50 and 75 ns operation G. Arduini – BE/ABP Contributions from: BI, ABP Collective effects Team, Cryo, Injection Team, OP, RF, Vacuum Team
Dec 30, 2015
50 and 75 ns operation
G. Arduini – BE/ABP
Contributions from: BI, ABP Collective effects Team, Cryo, Injection Team, OP, RF, Vacuum Team
Evian Workshop - 50 and 75 ns operation 2
Outline
– Effects on vacuum with 50 and 75 ns beams– Scrubbing effect on vacuum– Effects on cryogenics with 50 and 75 ns beam– Scrubbing effect on cryogenics– Beam stability– Summary and outlook for 2011
07/12/2010
Evian Workshop - 50 and 75 ns operation 3
Vacuum – 50 ns – 450 GeV/c– Vacuum rise in all uncoated warm-warm and warm-cold transitions
(where vacuum pressure measurements exist):• pressure rise is observed also in vacuum pipes with single beam (not for 150 ns)• threshold of 6-8×1010 p/bunch and with trains of 24 bunches• Electron-cloud builds up after more than 12 bunches• Electron survival time up to 8-9 ms
0.6x1011 p/bunch
0.8x1011 p/bunch
1.1x1011 p/bunch
12+1212+2412+36
07/12/2010
Evian Workshop - 50 and 75 ns operation 4
Vacuum – 75 ns – 450 GeV/c– Vacuum rise in all uncoated warm-warm and warm-cold transitions:
• pressure rise is observed also in vacuum pipes with single beam (not for 150 ns)• Threshold of 9-10×1010 p/bunch. Increase by a factor ~3 when going from 1.1 to
1.3 x 1011 p/bunch.• Saturation after 2-3 trains of 48 bunches each• Important pressure rise when we inject a larger number of bunches
(incompatible with ramp where we have additional effect of synchrotron light for p>~1.5-2 TeV)
Losses
07/12/2010
V. Baglin
Evian Workshop - 50 and 75 ns operation 5
Scrubbing – 50 ns• Comparison between pressure rise before and after scrubbing
run for 12+36 bunches at 450 GeV (reduction by a factor 7 after ~16 hours with pressures larger than 10-7 mbar) Time constant ~8 hours 04/11: max p @
LSS3 VGPB.2.5L3.B=7.5E-8 mbar31/10: max p @ LSS3 VGPB.5L3.B = 5.5E-7 mbar
G. Bregliozzi, G. Lanza
07/12/2010
Evian Workshop - 50 and 75 ns operation 6
Scrubbing – 50 ns
Linear scale
12+2x24 bunchesTime constant ~3.5 hours
J-M Jimenez
07/12/2010
Evian Workshop - 50 and 75 ns operation 7
StSt MD 3 (Oct, 2008)
StSt MD 8 (Nov, 2009)
CNe13 MD3 (Oct, 2008)
CNe13 MD8 (Nov, 2009)
StSt MD 9 (March 2010)
StSt MD 13 (Sept 2010)
-6
-5
-4
-3
-2
-1
0
25ns50ns75ns
log
(no
rmal
ized
e-c
lou
d
curr
ent)
StSt StSt (cond.)
a-C a-C StSt StSt
Scrubbing experience (SPS)
07/12/2010
M. Taborelli
• Scrubbing with tightest bunch spacing allows operation below multipacting for the larger bunch spacings
Evian Workshop - 50 and 75 ns operation
8
Solenoids
Pressure rise when ALICE solenoid off
• Experimental solenoids and their fringe fields are effective in suppressing multipacting (see CMS and ALICE). But no build-up = no scrubbing potential problem in case one of the solenoids is OFF during a run.
G. Bregliozzi
07/12/2010
Evian Workshop - 50 and 75 ns operation
9
Cryogenics: 75 ns vs. 50 ns75 ns up to 824 bunches (Beam 1). No detectable heat load due to e-cloud no scrubbing in the arcs
50 ns up to 444 bunches (Beam 1) heat load due to e-cloud scrubbing in the arcs
-1
0
1
2
3
4
5
6
7
8
9
11/1
9/10
0:0
0
11/1
9/10
1:1
2
11/1
9/10
2:2
4
11/1
9/10
3:3
6
11/1
9/10
4:4
8
11/1
9/10
6:0
0
11/1
9/10
7:1
2
11/1
9/10
8:2
4
11/1
9/10
9:3
6
[W p
er h
alf-
cell]
, [Te
V], [
1013
p]
Qbs21L3 (calculated with T increase) Qbs33L6 (calculated with T increase)
Qbs13R7 (calculated with T increase) Qbs (IC+SR calculated with beam parameter)
Beam energy Intensity Beam1
Intensity Beam2
-1
0
1
2
3
4
5
11/2
0/10
4:4
8
11/2
0/10
5:1
6
11/2
0/10
5:4
5
11/2
0/10
6:1
4
11/2
0/10
6:4
3
11/2
0/10
7:1
2
11/2
0/10
7:4
0
11/2
0/10
8:0
9
11/2
0/10
8:3
8
[W p
er h
alf-
cell]
, [Te
V], [
1013
p]
Qbs21L3 (calculated with T increase) Qbs33L6 (calculated with T increase)
Qbs13R7 (calculated with T increase) Qbs (IC+SR calculated with beam parameter)
Beam energy Intensity Beam1
Intensity Beam2
L. Tavian
For both beams significant heat load in the beam-screens of the triplets-cold D1 region (particularly L8)
Visible e-cloud activity~ 40 mW/m on beam1
07/12/2010
Evian Workshop - 50 and 75 ns operation
10
Effect of scrubbing (50 ns)
Before scrubbing (30/10): Heat load ~20 mW/m/beam
After scrubbing (19/11)Heat load <10 mW/m/beam. Only B2
-1
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
11/1
9/10
21:
36
11/1
9/10
22:
04
11/1
9/10
22:
33
11/1
9/10
23:
02
11/1
9/10
23:
31
11/2
0/10
0:0
0
11/2
0/10
0:2
8
11/2
0/10
0:5
7
11/2
0/10
1:2
6
11/2
0/10
1:5
5
11/2
0/10
2:2
4
[W p
er h
alf-
cell]
, [Te
V], [
1013
p]
Qbs21L3 (calculated with T increase) Qbs33L6 (calculated with T increase)
Qbs13R7 (calculated with T increase) Qbs (IC+SR calculated with beam parameter)
Beam energy Intensity Beam1
Intensity Beam2
L. TavianSame filling pattern (9x12 b) and bunch population (~1011 p). Scrubbing at 450 GeV effective also for 3.5 TeV in the arcs
-1
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
10/3
0/10
14:
24
10/3
0/10
16:
48
10/3
0/10
19:
12
10/3
0/10
21:
36
10/3
1/10
0:0
0
10/3
1/10
2:2
4
10/3
1/10
4:4
8
10/3
1/10
7:1
2
10/3
1/10
9:3
6
10/3
1/10
12:
00
[W p
er h
alf-
cell]
, [Te
V], [
1013
p]
Qbs21L3 (calculated with T increase) Qbs33L6 (calculated with T increase)
Qbs13R7 (calculated with T increase) Qbs (IC+SR calculated with beam parameter)
Beam energy Intensity Beam1
Intensity Beam2
07/12/2010
Evian Workshop - 50 and 75 ns operation 11
Beam stability at 450 GeV/c (50 ns)
• Build-up of the electron cloud over more than one train leading to instabilities and emittance blow-up along the trains.
• Compatible with electron cloud instability
F. Roncarolo
12+4x24 – 1.85 ms spacing
07/12/2010
H: ~1 s rise timeV: ~0.2 s rise time
E. Métral
Evian Workshop - 50 and 75 ns operation 12
Beam stability at 450 GeV/c (50 ns)
• Can be stabilized by increasing chromaticity (up to 18 units) and by transverse emittance blow-up in the injectors (~3-3.5 mm @ SPS extr.) Can be used during a scrubbing period.
F. Roncarolo
07/12/2010
Evian Workshop - 50 and 75 ns operation
13
Beam stability at 450 GeV/c (75 ns)
Q’H,V=14 Q’H,V=24
B. Salvant
Does not present characteristics of the ECI, quadrupolar mode visible when looking at single bunch coupled-bunch mode due to impedance in that case running close to Q’=0 could also cure the observed instability.
07/12/2010
Evian Workshop - 50 and 75 ns operation
14
Beam stability at 450 GeV/c (75 ns)High chromaticity allows controlling the beam also for tighter
batch spacing (1.005 ms) and above nominal bunch intensity but vertical blow-up can be observed although no evident sign of instability incoherent effects of e-cloud also for low e-cloud densitiesBunch population = 1.2×1011 p – batch spacing 1.005 ms
07/12/2010
F. Roncarolo
Evian Workshop - 50 and 75 ns operation 15
Beam stability at 3.5 TeV (50 ns)• Observed
instabilities (H faster) with 1x12+4x24 bunches at flat-top when the damper is switched OFF. Not observed for 9x12 bunches for the same machine settings.
• Rise-time: few tenth of s (H), 1-2 s (V)
• Can be cured by the transverse feedbackH. Bartosik, B. Salvant
07/12/2010
Evian Workshop - 50 and 75 ns operation 16
Summary and outlook• Reduced vacuum activity with 75 ns spacing as compared to 50
ns but important pressure rise observed for large number of bunches
• No visible heat load due to electron cloud in the beam screens for 75 ns beam (but therefore no scrubbing!). Important heat-load in the triplet-cold D1 region in particular L8.
• Typical signature of ECI observed with 50 ns. Instabilities observed with 75 ns beam have not ECI siognatures although blow-up is visible that could be related to incoherent effects below ECI threshold scrubbing
• Comparison of 50ns beam behaviour during the ramp and at 3.5 TeV before and after scrubbing at 450 GeV clearly shows improvement in the straight sections (factor 7 reduction in pressure rise after 16 h of effective scrubbing) and in the arcs (heat load reduction by > factor 2).07/12/2010
Evian Workshop - 50 and 75 ns operation 17
Summary and outlook• SPS experience + LHC observations scrubbing with 50 ns
beams should allow operation with 75 ns below multipacting threshold
• This would imply having a period of 1 week for scrubbing at 50 ns (based on extrapolations from experimental data to be corroborated by simulations – ongoing -) with ~1.3-1.5x1011
p/bunch in order to run with 1.3x1011 p/bunch (maximum possible in the PS at present) with 75 ns later
• Why a scrubbing run with 50 ns at 450 GeV? – Avoid using physics fills ( background) for scrubbing– Provide margin for acceleration– Scrubbing of the arcs and suppression of incoherent effects below ECI
threshold for 75 ns beam– Scrubbing of the experimental areas with expt. (and e-cloud suppression)
solenoids OFF (could be a bottleneck later)– Would allow machine development with 50 ns beams with large number of
bunches during the year.07/12/2010
Evian Workshop - 50 and 75 ns operation 18
Summary and outlook• For that:
– Injection of at least 4x24 bunches with 50 ns spacing should be set-up before the scrubbing run (up to nominal transverse emittance at least)
– Machine protection should be set-up for high intensity at 450 GeV/c before the scrubbing run
– RF should be conditioned for operation at high intensity before the scrubbing run
– Solenoids (experimental and anti e-cloud) should be OFF during scrubbing– Vacuum interlock levels should be temporarily set to 2x10-6 mbar
07/12/2010
Evian Workshop - 50 and 75 ns operation 1907/12/2010
Thank You!!
Evian Workshop - 50 and 75 ns operation 20
Why 1 week of scrubbing?• From scrubbing data:
– 6 hours is the time it takes to reduce the pressure rise by a factor 2 when we run at pressures larger than 10-7 mbar
– Pressure rise to 1.6x10-6 mbar and heat load of 40 mW/m/beam with 444 nominal bunches (trains of 48 bunches spaced by 3 ms and a single beam)
– Factor 3 in pressure rise when going from 1.1 to 1.3x1011 p/bunch (seen with 75 ns)
• Assumptions:– Linear scaling of the pressure rise vs. number of bunches– We can set the interlock level to 2x10-6 mbar and keep a factor 2 margin– Heat load margin for e-cloud for both beams = 2.3 W/m – 0.4 W/m (from
impedance) 1.9 W/m with factor 2 margin– 25 % efficiency
• Then:– For an injection of 48 bunches of 1.3x1011 p Dp < 10-6 mbar at the beginning
of the scrubbing. We can double the number of circulating bunches every 6 hours 1260 bunches (12+13*96) with pressure lower than 10-7 mbar after ~42 hours of scrubbing 168 h of time = 7 days
07/12/2010
Evian Workshop - 50 and 75 ns operation 21
Scrubbing – 50 ns No correlation between PBeam and position along the ring Pressure increase reduced by factor 2/3 within 3 hours
J-M Jimenez et al.
07/12/2010
Evian Workshop - 50 and 75 ns operation
22
Effect of scrubbing: vacuum and synch lightBefore scrubbing (30/10):
Pressure rise due to synch light
After scrubbing (19/11)No visible pressure rise due to synch light. Only B2
Same filling pattern (9x12 b) and bunch population (~1011 p). Scrubbing at 450 GeV effective for reducing pressure rise due to photo-electrons above 1.5-2 TeV
07/12/2010
Evian Workshop - 50 and 75 ns operation 23
High intensity effects
• Capture losses observed on beam 2 during Thursday night Cavity 7B2 is showing some noise on the accelerating voltage in the presence of high intensity. Possible sources:
– Multipacting in the antenna pot or cable connector perturbing the field measurement (and the cavity feedback).
– Micro-quenches in the cavity causing a real field drop (and also perturbing the feedback).
• Might be triggered by multipacting/electron cloud in the cavity.• Conditioning at higher voltage might help. Conditioning with
beam might be required.
A. Butterworth
07/12/2010