background issues in the new hera interaction region• at hera complications due to ﬁxation of...

Background Issues in the new HERA

Interaction Region

M.Seidel, DESY

January, 2004

• Overview on HERA Interaction Region• p-Gas Background• Synchrotron Radiation• e-Gas Background• Recommendations from HERA Experience

JJ II B OM.Seidel DESY, IR Mini-Workshop, IHEP Bejing’04

Goal-Parameters for theUpgraded HERA

e-Beam p-Beamenergy 27.5 GeV 920 GeVbeam current 58 mA 140 mAemittance 22 nm 5 µm /γemittance ratio 0.18 1betafunction IP β∗x 0.63 m 2.45 mbetafunction IP β∗y 0.26 m 0.18 mbeam size σx × σy 118µm×32µm 118µm×32µmtune shift/IP ∆νx,y 0.027, 0.041 0.0017, 4.6 · 10−4aperture 20 σ 12 σ

Luminosity 7.00 · 1031cm−2s−1

=⇒ specific luminosity is raised by factor 2.8


Overview on new IR

sc magnets

SR Lightp−beame−beamchamber

Absorber

GN GN GM GM GJ GI GO GM GM GN GNGG GI GI GJ

−0.5

−0.4

−0.3

−0.2

−0.1

0

0.1

0.2

0.3

0.4

−30 −20 −10 0 10 20 30

x [m

]

z [m]


Switching the Lepton Speciesby Quadrupole Shifting

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

-20 -15 -10 -5 0 5 10 15 20

x [m

]

z [m]

Offs

et 7

.5m

m

electrons, 20σxprotons (e-), 12σxpositrons, 20σxprotons (e+), 12σx


Side View of H1 Detector


GO magnet in H1


ZEUS Detector


Bridge rightMagnets open


Chamber Cross-Section


Septum Absorber - Beam View

SR Power: P ≈ 6 kWCrit. Energy Ec ≈ 150 keV principle: � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

� � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ��

20 mm

40 mrad


Absorber Coating Under Work

idea: thin low Z layer over thick high Z layerproblem: stability at brazing temperature

history of sandwich-layers tested

Mo − 100um

Cu − 15um

Ag − 25um (braze)Cu − 15um

Cu − bulkCu − bulk

Ni − 15um

Ni − 100um

Ag − 100um

Cu − bulk

Ni − 2umAu − 20umNi − 2um

Ag − 50umNi − 2umCu − 15um


Three ChambersBehind SeptumAbsorber


Mirror Plate Magnet GM


p-Chamber in GM, Septum Magnet


Overview on Types ofBackgrounds

we expected (and have evidence for) four types ofbackgound:

• Proton Halo Losses in low-β quads– critical when diffusion/emittance growth from beam-beam effects comes in

– controlled by two stage collimation system; relative beam size and position

• Proton Gas Scattering– lepton-SR induced Pressure Rise!, HOM heating!

– dominating effect!

• Synchrotron Radiation– masking of direct and indirect radiation

– orbit-control → bending in quadrupoles; direction of radiation

• Lepton Gas Scattering– wide energy spectrum of scattered leptons

– momentum collimation


Proton Gas Scattering

• most sensitive within 20 m from the IP (Monte Carlo)• verified with artificial pressure bumps (red)

• cross-section with gas species: σI ∝ A23

that the gas released when the TSPs are heated is primarily hydrogen. Reasonable agreement isseen, particularly when it is borne in mind that the systematic uncertainty in the measurementsis at least 20%, as is illustrated by the spread of the three measurements taken at 3.6 m, eachwith different values of Pmax. Further, the geometry of the pumps in the range 11 to 61 m issuch that the getter pump sees a larger proportion of the pressure induced in the heating of theTSPs than do the getter pumps closer to H1. The more distant values should thus be correctedupwards somewhat.

z/m

∆IC

JC1/I

p/p m

ax [µ

A/m

A/1

0-9 m

bar]

0 20 40 600

0.01

0.02

0.03

0.04

0.05TSP heating datapp MC (arb. units)

Figure 30: The sensitivity of the CJC currents to changes in the pressure in the beam lineinduced by heating the TSPs, shown as a function of the position along the beam line up toz = −60 m from the H1 detector for proton only operation. The points represent the data whilethe band is a Monte Carlo prediction assuming that heating the TSPs leads to the production ofhydrogen.

3.1.4 Switching off pumps

Experimental measurements of the sensitivity of the H1 detector to deterioration in the vacuumat different positions along the beam line may also be made by turning off selected ion getterpumps. This was attempted for the pump at NR 3.6 m. The results are shown in figure 31.There is no perceptible change in the CJC current during the period in which the high voltage ofthe getter pump was off. It is, however, likely that the pressure change resulting from switchingoff the getter pump was small. Firstly, the pumping speed of the getter pump is small comparedto that of the TSP at the same location, 60 l/s as opposed to 1000 l/s, and secondly, the getterpump continues to function as a titanium sublimation pump even when turned off until thetitanium surfaces become saturated; this may take hours or even days depending on the pressure.Despite this, the pressure measured at NR 6 m is seen to rise slightly. It is possible that thisis due to methane, which is not efficiently pumped by TSPs. This experiment was consideredinconclusive due to the various difficulties mentioned above and was therefore repeated using

30


IR Pressure in typical Run

1e-11

1e-10

1e-09

1e-08

1e-07

11/1220:00

12/1200:00

12/1204:00

12/1208:00

12/1212:00

12/1216:00

12/1220:00

13/1200:00

13/1204:00

13/1208:00

13/1212:00

13/1216:00

13/1220:00

Dru

ck [m

bar]

north L8north L6

north R3.5north R6north R8

0e+00

1e+01

2e+01

3e+01

4e+01

5e+01

11/1220:00

12/1200:00

12/1204:00

12/1208:00

12/1212:00

12/1216:00

12/1220:00

13/1200:00

13/1204:00

13/1208:00

13/1212:00

13/1216:00

13/1220:00c

urre

nt [m

A],

ener

gy [G

eV]

Zeit

energycurrent


Simulated Pressure Distribution

1.0e-11

1.0e-10

1.0e-09

1.0e-08

-20 -15 -10 -5 0 5 10 15 20

P[m

bar]

IP

zeush1

0.0e+001.0e+022.0e+023.0e+024.0e+025.0e+026.0e+02

-20 -15 -10 -5 0 5 10 15 20spe

c. s

peed

[l/s

m]

pos[m]

pumping speed


Old and New Radiation Mask in H1


Temperatur GG Flange at H1H1 Beam pipe temperature

0

5

10

15

20

25

30

35

40

45

50

9 12 15 18 21 24

Wed 27.11.2002time [hours]

T[o

C]

Wed 27.11.2002Wed 27.11.2002Wed 27.11.2002Wed 27.11.2002Wed 27.11.2002Wed 27.11.2002


Pressure Development 2002/03


Development of Drift ChamberCurrents at H1 CJC2 current history

10-1

1

10

100 200 300 400 500 600 700days from 1.1.2002

( CJC

2 - I

0 - S

R )

/ ( S

ep ‘0

2 pa

ram

eter

izat

ion

e-ga

s + p

-gas

)

10-1

1

10

IeIp < 150 mA2

IeIp < 500 mA2

IeIp > 500 mA2

IeIp > 1000 mA2

τ = 100 days

leakNL 11m

GO/GG110K170K

leakNR11m

leakNL

11m

for 5000 mA2

DP 04/01/2004 13.56.09JJ II B O

M.Seidel DESY, IR Mini-Workshop, IHEP Bejing’04

Synchrotron Radiation

• beam separation and quads produce synchrotron radiation (SR);small powers (10−8 of peak density) already problematic!

• orbit mis-steering:– wrong direction → photo desorption; back scattering; heating– possibly much more power produced; higher critical energy

• orbit control via BPM’s, beam based methods• continuous quad aligment monitoring• tails in particle beam problematic


Radiation from Quads is Wider!

phase space distribution of radiation from a thin slice of combinedfunction magnet (γ−1 opening angle neglected):

dP

dl(x, x′, y, y′) =

C04π2εxεy

{(1ρ

+ Kx)2

+ K2y2}

exp(−γxx

2 + 2αxxx′ + βxx′2

2εx

)exp(. . .)

step through the beamline and integrate projections of suchdistributions:

1e-05

1e-04

0.001

0.01

0.1

1

10

100

1000

0 0.02 0.04 0.06 0.08

W/m

m

x [m]

chamber wall

BOGJGI

GOall

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

-10 -8 -6 -4 -2 0

x [m

]

z [m]

chamber wall

BO GJ GI GO


2D Simulation -Nominal and Distorted Orbit

fan projected

to 3.5m right

of IP

1

10

100

1000

10000

100000on axis orbit [W/cm2]

-4-2 0 2 4

horiz. dist., 3σ offset

-4-2 0 2 4

vert. dist., 3σ offset

hor. position [mm]

vert.

pos

ition

[mm

]

pow

er d

ensi

ty [W

/cm

2 ]

-20 0 20 40 60 80 100

-4-2 0 2 4


Beam Based AlignmentTwo Methods

QuadQuadKick Kick

difference amplitude around the ringglobal optics errors!

kompensation of kick by two correction coilscalibration of coils! statistics of all monitorscontributes

specific problems at HERA:

• nominal offset in x not zero!(paper by G. Hoffstätter, F. Willeke)

• coupling in vertical plane• IR quadrupoles are thick lenses• advantage: lumidetector fixes IP angle


Stretched Wire Alignment System

• stretched gold-plated wire as reference line• 100 MHz signal on the wire is detected in BPM-like monitors• resolution better 1 µm possible; demonstrated at SLAC FFTB• at HERA complications due to fixation of the wire end point on a

magnet support structure

��

��

Tunnel

Bridge

Hall Floor

Magnet

Wire

Magnet Magnet

Detector

IP

remote adjustable

remote adjustable


Effects from Electron beam Tails

particle beam tailsgenerate tails in theradiation fanscraper measurementsfrom HERA-e(A.Meseck, 2000)

�� !�#"!��$�!%&�

100000

1e+06

1e+07

1e+08

1e+09

-10 -8 -6 -4 -2 0 2 4

coun

t/sec

ver. scraper position in mm

"d011958a.dat" using 9:($19*29000)"d15371.d" using 1:6

"d022002a.dat" using 10:($19*36000)"d25371.d" using 1:6

10000

100000

1e+06

1e+07

1e+08

1e+09

-10 -5 0 5 10

coun

t/sec

hor. scraper position in mm

"d031306a.dat" using 11:($19*29000)"d035051a.dat" using 12:($19*36000)

"d1800.d" using 1:6"d2800.d" using 1:6

'�(*),+.-0/#132*4,57698;:=< >@?A8CBEDGF!HGI�J�:LKMB�N OM8QP;:SRTP;>MR*:SN8CBUDVRTWXKM8;J7Y@8;:[Z\RT]L<@aMeM?:SB@8gfG]MP;>M<W*KMe�<PCOM8;Z!8gd�WTN OM]M>MNhY@8;:LZjikY@8;:LZlW*KLN8;?m:SPCNWT]bKcn


Simulation of the SR Fanincluding beam tails

beam model distribution

1e-14

1e-12

1e-10

1e-08

1e-06

1e-04

0.01

1

-2 0 2 4 6 8

Dic

hte

x in σx

normal distributiontail distribution

sum

resulting distribution of thefan at +3.5 m

1e-06

0.0001

0.01

1

100

-0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1 0.12

P [W

/mm

]

x-position [m]

chamber wall

electrons, Gaussianelectrons, tails

electrons, tails scraped


Lepton Gas Scattering

• electrons scatter with residual gas molecules and lose energy;if this happens in the straight section upstream of the detectorthey get lost preferentially in the separation field

• wide energy range for scattered leptons σ ∝ 1/E• strong Z dependence for rest gas σ ∝ Z2

• sensitivity over long distance

comparison of old new HERA IR

Detector

now

IP

Beam

old setup


Dispersion Function forDifferent Energies

problem: scattered particleslose energy; are bent strongerthan nominal inside the de-tector; produce background

remedy: energy collimation;dispersive section

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

-60 -50 -40 -30 -20 -10 0

x [m

]

Dipol

Dipol

electrons, 20 sigma

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

-60 -50 -40 -30 -20 -10 0

D =

(x(p

)-x(

p 0))

/(p-p

0) [m

]

BH BN

γ

γelectrons

electrons

p = 0.995*p0p = 0.950*p0p = 0.850*p0p = 0.750*p0p = 0.500*p0


Simulation w/o Energy Collimation

only beam particles that hitdetector pipe are plotted

particle energy as functionof distance to IP, where theresidual gas interactionoccurs

(U. Kötz, ZEUS)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

020406080100

∆p/p

z [m]

with chicane (738 hits)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

020406080100

∆p/p

z [m]

without chicane (1226 hits)


Recommendations fromHERA Experience

• Synchrotron Radiation– careful simulation and prediction of fan locations required

– error analysis of orbits and magnet positions

– shielding of critical vacuum components (bellows, flanges)

– HOM heating produces outgassing!

– consider effect of beam tails; possibly scraping required

• Beam Gas Background– simulation of beam losses from gas scattering (energy loss)

– consider momentum collimation

– careful layout of vacuum system at critical locations

– estimate and prediction of vacuum conditioning


background issues in the new hera interaction region• at hera complications due to ﬁxation of...

Documents