Run2 Problem List (Bold-faced items are those the BP Department can work on) October 4, 2002

Transcription

1 Run2 Problem List (Bold-faced items are those the BP Department can work on) October 4, 2002 Linac Booster o e12 ppp at 0.5 Hz o Space charge (30% loss in the first 5 ms) o Main magnet field quality (measurement at the E4R) o Orbit error and skew sextupole field of the dogleg magnets o Coupled bunch instability after transition Longitudinal damper (being commissioned) h=83 and h=85 o Low p/p bunch production Bunch rotation o Longitudinal matching between Booster and MI o Collimator commissioning o MiniBooNE operation at 5 Hz (concern: beam loss and activation) MI-8 line o Trajectory and optics matching to the MI Main Injector o $29 stacking cycle: 1 Booster batch (84 bunches), 4.5e12 every 2.2 sec. ε L doubled during cycle ( ev-s) ε H increased by 20% (16π 20π, improved?) Mainly during transition crossing Barrier RF stacking or slip stacking o $2B p coalescing cycle: 7 bunches into 1, 28-32e10 each, 36 times. Coalescing efficiency 75-80% (12-15% dc beam, 8-10% satellite beam) ε L increased by 30% (?) (goal: 2 ev-s, actual: 4 ev-s) ε H increased by 20% o $2A pbar coalescing cycle: 4 batches of pbar, 7-11 bunches per batch, into 4 bunches, 396 ns spacing, 9 times Acceleration efficiency 99% Coalescing efficiency 70% (65-90%, depending on ε L ) ε L increased by 30-60% (affected by beam loading during coalescing) ε H increased by 20% o Beam loading compensation on 2.5 MHz and 53 MHz rf o pbar injection damper commissioning P1-P2-AP0-AP1-AP2/AP3 line o Long (900 m), three energy modes (8, 120, 150 GeV) o Mismatch to the MI lattice Debuncher o Momentum cooling reaches only 1/10 of the theoretical value Accumulator - 1 -

2 o ε H blowup: from 8π to 17π at 100e10 (resolved, see below) o Due to IBS: Lattice γ t increased from 5.5 (Run1) to 6.55 (Run2) (η from down to 0.012) leading to larger dispersion function o Dual lattice solution: Stacking lattice η = (at the core) and (at the extraction orbit); shot lattice η = (at the core) and? (not measured at the extraction orbit) o Trapped ions could be another main source o Problems with larger stack size Slower cooling rate (12 ma/hr at ma, down to 6 ma/hr at 100 ma) Larger transverse emittance o Low longitudinal phase space density ( large ε L ) Four 2.5 MHz (ARF4, h = 4, max 900 V) bunches per transfer, each bunch ev-s, total 8-12/14 e10 pbars, 9 transfers. Some recent shots measured 14 transfers, average ε L = 1.37 ev-s. At large densities, transverse emittance becomes large. o Frequency mismatch between Accumulator and MI (due to shift of the extraction orbit in the new lattice; dual lattice also helps solve this problem) o To shift pbar momentum by 40 MeV/c to match the Recycler lattice o Extraction: Four 2.5 MHz bunches per transfer, each bunch ev-s, 8-12/14 e10 pbars total, 9 transfers. o pbar transfer without coalescing A1 line o MI - Tevatron pbar transfer efficiency 88% (Note: New calibration makes it near 100%) o Mismatch between MI, A1 and Tevatron (which may cause large pbar emittance dilution, see below) Recycler o Low proton injection efficiency Mismatch due to feed down from sextupoles at the injection point Aperture limited at 30 π o Emittance blowup at injection due to mismatch. o Proton life time 70 hrs, limited by vacuum o Proton emittance growth rate 5 π/hr, due to residual gas scattering o Closed orbit affected by MI ramp. o pbar injection efficiency 80% (Goal: 75%) o pbar life time 100 hrs at 70e10 (Goal: 100 hrs at 50e10) o pbar extraction needs work o Stacking and cooling of high intensity beams (barrier rf, p/p measurement, beam loading) Tevatron at 150 GeV o Poor lifetime due to long range beam-beam pbar loss 20-25% p loss 10-15% o Large pbar emittance dilution - 2 -

3 ε H, ε V increased from 6-8 π (extraction from the MI) to π (in the Tevatron) o Longitudinal coupled bunch instability (bunch length blowup) Bunch-by-bunch damper o Longitudinal dancing bunches h=1112 and h=1114 o Slow head-tail instability (single bunch) Octupole commissioning h=1112 and h=1114 o New helix o Physical aperture limit at C0 from the old abort system. After its removal, the helix can be expanded. o Transverse damper commissioning Tevatron at ramp o pbar loss 10-15% o Control of orbit, tune, chromaticity and coupling Tevatron at low-β squeeze "step 13" o pbar loss was 15-25%, now 3-5% Tevatron at collision o DC beam o CDF and D0 background (improved) o Slow head-tail instability in vertical plane o Transverse emittance growth rate π mm-mrad/hr (both p and pbar, both h and v), higher than Run1 ( ) (to be checked, could be improved by solving the vacuum problem) o p intensity: goal 27e10 per bunch, achieved 20e10 o pbar intensity: goal 3.3e10 per bunch, achieved 1.3e10 o Overall pbar efficiency: In the Tevatron: 70% From Accumulator to HEP: 40% - 3 -

4 Table. Comparison of Run1 and Run2 Parameters (Sept 24, 2002) Parameter Run1 (6 6) MI TDH Run2a Goal (36 36) Run2a Achieved (36 36) L (peak) (1e31) 2.5 (record) 1.6 (typical) (record) (typical) L (weekly) (pb -1 ) p/bunch (1e10) (p total) 23 ( 6 = 140) ( 36 = 1000) 20 ( 36 = 700) p-bar/bunch (1e10) (p-bar total) 5.5 ( 6 = 33) (inj. to Tev 200) 3.3 ( 36 = 120) 1.3 ( 36 = 47) ε (π mm-mrad) 23 (p) 13 (p-bar) 20 (p) 15 (p-bar) (p) (p-bar) rms bunch length (ns) (p) 2 (p-bar) p on target (1e15/hr) (5e12 per 1.5 s) (4.5e12 per 2.2 s) p-bar stack rate (1e10/hr) p-bar stack size (1e10) (inj. to Tev 200) p-bar efficiency (%) 80 40

5

6 Tevatron Luminosity 1 of 2 10/9/2002 5:34 PM Tevatron Luminosity Luminosity Explanations

7 Tevatron Luminosity 2 of 2 10/9/2002 5:34 PM last modified 4/9/2002 Fermilab Security, Privacy, Legal

8 Figure 1. Proton and pbar intensity plot of Shot #1764.

9

10 f Proton Source Department Intensity and Energy Lost vs. Time at High Intensity CHG0 is Booster charge through acceleration cycle. BELOST is integral beam energy lost during acceleration NOTE: 300 joules at 5Hz over 470 meter circumference would be 3.2 Watt/meter

11 f Proton Source Department Beam Energy Lost in Ring vs. Per Pulse Intensity Beam Energy Lost During Acceleration 10/9/2000 Data (Notch off & excluding extraction) Kilojoules/Pulse Lost in Ring Protons/Pulse (E12) at 8GeV

12 Barrier RF Stacking Study 1. Goals: Near term - To increase proton intensity (protons per second, or pps) on the p-bar target by 50%: o Present: 4.5e12 protons every seconds (22 Booster cycles). o Stacking goal: 9e12 protons every 2 seconds (30 Booster cycles). Medium term - To increase proton intensity (protons per second, or pps) on the NuMI target by 60%: o MI baseline: 3e13 protons every seconds (28 Booster cycles). o Stacking goal: 6e13 protons every seconds (35 Booster cycles). 2. Status as of September 2002: Paper study is finished. Main results: a. Can stack 2 Booster batches into the size of one Booster batch in the MI for p-bar production. b. Can stack 12 Booster batches into the size of six Booster batches in the MI for NuMI. c. Longitudinal emittance blow-up is about a factor of 3 (from 0.1 ev-s to 0.32 ev-s). d. Key issue is that the Booster beam must have a small p/p to start with (required E about ±6 MeV). e. Compared with the slip stacking (another proposed scheme), one main advantage of the barrier rf stacking is smaller beam loading effects thanks to lower peak beam currents. Beam experiments and hardware are in preparation. 3. Experiments in the Recycler: Goal: To verify the simulation results. 1

13 Recycler proton beam: 4 bunches (2.5 MHz) per injection from the MI. Longitudinal emittance = 2.5 ev-s per bunch. Assuming parabolic distribution, this gives ±4.7 MeV in energy spread. Recycler wideband rf: 2 kv maximum. Two experiments: a. Beam squeezing at various barrier bucket moving speed. b. Stacking using off-momentum injected beams. Key parameter p/p measurement: a. Schottky spectrum: Doable for unbunched beams. Difficult for bunched beams. Our case is somewhat in between: The beam is debunched (no synchrotron oscillation), whereas there is a strong coherent gap signal. b. Bucket scanning. 4. Experiments in the Booster: Goal: To obtain small p/p beam at Booster extraction. Experiments: a. To suppress coupled bunch instabilities: Feedback - Longitudinal damper. (D. Wildman) Landau damping - To operate one rf cavity at h=83. b. To lower p/p of the beam at extraction: Bunch rotation. Adiabatic compression. 5. Hardware Development: 1) Barrier rf system for the MI: Goal: To build an 8-10 kv barrier rf system using Finemet cavities and high voltage fast switches. Cavity: Based on the design of an rf chopper that was built by a Fermilab-KEK collaboration through a US-Japan Accord. Hitachi Metals will supply the Finemet cores. 2

14 Switch circuit: Also based on the design of the rf chopper. Behlke will supply the switches (solid state HTS series). However, there is an important difference between the chopper circuit and the barrier rf circuit. The former uses a pair of +V and -V pulses. The latter requires a zero-voltage gap between +V and -V pulses. Therefore, the circuit needs to be modified. 2) MI rf modification: (J. Griffin) Goal: To reduce the beam loading effect by a factor of 3-4. Method: To lower the screen voltage from +2 kv to -100 V. 3

15 Barrier RF Stacking for NuMI 6-Batch Operation (B. Ng) 4

16 5

17 Barrier RF Stacking for p-bar 2-Batch Production (B. Ng) 6

18 Barrier RF Experiments in the Recycler (C. Bhat) 7

19 RF Chopper Cavity made of Finemet (Fermilab/KEK) 10

20 Comparison: Ferrite (4M2) vs. Finemet 11

21 $29 Cycle Intensity AAC Review May 13-15, 2002

22 $29 Longitudinal Emittance vs Energy AAC Review May 13-15, 2002 Energy (Gev)

23 $29 Horizontal Emittance vs Energy Energy (GeV) AAC Review May 13-15, 2002

24 $2B Longitudinal Emittance vs Time Time (sec) AAC Review May 13-15, 2002

25 Pbar Coalescing AAC Review May 13-15, 2002

26 Pbar Longitudinal Emittances vs Energy for one Store 2.5 T2 T3 T4 T5 T6 2 T7 T8 T AAC Review May Energy (GeV) 13-15, 2002

27 Pbar Longitudinal Emittance after Coalescing vs Batch Number AAC Review May 13-15, 2002 Batch number

28 Pbar Horizontal Emittance vs Energy for 5 Different Stores Energy (GeV) AAC Review May 13-15, 2002

29 Pbar Transfer Without Coalescing (Draft, 9/30/02) One main reason for not being able to reach the design luminosity in Run2 is the low overall pbar transfer efficiency, which stays around 40%. Where do we lose them? From the pbar intensity plot (Fig. 1), there are two obvious places. One is in the Tevatron, about 30% lost from injection to collision. Another is in the Main Injector, about 30% lost during pbar coalescing. It is the latter that we propose to get rid of so that the overall efficiency could be improved. The scheme works as follows: 1. The required longitudinal phase space density in the Accumulator is 12e10 pbars per ev-s. This is a modest number compared with what we achieved in Run1b. (Fig. 2) 2. Use h = 4 RF (ARF4, 2.5 MHz, 4.5 V) to capture 4 bunches from the cooled core. The bucket size is 0.25 ev-s. Each bucket contains 3e10 pbars. 3. Move these bunches to the extraction orbit and increase the RF voltage to the maximum (900 V) to narrow the bunches. 4. Extract the 4 bunches and inject them into the 2.5 MHz RF bucket in the Main Injector. 5. The MI 2.5 MHz RF is set at its maximum voltage (60 kv). A bunch rotation will occur immediately after the injection. The bunch length will be compressed to about ±7.4 ns. 6. Capture these bunches by the 53 MHz RF (at 110 kv) in the MI. The bunch spacing will be 396 ns. 7. Adiabatically increase the 53 MHz RF voltage to narrow the bunches. Accelerate the 4 bunches to 150 GeV and inject them into the Tevatron. 8. Repeat this process 9 times. Calculations show this scheme could work. No new hardware is needed to implement this scheme. One technical challenge is the regulation requirement of the ARF4 to the level of several volts in order to create a 0.25 ev-s bucket.

30 Figure 1. Pbar intensity plot of Shot #1764. Figure 2. Longitudinal phase space density at the peak of the antiproton momentum distribution just prior to unstacking as a function of stack size. (FERMILAB-TM-1970)

31 Accumulator Lattices Present Lattice η=0.012 New Lattice η=0.022

32 Measured Transverse Emittance Growth Rates Emittance (pi-mm-mrad) y = 19989x x x y = 53722x x x Horz. η=0.012 Horz. η=0.022 Growth rates were measured with the same beam on the same day. Emittance initial conditions for both measurements are close but not exactly the same 2 1 Vert. η=0.012 y = 6.279x Vert. η=0.022 y = x :00:00 0:10:05 0:20:10 0:30:14 0:40:19 0:50:24 1:00:29 Time

33 Graphic 1 of 1 10/9/2002 4:36 PM Close Duplicate

34 Remaining RF Extraction Problems Momentum width of core held large because of IBS Core is diluted after nine transfers Need to optimize momentum cooling First Transfer Ninth Transfer RF Curves

35 Graphic 1 of 1 10/9/2002 4:29 PM Close Duplicate

36 4. Longitudinal emittance growth and blow-ups a) initial emittance is large ( ½ RF bucket area at flat top) b) σ s =2 + ( )ns/hr growth consistent with known RF noises (some 60 urad in RF phase or/and 70V in voltage), microphonics idea