Background in the Interaction Regions

1
Background in the Interaction Regions

• Manuela Boscolo

Italy- Japan Symposium, Frascati, February
2002 DAFNE - KEKB Meeting
2
Outline

• Present situation
• Simulation study based on home-developed
  MonteCarlo (STAR)
• Effects of machine non linearities
• Comparison with measured collimators efficiency
• Studies of additional scrapers (just installed 3
  per ring)
• Work in progress and conclusions.

3
DAFNE background and beam lifetime in the
experiments are dominated by Touschek scattering,
that is an elastic Coulomb scattering of pairs of
particles within a bunch. The two emerging
particles have the same momentum deviation one
gains and the other looses it. They start to
oscillate with a betatron amplitude
N particles/bunch V bunch volume e momentum
acceptance
The Touschek rate is approximately
Touschek lifetime is determined by momentum
acceptance and bunch density integrated over the
lattice structure.
Off-momentum particles can exceed the momentum
acceptance given by the RF bucket, or may hit the
physical aperture when displaced by dispersion.
In addition, a betatron oscillation is excited if
the momentum change happens in a dispersive
region.
4
KLOE Physics Luminosity arrow
5
Present situation

• At the beginning of KLOE operation the high
  background rates limited the maximum stored
  current.
• Background reduction obtained by
• tuning orbits and optical functions in the IRs
• adjusting sextupole strengths
• optimising scrapers configuration.
• Jan.01 Scrapers have been upgraded to improve
  absorption efficiency.
• Jan.02 Additional scrapers have been installed,
  3 per ring.
• The background to luminosity ratio decreases by
• a factor 2-3 (e-e-) with respect to June 2001.

6
October 6th 2001 24hours KLOE data taking
7
June 30th 2001 24hours KLOE data taking
8
June 30th 2001 24hours KLOE data taking
9
October 6th 2001 24hours KLOE data taking
10
Effects of non linear elements on backgrounds
Modelling of Touschek background particles
all magnetic elements have been considered,
including sextupoles and octupolar components in
the wigglers.

• When non linear terms are included in the
  simulation beam losses at KLOE increase.
• The closest and the farthest arcs from
  experiments produce most of the Touschek
  particles eventually lost at IRs.

11
DAFNE layout
PL1
PS1
Ecm 1019 MeV q 25 mrad C 98 m
PS2
PL2
12
Relevant beam parameters used for Touschek
simulations
Touschek particles are generated according to
H.Brucks formula and tracked through the machine
13
Simulations results at KLOE (1bunchIb10mA)
KLOE detuned optics
PL1
PL2
PS2
PS1
14
Summary of simulations results at KLOE
(1bunchIb10mA)
Contributions of the four arcs to beam losses
along the ring and different non linear
configurations.
Expected beam losses at the KLOE IR significant
contributions are expected from the two arcs PL1
and PS1.
15
Losses at KLOE IR coming from PL1 vs turns
16
Simulations results on losses at DEAR
(1bunchIb10mA)
last DEAR optics (Dec.01)
1 turn
PS2
PS1
1 turn
PL1
1 turn
PL2
2,5,8 turn
17
Sextupoles Optimization

• Losses at experiments are very sensitive to
  sextupoles setting values both from simulations
  and experimentally.
• Experimental careful tuning of sextupoles led to
  a background reduction, by a factor 2 for
  electrons and smaller factor for positrons.
• Comparison between expectations and measurements
  is not easy, as adjustments on machine are
  adiabatically done during machine runs with
  colliding beams
• A systematic comparison has been done close to
  sextupoles optimized setting values.

18
Variations induced on sextupoles
Detuned optics
Detuned lowb optics
19
Comparison between expected and measured
background rates at the KLOE calorimeter versus
sextupoles strengths
The MC reproduces actual behaviour of background
vs sextupoles strengths
Simulated loss rates have been normalized to
measurement (27/09/01) nb47 Ibunch5mA
(Itot100mA) and normalization is made for
R0.1 (Rsy/sx)
20
KLOE IR
evaluation of calorimeter acceptance is essential
for a comparison between measured and simulated
background rates
ECM2
negligible contribution to bkg
ECM2
ECM2
21
Collimators at the two DAFNE rings

• using the non linear model previously discussed
• AND
• after the calorimeter efficiency evaluation

a comparison between predictions and measurements
on the collimators efficiencies has been done
22
In order to protect the detectors of the
experiments from off-momentum particles remote
controlled scrapers have been installed for the
incoming beams on either side of each experiment.
They are placed before the splitter magnets,
about 7.0m from the IP
SCHPL101
SCHEL101
16 r.l.
5.5cm
x(mm)
44
W
19
BEAM
0
S
Two horizontal jaws per scraper, external and
internal, are used to intercept the two
off-energy particle families. the scrapers
consist of tungsten blocks of 55 mm thickness,
which corresponds to 16 r.l.
-19
W
PIPE
-44
23
Scraper
The inner surface of the new scraper blocks is
divided into two flat parts. A first 10 mm long
section with a slope of 100 mrad towards the
beam, in order to increase the impinging angle
into the block for most particles. This is
followed by a second section of 45 mm length
which slopes by 10 mrad in the opposite direction
to avoid foreward scattering of electrons back
into the beam pipe. The total scraper thickness
of 55 mm (about 16 r.l.) reduces the punch
through probability of 500 MeV electrons to below
10-6.
24
KLOE scraper scan
single beam measurements (June 01 )
sx 2.9 mm at KLOE scraper 26mm 9 sx max.
efficiency 19mm 6.5 sx max. jaw position
25
DEAR scraper scan
single beam measurements (June 01 )
sx 3.4 mm at DEAR scraper 25mm 7.4 sx ? D.A.!
26
Collimators Modeling

• Perfectly absorbing scrapers
• No width

actual behaviour is reproduced BUT Edge effect
is missing
It has been found that most of the particles are
scattered by the scraper edge, instead of being
absorbed, thereby producing additional background
to the experiments.
WORK IN PROGRESS real scraper shape included in
simulation and edge effect has been preliminary
simulated
27
First results with real scraper in code STAR

• Electron interaction Multiple scattering,
  ,Bremsstrahlung, de/dx simulated by a toy MC
• many particles to be tracked

Electron interactions at scraper edge
Zoom at scraper
Only additional background to KLOE IR from edge
effect displayed
28
DAFNE scrapers position
A further reduction of KLOE background is
expected with additional scrapers in regions with
non-zero dispersion.
29
new scrapers simulated scans
Scrapers assumed perfectly absorbing and
infinitely thin
sx 2.9mm at SCNPS1 scraper 20 mm 6.9sx max.
jaw position
sx 2.95mm at SCNPS1 scraper 20 mm 6.8sx max.
jaw position
SCNPL1
SCNPS1
30
DEAR scraper efficiency with KLOE scraper in
With the KLOE scrapers at 30mm the DEAR scrapers
are expected to further reduce the KLOE
background by a factor 10. Observed data show
just about 20 improvement.
the expected KLOE background rate has been
counted in the whole interaction region (IR),
from splitter to splitter, as in principle all
the particles hitting the vacuum chamber in this
region are dangerous.
31
New scrapers efficiency with existing scrapers in
SCNPL1
SCNPS1
reduces KLOE loss rate for a factor 2.5 with no
other scrapers
reduces KLOE loss rate for a factor 2 both with
no other scrapers and with (black) the scrapers
upstream the two experiments inserted.
KLOE scraper jaws at 30 mm DEAR scrapers at
26 mm ?no loss rate reduction is foreseen in
KLOE IR.
32
KLOE IR- simulation results
47 bunches,Ibunch20mA, sz2.4cm, k0.01? I0.94
A Octupolesextupoles
KLOE IR (splitter-splitter) BKG(MHz)
Scrapers efficiency ()
part. from PL1
part. from PS1
33
Summary of scrapers efficiencies at KLOE
20 bunches,Ibunch5mA, sz1.9cm, k0.01?
I100mA Octupolesextupoles
Expected KLOE IR (splitter-splitter) BKG(KHz)
expected Scrapers efficiency
measured Scrapers efficiency
34
New scrapers efficiency to DEAR background rate
KLOE
DEAR
Last DEAR optics Dec. 2001
Losses at DEAR vs turns (with no scrapers)
35
DEAR IRSimulation results(DEAR optics)
47 bunches,Ibunch20mA, sz2.4cm, k0.01? I0.94
A Octupolesextupoles
DEAR IR (splitter-splitter) BKG(MHz)
Scrapers efficiency ()
part. from PS2
part. from PL2
36
Conclusions

• Significant progress has been made in the
  understanding of the machine induced background
  to the KLOE and DEAR experiments
• Touschek background modelling has been improved
  and MC reproduces effective behaviour
• Beam losses have been found very sensitive to
  machine non linearities, tunable octupoles are
  foreseen as knob to further reduce background.
• Additional scrapers have been installed this
  January, simulations predict some reduction of
  beam losses at KLOE IR.

• FUTURE PLAN
• Include real octupoles in simulations and find
  their optimum setting values for background
  minimization.
• Simulate real scrapers with edge effect.

37
Beam Lifetime

• The lifetime in DAFNE is limited by the Touschek
  effect.
• It is short 1/2 h in typical conditions of KLOE
  data taking (coupling is lt0.5 much lower than
  design)
• It influences strongly machine operation ?
  frequent refills and high bgk rates.
• Third harmonic cavity is foreseen to increase
  bunch length.

38
Typical behaviour of the measured background rate
as a function of scraper openings
(dashed blue line) It appears that the external
jaw reduces the background by about a factor of
2.3 at a distance of ?25 mm from the beam axis,
while the closed orbit was measured at -2.7 mm
from the center of the chamber. A steep increase
of the background rate is observed when the
scrapers are closed to less than about 23 mm
(9sx) from the closed orbit position. Apparently,
from this moment on, the scrapers are producing
more background particles than they are stopping.
The background rate measured by KLOE for an
electron beam as a function of the KLOE scraper
setting is shown
(full red line) It is shown the behaviour of the
new scraper. The external jaw reduces the
background by a factor 2.9 at a distance of ?20
mm from the chamber axis. No background reduction
is found when moving in the internal jaw,
however, the previously observed strong increase
is no longer present (dashed line), indicating an
improved stopping efficiency of the scraper
blocks.
39
Scrapers scans performed with single beams
Rates of the two forward calorimeters in KLOE
have been taken as a function of the opening of
the horizontal scrapers around IP1. The two
rates from the calorimeters ECM2 and ECM4
integrate the signals over the four innermost
sectors of the west and east forward calorimeters
each.

Simulated background rates have been normalized
to experimental situation 20 bunches with
Ibunch5mA (Itot100mA) and R0.1 (Rsy/sx)