Minutes: ACTAR collaboration meeting
Santiago March : 11-12
Patricia Chomaz made an introduction about ACTAR project status.
Deliverables : Reports before the end of the year 2008 (End of the JRA)
Task 1.1 : Yellow book -> done
Task 1.2 : Detailed simulations
Task 2 : Test modules
- Detection/target gas
- Readout chamber : Wires (GANIL), GEM (CENBG), Micromegas (IPNO)
- Magnetic field configuration tests
Task 3 : Electronics and data acquisition
Task 4 : Data tracking
What remains to be spent : Saclay Engineer : see Lolly (Urgent)
There are still many things to do to provide deliverables before the end of 2008
Hector Alvarez : Presentation of the work programme ; ActarSim development programme
Hector made a sort review of the ActarSim code base on GEANT4/ROOT
Geant4 is used for Physics process and Geometry and Root for Drift , Digitilization (Maya like, Tactic like, TPC like) and Tracking process and data storage.
Code development: a version control system is available on the web site (document modifications, …)
Work to be done : ancillary detector coupling, pads geometry, … a detail work-plan has been defined for both geometry
3 developers: E. Benjamin, P. Danyang and Hector Alvarez
Tracking formalism is not yet ready
Pang Danyang reports on his simulations about 78Ni(d,p)79Ni @ 5-10 MeV/u and ActarSim development.
A beam energy of 10 AMeV is mandatory to get enough proton energy to be detected inside ACTAR.
Different Range are obtained with different GEANT4 libraries (to be checked)
Efficiency of the reaction 78Ni(d,p)79Ni @ 10 MeV/u have been shown for cubic geometry using Si wall (left, right, front side), an average efficiency of 50 % is obtained.
Work has to be done now to optimise geometry with drift, digitalisation and tracking.
A discussion starts around amplification GEM, Wires and charge spread.
Even without charge distribution on neighbour pads, we can measure it on PA current distribution. Shaping of PA could be adapted as well as the distance between GEM and pads.
Action : Some tests are scheduled at MSU (Wolfi).
E.A. Benjamin reports on her simulations about 78Ni(d,p)79Ni @ 10 MeV/u and ActarSim development.
Cylindrical geometry : radius 30 cm and length 50 cm with magnetic field 1-2-3 T, and pressures 400, 850, and 1013 mbar.
Efficiency plots were shown for a vertex position Z=10cm and 25 cm.
The efficiency for the protons which stop inside the gas, for all studied cases is :
1) within 45% to 65%, for Z=10 cm
2) within 54% to 64%, for Z=25 cm
Future : projection into the pad plane, tracking ….
Simulation work plan:
- Cubic geometry : Super Maya
ancillary detectors coupling (Si and CsI), optimisation volume/pressure/energy loss of the beam/pad side.
Key points : ancillary detector vs electronics implantation (pads size)
Goal : decrease the pad size
- Cylindrical geometry: Magnetic field
Projection on the end caps
For MSU 1m long and 50 cm diameter
Pads geometry = 10000 channels
Triangle shifted by ¼ in case of GEM and Micromegas (no charge distribution)
For ACTAR 50 cm long and 30 cm diameter
Number of Pads : 10000 to be checked
After 72 channels : 7cm x 0.5 cm
Wire and Hexagonal
Meeting : Dedicated visio conference before summer.
R. Raabe : Maya Status
- CsI test (80) Alpha source bias 50 V and 30 V
- Resolution about 3% (Only 3 detectors are not working properly)
- Separate HV control is desirable
- New anode with new Gassiplex (SHARAQ)
- Gas mixture test : He + CO2 : (3He,d) and (3He,4He)
- looks ok difficult to control mixture
- Additional test necessary
- Proposed experiments : (3He,4He) Ricardo and Roy and resonant elastic scattering (check with Alain)
H.Savajols: ACTAR@GANIL
The effectiveness of the active target concept was presented. This was illustrated by impressive results obtained in the domain of extremely neutron rich nuclei, new radioactivities, collective excitations of exotic nuclei. The active target enabled important breakthroughs, allowing to perform experiments which were not possible with any other method. However, the first generation active targets present intrinsic limitations, linked for most of them to the electronics available at the time of their conception.
A cubic geometry device was proposed for the studies of reactions induced by SPIRAL2 beams with the following characteristics :
chamber type / rectanglesize: / diameter / cm
width / 20 / cm
length / 20 / cm
heigth / 30 / cm
Readout plane / Micromegas, Wire
Dead Zone / 0 / %
Pad size / length / 2 / mm
width / 2 / mm
number of Pad / 10000
gaz / ALL available
drift speed / 200ns/cm to 2µs/cm
drift Time / 4µs to 50µs / µs
environment / Electric Field / V/cm
Magnetic Field / No / Tesla
Vacuum / 10-6 / TOR
Pressure / 20mb tp 2 atm / mBar
temperature / ambiant / °C
Temp. Variation / 5 / °K
mean event rate / 100 to 1000 / Bq
number of touched pads/event / 1 to 10 / %
Synergy with GASPARD.
The final decision on geometry for active target dedicated to reaction studies will be taken at the end of the simulation work.
W.Mittig (MSU) : AT-MSU
Boundary conditions : 1 M Dollars
New accelerator : 3 MeV/u available in 3 years
Dipole or Solenoid : 32Mg(d,p) simulations : At this energy you are below the Coulomb barrier and the maximum cross section are at forward laboratory angle
Is it really direct reaction ? Determination of l value could be done via excitation cross section
2T and 0.500 atm
Best is a solenoid cylindrical for trajectory reconstructions and beam shielding
Case 70Ni(d,d’) at 100 MeV/u
chamber type : / Cylindricsize: / diameter / 50 / cm
width / cm
length / 120 / cm
heigth / cm
Readout plane / Gem, Wire
Dead Zone / 5 / cm
Pad size / length / 5 / mm
width / 5 / mm
number of Pad / 10000
gaz / ALL available
drift speed / 200ns/cm to 2µs/cm
drift Time / 4µs to 50µs / µs
environment / Electric Field / V/cm
Magnetic Field / No / Tesla
Vacuum / air / TOR
Pressure / 20mb tp 2 atm / mBar
temperature / ambiant / °C
Temp. Variation / 5 / °K
mean event rate / 100 to 1000 / Bq
number of touched pads/event / 1 to 25 / %
MD Salsac (IRFU) R3B TPC
TPC after the Magnet
A versatile setup used for multipurpose physics
6000 channels for the readout
amplification : micromegas 2x5 m2
0.9x1.2x3.2m
Z=1 to Z=92, 104 dynamics !!!! so 2 gaz amplifications
12 bit ADC for Z determination
Horizontal position : Charge amplification
Vertical position : time projection
End 2012 – Saclay collaboration
P.Egelhof (GSI): Active target requierement for R3B status
- GSI today Prototype experiment : IKAR in front of ALADIN was described
Elastic proton scattering (matter distribution oh 8He,11Li,14Be,8B)
- With FAIR : Gain facteur 104 more intensity than present GSI
Super FRS + High energy cave or storage ring
Physics : Light ion induced direct reactions at low momentum transfer : Matter distribution, Giant monopole resonance (nuclear compressibility and new collective mode) and GT strength.
Requirement targets : H and He target and Low recoil energy detection
- EXL project is the best place to study low momentum transfer BUT at R3B and active target could overcome the half life limitation time (Not accessible to EXL)
- R3B active target : H2 at 20 bar (1m by 1m) with a beam shield
Beam tracking + vertex reconstruction
According to luminosity R3B TPC will not compete with EXL but will able to address the most exotic nuclei with short live isotopes.
Time schedule : first beam 2012 super FRS, but delayed 2014.
R3B active target is somehow complementary to EXL (short lifetimes, small beam intensities)
Questions :
- requirement for other reactions
- magnetic field or high pressure
- shielding of the large emittance beam (2cm)
General discussion about synergies :
FEC board : 30cm x 1cm connector for 256 channels
- Super Maya Type : if we have 20 x 30 cm we can put 5120 channels pads limitations is 10 mm2
- MSU cylindrical : 25 mm2 pads size
BUT limitation is coming from the FEM board which is managing 6 FEC board.
Common DAQ: See Lolly
Dynamic ranges 1pC for Bordeaux
Rate ranges : MSU 1000/s is not possible but one way is not to read the 511 cells but 128
With T2K equipement, the limitation is 7cm in between each FEC board !!!!
Need to know readout wire, GEM or Micromegas (protection are different).
What about the cooling ?
Front-end part, we can gain a factor 2 in density as compared to what exists now.
à DAY 2 discussion :
Wolfgang Mittig : AFTER+ and application
ACTAR
TPC/GLAD/R3B/FAIR
BordeauxTPC
Is it possible to use the same asic for Si detector ? Sampling will be different Ge, CsI and LaBr3.
CC After+
72 channels is now available, but this can be reduced 64
What happens if a channel is not working ?
Dynamic : 120fC to 10pC
Peaking time increase to 3micros for CsI
External PA options
Sampling 200 MHz
Selective readout on the ASIC and FEM
ADC 13 bit
Time synchronisation
Rates : 1000 Hz events recorded
Working hypothese : Minimum changes
With 50ns x 511 x 72 = 1.8 ms can we reduced the number of cells to 128 to gain a factor 4
Trigger : external and internal
External trigger : be able to put delay to make coincidence with pads signal (leading edge)
Internal trigger : multiplicity, readout SCA nsca =2 x tdrift x fsampling
FEC -> FEM data transfer no limitation
RIB 1/1000 of the total words are transmitted
Collision x 25
Eric Delagne
What is different from high energy phys
- auto-triggerable
- large dynamic range
- low noise
- multiple sampling (For tracks with angles, for common mode rejection)
Two chips were presented : altro
No dead time
most advanced chip including A/D
zero dead time operation
power 35mW/ch
16 ch/chips only
external PA
PASA treats only cathode signals
Baseline correction limited
Work on the go : integration 1x1mm pads size for ILC
F. Druillole
Brief presentation of the T2K project (124000 channels)
FEC + FEM + DCC (for a large number of channels, need of 2 levels concentration)
CERAT : Complete electronic readout active target
Phase A organisation => conceptual design report
Scientific Manager and Technical Manager
Goal : definition of needs, feasibility studies
- Physics information
- Detector information
- System information
Exel file to be fill up
As we only manage 10000 channels, can we go directly from the FEM to the computer (without DCC)
Example of Antares have been shown, when they reach 350kBq with an event size of 6 bytes
Antares like data stream is proposed
Action: Write a technical specification report before the end of the year
P. Baron A review of AFTER+ was presented
- 3 charge ranges : 120fc (750keV), 1 pC (6.25MeV), 10 pc (62.5MeV) adjustable channel
- Charge measurement : Output dynamic range 2Vmatch the ADC specification (12bit ADC)
With INL <2%
- Peaking time : 16 values: 50ns to 1 ms; adjustable/chip
- Charge resolution
Configuraion: Charge Range 120fc, peaking time 200ns, Cin Asic < 30 pF
Asked < 600 electrons rms; possibility < 850 electron rms
- SCA : memory cells 511
- Sampling frequency : 20 Mhz to 25 MHz
- Time resolution : correlated to the sampling frequency, Jitter<2ns
- Discriminator : LED, inhibition / channel
- Trigger output : current : Sum of 72 discriminators
Trigger time resolution : the trigger time resolution will depend on the input charge, threshold & peaking time value
- Input dynamic range :5% of asic input dynamic ; INL < 5%
- Threshold value : common DAC 3 bits + 1 bit of polarity
- Individual DAC: 4 bits To be confirmed
- Comment : DACLSB = 0.04% of asic input dynamic range
- Minimum threshold value: 120fc; 30pF; 200ns. Minimun = 3 keV (0.5fc; 0.4% of asic dynamic range)
- Readout : 20 to 25 Mhz
- Readout mode : all channels, Hit channels or Specific channels
- Reduce the number of the SCA : 511, 256 or 128 SCA cells
- Internal readout buffer :
Calibration : External capacitor (test 1 channel/72)
Test : 3 internal capacitors
Functional : 1 internal capacitor/channel; test on 1, few or all channels
- Counting rate : 1 kHz
- Power consumption < 10 mW/channel
- Digital multiplicity or not !!! We keep the analog multiplicity output
Gilles Wittwer : Reflections on how to deal with the DCC+ and Integration in the GANIL acquisition
Questions :
What will be the data rates ?
How will be used the multiplicity ?
Locally on the FEMs
What kind of intelligent trigger ?
Preprogrammed at the front end but needs also an external validation
What about clock and clock distribution ? Centrum like 100MHz and 48 bits
Existing COTS : example of a slow control and readout without DCC
1000 Hz and 10000 channels need to make zero suppression
Data flow Gbits
Some tracks for a concentrator (controller) were presented. A mix of commercial and dedicated electronic is probably a good solution, especially with the lack of manpower in Lab.
Two years of development
Budget : a board = 5000 Euros
General discussion
For electronics and DAQ : minimum of 3 years developments
AFTER+ : Saclay
FEC : Saclay – Bordeaux – UK ???
FEM - Saclay– Bordeaux – UK ???
Control – PC: GANIL/MSU/UK ???
Bordeaux : Summer 2008
UK : beginning 2009
MSU :
GANIL : beginning 2009
Cost estimate of the new electronics : AFTER+ & FEC & FEC & DCC
Prototype : 50 kEu
Cost : 150 kEu ASIC production for 30000 cx
FEC : 500-600 Eu card : 256 channels
FEM : 1500 Eu card
Concentrator 5000 Euros (2 FEM)
Controler : 5000 Eu card