A Beam Condition Monitor Investigation for CMS

1
A Beam Condition Monitor Investigation for CMS

• Beam accidents scenarios.
• The machine Interlock System.
• The DCS and the DSS.
• The BCM.
• System possibilities.
• Proposed prototypes
• Test beam at T7
• Conclusions and Outlook.

Luis Fernández Hernando, UNIL- EST/LEA-CMS 2003
2
Beam accidents scenarios The dose rate during
normal operation is 16 mGy/s Unsynchronized beam
abort dose rate is 38 kGy/s ie 106 orders of
magnitude increase Our Question Can we
implement a monitoring system to provide
protection for our detectors?
Worst Case Scenario Unsynchronized beam abort.
Occurs over 300 ns.
Deterioration of beam conditions due to equipment
failure will look similar to the above, but will
develop over the ?sec, msec timescale.
3
Failures that lead to beam loss where the BCM
should act in time to prevent major damage
Name Operation Mode Loss Location ?T
D1 warm Collision Triplet/collimator 5 turns
Damper Injection Arc/triplet 6 turns
Warm quadrupoles Any Collimator 18 turns
Warm orbit corrector Collision Triplet/collimator 55 turns
RF Any Arc/triplet/septum 55 turns
D1 warm Injection Arc/triplet/collimator 120 turns

• The BLM has one turn resolution.
• A D1 failure is the most critical. Dipole magnet
  failures cause orbit distortions.

4
D1 Failure
A power converter failure for the D1 magnets in
IP5 leads to a particle impact at the primary
horizontal collimator in IR7. It takes 12 turns
until the displacement of a fraction of 10-5 of
the initial number of particles has exceeded 6
sigma in that place.
5
Machine Protection
Primary collimator
Secondary collimator
Triplet
Absorber
Absorber
Critical apertures in units of beam size ?
14?
10?
7-8 ?
6?
I.R. 5
I.R. 6
I.R. 7

• The machine protection already ensures the
  integrity of CMS in case of unsynchronized beam
  abort.
• The BCM will be an auxiliary (and monitoring)
  system for protecting the experiment.
• In case that the beam arrives to the collimators
  with a deviation of 2-3 ? it could scrap the
  triplets in I.R. 5.

6
Machine Groups Interlock System

• 16 Beam Interlock Controllers
• 2 fast links
• if one loop open ? Beam Dump

BIC
BIC
BIC
BIC
BEAM DUMP
Pt.5
CONTROLLERS
BIC
BIC
CMS
Pt.4
Pt.6
RF
Beam Dump
BIC
BIC
Betatron
Pt.3
Momentum
Pt.7
cleaning
cleaning
BIC
BIC

• 2808 bunches on beam separated 25 ns
• Kicker magnets rise time 3 µs
• Gap in beam of 3 µs

LHC-B
ALICE
Pt.8
Pt.2
ATLAS
BIC
BIC
BEAM 1
Pt.1
clockwise
BIC
BIC
BEAM 2
counter-
clockwise
BIC
BIC
BPC
BPC
BPC
7
Interlock System
Inputs
Output
Beam Permit
Collimators
O R
Powering Interlock
8

• DCS
• Monitoring and control of the detector
• DSS
• Safeguard of experimental equipment
• BCM
• Input into DSS.
• Protect subdetectors from adverse beam conditions

9

Beam Conditions Monitor Protection against fast
beam losses Independent action from the DSS 2
collars of sensors around the beam pipe near
the pixel region and more sensors located near
the TAS BCM geometry must allow for the
detection of showers within the experiment that
result from beam deterioration
4.3 cm
2 m
I.P. 5
Decision box
Digital signal to interlock
Digital signals from sensor readout
Digital signal to DSS
DSS abort signal
BCM sensors
DSS backend
Analog signals from sensor readout
10
System Possibilities

• The sensors that can be used are
• CVD diamonds good radiation hardness.
• Will get samples for next test beam experiment.
• Silicon widely used in other applications.
• May be suitable for more accessible locations.
• CdTe Being considered.
• Quartz No need of biasing the sensor and fast
  signal.
• Yet to be investigated.
• System readout for the diamond/silicon/CdTe
  approach
• Current amplifier simplest solution. Analog
  reading.
• APVB binary response chip. More complicated.
  Signal already treated.
• Readout chain available and preliminary test
  setup built.
• CARIOCA Fast amplifier, and comparator. Test
  boards available next week.
• APV25 Investigating possibility of running in
  conjunction with APVB.

11
CVD Diamond Sensors
Material with outstanding radiation
hardness Ionization chamber. The energy necessary
for creation of an electron-hole pair in diamond
is 13 eV (in Silicon is 3.6 eV) A mip traversing
100 µm of material produces 3600 eh-pairs (in
Silicon 8900) The bias is of the order of 1
V/µm Fast charge collection Silicon shows better
resolution than diamond for tracking of particle
hits but for the BCM spatial resolution is not as
important as radiation hardness
12
CVD Diamond Sensors

• For a 1 cm2 sensor area, with collection distance
  of 150 µm, located at a radial distance of 4.3 cm
  from the beam we have that
• During normal operating conditions, dose rate of
  1.66E-2 Gy/s, per 100 ns time bin the MIP
  equivalent fluence passing through the sensor is
  on average 5.9 MIPs. Expected signal of 51 nA.
• In the case of D1 failure at the same level as
  an unsynchronized beam abort the flux per 100 ns
  is 2.2E9 MIPs. This implies a current of 20 A
  !!!!.
• These two extremes imply a large range of signal
• Not possible to deal with the full range !
• Will focus on the need for a readout chain that
  is sensitive to the development of adverse beam
  conditions.

13
A pattern of bits, with a clock signal and a
command line, is sent by a data generator to the
chip
System Readout The readout chip that has been
tested for the preliminary investigation is the
APVB This chip has an internal frequency clock
that can be adjusted for seeing the beam
crosses It reads the current signal from the
sensors and compares it with the set threshold,
giving a binary response This digital response is
afterwards treated in the decision box
The APVB sends a string of 0s and 1s that has
to be decoded This response is given after 7 µs
processing delay, limiting the readout frequency
to 0.14 MHz
Sensors
PLD/FPGA
Response to be treated in the Decision Box
Output data can be handled by an FPGA or a
Programmable Logic Device
14
Strategy for readout   The readout from the
sensors is compared with 2 threshold levels.  
Low threshold
High threshold
I.P. 5
0
0
15
Test Beam Plans

• Date During the 8th- 20th Oct LHC irradiation
  period
• Place T7 irradiation facility in the CERN East
  hall
• Beam 24 GeV protons in fast extraction spill
  from the CERN PS
• Each spill 3.6 x 1011 protons
• Beam Time one 8-hour machine operators shift
• 2-stage programme is proposed
• Stage 1 Repeat of the 1-shot testbeam
• 2 spills separated by 256ns
• Target flux 109 protons/cm2 at centre of beam
  spot
• Approximates to unsychronized beam abort
  scenario
• Stage 2 Single spill running
• Lower intensity beam spot
• Controlled beam loss on the T7 beamline to be
  attempted
• Programme to be set out once sensors are up and
  running
• Test beam to be done in close cooperation with
  the PS machine operators

16
Conclusions

• Have identified beam loss scenarios that could
  be problematic to CMS sub-systems.
• The worst case unsynchronised beam abort is
  used to define the fluence, and this sets the
  sensor constraints and overall system design.
• A BCM development is in terms of beam loss
  scenarios that we can detect and react to.
• CVD diamond sensors now metalised and arriving
  in June. CVD diamond is our primary sensor
  candidate for the upcoming Test Beam.
• An APVB setup has been built and tested.
• The Carioca chip will be tested this month
• A test beam programme is in preparation (October
  2003).
• Present efforts done on a restricted equipment
  budget ( help from friends)
• and still considering different BCM design
  options