The Injector Design C.Y. Tan 01 Sep 2011 1
The Injector BNL (2009 line) FNAL Proposed 1 x 35 keV H- source (round 2 x 35 keV H- sources (round ● ● type) type) on a slide LEBT (~ 4 m) For redundancy ● ● Short LEBT (~118 cm) 2 x solenoids ● ● Xe gas for neutralization 2 solenoids for focusing ● ● 1 Einzel lens Xe gas for neutralization ● ● 1 electrostatic chopper. Einzel lens chopper ● ● RFQ 201.25 MHz, 1.5 m long RFQ 201.25 MHz, 1.2 m long ● ● Short MEBT (0.73 m) Short MEBT (1 m) ● ● ● 2 sets of doublets 1 set of triplets ● 1 buncher ● 1 BNL style buncher ● 2
H- Source ● BNL style round (dimpled) magnetron source. ● More discussion in D. Bollinger's talk. ● BNL has very good experience with this type of sources. Very reliable. ● FNAL has a lot of experience with slit type magnetron sources. Easier transition to round type. ● Good choice because it can use the same type of hardware FNAL already has. 3
LEBT 46.55” 4
LEBT (cont'd) ● BNL source emit- tance numbers ● RFQ input numbers from manufacturer ● Assume beam is neutralized. Edge of beam pipe 5
Solenoid Settings in Trace2D Focal length is consistent with as built solenoid set to 500 A. 6
FNAL Solenoids Solenoids run DC. Water cooled. Settings probably > 500A. Designed by A. Makarov and V. Kashikhin. 7
B field Measurements 8
Small x-offset (~ 0.2 mm) in position and angle when current is changed from 500 A to 600 A. Do not expect to work below 500A. No change in y. Should be correctable with dipole correctors. 9
Neutralization using N2 in test stand N2 used. Good vacuum 1e-6 torr. Bad vacuum 1e-4 torr. Note: pressure gauge quite far from beam in test stand. Faraday cup measurement 10
Chopper ● Einzel lens is used as a chopper ● Suggested by D. Raparia ● Lens placed close to the entrance of the RFQ – Keep de-neutralization region to a minimum. – Beam is already tightly focused by the solenoid at this location. So neutralization is probably at a minimum here. ● Einzel lens as chopper experiments in the test stand have verified that method works. 11
Einzel Lens Chopper ● Simulation using SIMION ● Optimized lens ● 2” long ● 1.75” diameter ● -37 kV to stop 35keV beam. 12
Einzel Lens as Built Designed by A. Makarov 13
Chopper Timings 14
Einzel Lens Test F. Cup Toroid Note there is scraping at the beginning of the pulse in the test stand. This is a different Einzel lens than the one that we will be using. This Einzel lens is 2” in diameter and 2.5” long. 15 Note: injector lens is 1.75” diameter and 2” long.
Einzel Lens Test (cont'd) Rise time on Faraday cup is bet- ter than 50ns. Con- sistent with thyrat- ron turn on of 35 ns. Ringing from finite bw of Faraday cup. 16
LEBT Corrector Dipoles Designed to fit over 4” beam pipe. Corrects drop in integrated Bdl when placed close to iron. 10A corrects 1 deg. H+V in one < 1.5” package. 17
RFQ Designed by A. Schempp. Tuned by J. Schmidt and B. Koubek. Rod type RFQ. > 18
19
As found RFQ parameters. Note frequency is 190 kHz too low.To be cor- rected with 20 tuner.
PARMTEQM Simulations Transmission effi- ciency > 99% for 60mA beam. 21
PARMTEQM Simulations (cont'd) Input Output 22
MEBT Doublet – Buncher – Doublet (30 cm longer compared to BNL MEBT) 23
MEBT Simulations PARMILA shows 95.1% of the beam is captured at the end of DTL 1. Note: Lattice of DTL not well understood! Using “as found” DTL lattice from M. Popovic. Note: Quads are very strong 24
Before doublet Before Tank 1 After Buncher End of Tank 1 25
Buncher Same as BNL (except FNAL buncher is made of copper while BNL buncher is made of aluminium) Designed by M. Okamura. Must have grids to get transit time factor correct. Low power test without grids are complete at FNAL. With grid has been done in Japan. 26
Grids Grids to correct transit time factor. Picture is BNL grids. 27
Bead Pull Data FNAL With grid effectively re- duces gap or increases transit time by 40% Japan Full width half height ~ 30mm gap. 28
Doublets (Quads with embedded corrector dipoles) Quadrupole runs DC and water cooled. Iron core. Has embedded corrector dipoles. Overall length of doublet is 7”, 1.5” beam pipe. 29
Quadrupole Design 400A and 11 turns per pole Field inhomogeneity is 1.7% @ 1 cm with F/D at 4.4kA/pole. Doublet Single quad. 4.4 kA/pole 4.4 kA/pole 30
Sextupole Component from Dipole The dipole contributes a sextupole component of Sint = 0.564 10 -4 Tm when powered to de- flected beam 0.2 deg. (gives 1mm deflection from last corrector to the DTL, Bdl = 0.45 10-3 Tm) If integrated sextupole field A3 < 0.5% of integ- rated quadrupole field, transverse emittance Integrated vertical dipole field at centre blow up < 1%. And 1 cm 31
Coupling If quad random rolls < |0.5| deg for all 4 quadrupoles, emittance growth is < 1% at the start of DTL 1. 32
Performance Goals At minimum, must perform as well as present system. 33
Conclusion ● Nothing in the design is technically unachievable. ● Not pushing any envelopes! ● However, there are some concerns. See “Outstanding Issues and Contingencies Talk” ● Copying as much as possible BNL design has helped the injector design. ● There will be a lot of work to validate design in the test set up. 34
Acknowledgements ● D. Raparia, J. Alessi, M. Okamura (BNL) Generous hosts when we visited. ● Shared drawings of solenoid and quadrupoles. ● Allowed us to copy BNL buncher (Okamura) ● – Fixes to BNL buncher that were incorporated into FNAL buncher. Discussions with D. Raparia ● – Supplied initial input files to Trace, Parmila and Parmteqm. ● A. Makarov, V. Kashikhin, G. Velev (FNAL, TD) Designed and built solenoids, einzel lens and quads. ● ● All the personnel in Proton source, Mech. Support, EE Support. 35
Backup Slides 36
Test Stand Einzel lens is 2” in diameter and 2.5” long. Note: injector lens is 1.75” diameter and 2” long. 37
38
Recommend
More recommend
