/ 2mm raster 39

40 Kinematics 40