Transcript Accelerator Layout and Parameters R. Brinkmann, DESY R. Brinkmann

DESY
Accelerator Layout and
Parameters
R. Brinkmann, DESY
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
TESLA TDR March 2001: integrated XFEL
R. Brinkmann
DESY
ESFRI XFEL Workshop 30./31.10.2003
DESY
411 Authors TDR Part-II: Accelerator
1000
100
10
R. Brinkmann
CERN
USA
UK
Spain
Russia
Netherlands
Japan
Italy
Germany
France
Finland
China
Armenia
1
ESFRI XFEL Workshop 30./31.10.2003
Basis for TDR: high-performance sc cavities
DESY
( Talk H. Weise & poster session)
cw-cavity tests, 1st and best results
35
best test
g [MV/m]
30
25
20
15
10
1st test
5
0
1/1/95
5/15/96
9/27/97
2/9/99
6/23/00
date of measurement
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Update 2002: separate XFEL linac
beamline #2
DESY
beamline #1
Vert. bend
Collimation
Diagnostics
2.4 km
20…40GeV
LINAC
Upgrade
10…20GeV
1.4 km
Extraction
switch
20 + 2 units
LINAC
1st Stage
BC-III
3.9km
2.5GeV
BC-II
Injector
R. Brinkmann
BC-I
0.5GeV
0.1 km
3 + 1 units
ESFRI XFEL Workshop 30./31.10.2003
Update 2002 cont’d
DESY
• de-coupling LC - XFEL regarding construction & operation (and:
approval), maintaining common site; keep cost close to TDR2001
figure by reduction of (1st stage) energy to 20 GeV, # of beam lines
10  5
Parameters 1Å SASE, fixed gap undulator:
Eb
Ipk
KU gap Lsat Ltot
e
sE
lU
[GeV] [10-6m] [MeV] [kA] [mm]
[mm] [m] [m]
TDR2001 25
1.6
2.5
5
45
4
12
210 311
Update02 20
1.4
2.5
5
38
3.8 10
145 213
(FEL process & parameters:  talk by M. Yurkov)
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Present: detailed project preparation for
construction start in ~2 years
DESY
• Choice of DESY-near site de-coupled from TESLA LC
• Linac in TTF-like technology, 20 GeV at 23 MV/m,
abandon length reserve, optional higher energy from
improved cavity performance
• Investigation of operating parameter flexibility, in
particular regarding duty cycle / repetition rate
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Present accelerator layout
DESY
beam lines
low-E beam
line (?)
0.5 km
Beam distr.
Collimation
Diagnostics
10…20GeV
Extraction
switch (?)
BC-III
R. Brinkmann
2.5GeV
16 modules
4 RF stations
BC-II
Injector
1.5 km
100 modules
25 RF stations
BC-I
0.5GeV
0.1 km
Main
LINAC
ESFRI XFEL Workshop 30./31.10.2003
Accelerator layout cont’d
waveguide
tuner (var.
Q_ext)
eight 9-cell Nb
cavities at 2K, Q0=1010
R. Brinkmann
5 MW RF
source
De-rated
10MW MBK
DESY
Bouncer-type
modulator
12m TTF-like
acc. modules
ESFRI XFEL Workshop 30./31.10.2003
Linac tunnel layout
DESY
Accelerator housed in a ~5m diameter tunnel ~12m underground
Klystron in tunnel are connected to modulators in external hall by 10kV
pulse cables
Two options for tunnel layout being reviewed
30 cm
125 cm
275 cm
520 cm
190 cm
90 cm
80 cm
210 cm
440 cm
Figure 3. Main LINAC, Damping Ring & Klystron Station
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Reference parameters
Main linac
Energy gain
# installed modules
# active modules
acc gradient
# installed klystrons
Bunch spacing
beam current
powerbeam p. klystron
incl. 10% + 15% overhead
matched Qext
RF pulse
Beam pulse
Rep. rate
Av. Beam power *
Total AC power
DESY
0.5  20 GeV
116
104
22.9 MV/m
29
200 ns
5 mA
3.8 MW
4.8 MW
4.6×106
1.37 ms
0.65 ms
10 Hz
650 kW
» 9 MW
* Power limitation to ~300kW per beamline  solid beam dump possible
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Parameter flexibility: rep rate
DESY
max rep rate and beam pulse length vs. E_acc
Q_ext varied with E_acc (I_beam=const.)
160
0.80
f_rep [Hz]
T_beam [ms]
140
f_rep [Hz]
120
0.60
100
80
0.40
60
40
0.20
20
0
0.00
0
5
10
15
20
25
30
E_acc [MV/m]
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Longer term perspective: CW an option??
DESY
• At 20 GeV: cryogenic plant 3 times capacity of 500 GeV TESLA
LC… unreasonable!
• At lower energy: worth a serious thought…consider following option
for low-current, CW-mode (e.g. 1nC bunches at 10kHz):
–
–
–
–
At 23MV/m power per cavity is 30kW at ~zero beam current
Energy/sqrt(10) (because of cryogenics)  power per cavity 3kW
Increase Q_ext ~ factor 3  power <1kW (+regulation reserve)
 linac can deliver ~6GeV CW beam without modifications –
except for (of course) additional RF system with ~50kW CW
klystrons (or: IOTs)
• careful: we don’t have a suitable injector yet!
• No need to rule out ERL-type of operation in further distant future –
except for few basic considerations, can’t be focus of attention for
next years
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
(some of the)
Accelerator sub-systems
DESY
• Photocathode RF gun and bunch compressors: generate and
preserve excellent beam quality (peak current 5kA, normalised
transverse emittance 1.4 mrad*mm) ( talks by K. Flöttmann, W.
Sandner, poster session)
• Beam delivery system: ( poster session)
– Beam collimation: protect undulators from halo and mis-steered beams
– Diagnostics: determination of beam properties in 6D phase space with
complex structure
– Beam distribution: slow and fast devices for multi-user operation
– Feedback systems: intra-train stabilisation (position, energy, timing, …)
using “pilot bunches”
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Beam distribution options
DESY
• Straight forward and simple: pulse-to-pulse switch magnet
• Much better, but more challenging: fast (intra-train) kickers
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Different users – different time structures
DESY
• Generation of bunch train
patterns:
– At the source  varying transient
effects in the entire accelerator
(handled e.g. by the LLRF
system)
– At the beam delivery/distribution
system  more challenging kicker
devices
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
One source of jitter: ground motion
DESY
y_rms [nm]
150
100
50
0
1
2
3
4
5
6
7
8
9
10
position along site
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Intra-train beam stabilisation
DESY
• From ground vibration: jitter ~ 0.1s at end of linac
– Can be enhanced during “single events” e.g. heavy traffic, and by quad
support eigenmodes
– Other effects: stray fields, HOMs, …
•  feedback system between linac and distribution to undulators
user bunches
Alignment bunches
Also active stabilisation of energy and possibly other beam parameters
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003
Conclusions
DESY
• The 20 GeV s.c. linac based on the technology developed by
the TESLA collaboration and successfully demonstrated at TTF
is an ideal driver for the Free Electron Laser facility – offering a
broad range of operating parameters in its baseline design and
with future upgrade options
• Experience at test facilities creates a solid ground also for the
design of other accelerator sub-systems (injector, bunch
compressors, diagnostics, feedback systems, …) of the XFEL
• The project preparation is well under way and will lead into the
construction phase in ~ two years from now
R. Brinkmann
ESFRI XFEL Workshop 30./31.10.2003