NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)1
5th September 2006
Operational Experience with the CNGSBeam Position Measurement System
OUTLINE
1. Introduction to the CNGS BPM system
2. Laboratory results
3. Performance of the system with beam
4. Problems encountered during commissioning
5. Summary
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)2
5th September 2006
The CNGS Beam Line Position Monitors
CNGS Beam Position Measurement Requirements
18 Button ElectrodeBPMs in TT41
60mm Aperture
Intensity Range:11012 to 3.51013
110
340
110
BPCKD (double)
short circuits
BEAM 4 Stripline Couplers in TT40will see both LHC & CNGS beams
Original
Modified
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)3
5th September 2006
Construction of the CNGS Target Station Monitor (BPKG)
The CNGS Target Beam Position Monitor
Coupler Body- Aluminium alloy- lowers remnant radiation
Outer Surface Treatment- penetrating oxide layer- withstands radiation effects- gives thermal stability
Inner Surface Treatment- 30m gold layer- withstands radiation effects- maintains good conductivity
Feedtroughs- Ceramic dielectric (vac seal)- simple 50construction
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)4
5th September 2006
The CNGS BPM Electronics
CNGS Beam Position Measurement Acquisition System
LPFilter
LPFilter
A
B
LogampI to V
Converter
Diff .Ampli
Position = K * VoutLogamp
I to VConverter
BPFilter
BPFilter
LPFilter
LPFilter
LPFilter
LPFilter
A
B
LogampI to V
Converter
Diff .Ampli
Position = K * VoutLogamp
I to VConverter
BPFilter
BPFilter
BPFilter
BPFilter
• Each signal is compressed by a logarithmic amplifier, filtered and applied to a differential amplifier.
• The position response is: Pos [log(A/B)] = [log(A)-log(B)] (Vout) where Vout is the voltage difference between the log-amp outputs
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)5
5th September 2006
The CNGS BPM Electronics
CNGS Beam Position Measurement Front-end
D
Calib.Sequencer
VInt.
ΣInt.
HInt.
Analogue
Mux
Logic
U-D
Dataout
Control
ADCΣ
U
L
R-L
U + D+ L + R
Manchesterencoder
Xilinx XC9500 CPLD family
RCoupler
Coupler
Coupler
Coupler
Splitter
Splitter
Splitter
Splitter
AutoTrigger
200MHz BPFilters
Log Amps
Pickup
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)6
5th September 2006
The CNGS BPM Electronics
CNGS Beam Position Measurement Acquisition SystemBased on LHC BPM system:
VME64x Digital Acquisition Board- DAB64x (TRIUMF, Canada)
Altera Stratix (EP1S20) FPGA- use same code as for LHC BPMs
Mezzanine Card- receives info from 6 front-ends- performs manchester decoding- Xilinx FPGA treats data to give
correct input for DAB64x
Final configuration- 2 DAB64x with 2 mezzanines- processing data from 23 CNGS PUs
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)7
5th September 2006
The CNGS BPM Electronics
CNGS Beam Position Measurement Acquisition SystemWhy this choice of front-end?
low cost – as it requires only 1 coax cable per pick-up. large dynamic range without requiring gain switching.
simple engineering auto-triggered - no requirement for external timing in the
tunnel
Why this choice of digital acquisition? uses standard LHC BPM digital acquisition card
- software architecture already in place- guarantees hardware & software support
minimal development for CNGS team- develop a single (relatively simple) mezzanine board to
convert CNGS signal to compatible format
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)8
5th September 2006
Combining a Button Pick-up with aLog Amp Acquisition System
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
-25.0 -15.0 -5.0 5.0 15.0 25.0
Measured Position (mm)
No
rmal
ised
Po
siti
on
-15
-10
-5
0
5
10
15
Res
idu
al E
rro
r (%
)
BPG ResponseCombined ResponseLog Amp ResponseResidual Error (%)
Log(A/B)
ButtonResponse
Residual Error
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)9
5th September 2006
Laboratory Results of theCNGS BPM SystemLinearity of Complete BPG Measurement Chain
-250
-200
-150
-100
-50
0
50
100
150
200
250
-10 -8 -6 -4 -2 0 2 4 6 8 10
Antenna Position (mm)
Err
or
in P
osi
tio
n R
ead
ing
(m
icro
ns)
Horizontal ErrorVertical Error
Last 2 BPG & BPKG
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)10
5th September 2006
Performance with Beam -One Day CNGS Beam Stability
Unstable beam intensity from PS
No beam
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)11
5th September 2006
Performance with Beam –Stability of CNGS Target BPM Reading
1 ADC bit ~ 7m
H-plane dominatedby beam stability
V-plane close toelectronic resolution
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)12
5th September 2006
Performance with Beam –Target Position Reading with Intensity
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)13
5th September 2006
Problems EncounteredDuring BPM Commissioning
Auto-trigger Circuit:— No triggers at all for low intensity & haphazard
triggering for high intensity Same circuit used for years in PS to SPS transfer
line— Circuit tested with circulating beam in SPS
Trigger found sensitive to bunch length & batch structure Injection
Extraction
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)14
5th September 2006
Problems EncounteredDuring BPM Commissioning
Auto-trigger Circuit (cont):— Diagnosis
Short bunch lengths produced switching times which were too fast for the auto-trigger circuit to deal with.
—Solution At first possible access, all front-end cards were
removed All were modified with a quick-fix & re-installed
within 1 day
— Performance after quick-fix Correct functioning for high intensity, still a few
circuits which did not trigger all the time with very low intensity
— Longer Term Solution Re-design of auto-trigger circuit based on this
experience with beam!
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)15
5th September 2006
Problems EncounteredDuring BPM Commissioning
Other Problems:— Large offsets in 2 monitors after 1 week of running
Suspected short circuits in recuperated LEP buttons.
Will attempt to “burn away” shorts using HV discharge.
Horizontal plane
— Last two line monitors seen to give spurious readings on first shots when beam turned back on after being off a few cycles Currently no explanation for this based on BPM
system. Same front-end crate as 3 other monitors which
work OK Requires more study with correlation between BPM
& BTV to determine if it’s beam or monitor related
NBI2006 - Thierry Bogey, Rhodri Jones & Ralph Steinhagen (CERN - AB/BI)16
5th September 2006
Commissioning of theCNGS BPM System
Summary of NBI2005 – Open Issues— Linearity of BPM system
Similar front-end gives good results in PS to SPS line
Test bench available once electronics ready
— Performance of target pick-up in air Will only be possible to evaluate with beam!
Summary of NBI2006— Linearity of BPM system
Accuracy seen to be better than 250m over ±10mm range
Resolution at the 20m level for nominal beam
— Performance of target pick-up in air Performs with the same characteristics as the line
BPMs— Problems encountered during commissioning
A few teething problems that will be solved for future runs
