Haik Simon
Haik Simon
Haik Simon
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
Time stamping and online data processing for<br />
NUSTAR@FAIR.<br />
H.<strong>Simon</strong> - GSI<br />
Exotic Nuclei<br />
• Spectroscopy<br />
• Reactions<br />
• Mass/gs. prop.<br />
schematic<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Beam Tracking/ID for Experiments !<br />
TOF, dE<br />
TOF<br />
TOF, dE<br />
TOF, dE<br />
TOF, dE<br />
10 12 /s
Continous beam ID is integral part of experiments<br />
Example: 132 Sn PDR studies (reaction study @ CaveB)<br />
Primary: 3*10 8 238 U/spill @550Mev/u<br />
Secondary (mixed): 50 ions 132 Sn/spill<br />
A<br />
Z<br />
=<br />
m u<br />
c Bρ<br />
e βγ<br />
Bρ – from position at<br />
middle focal plane<br />
of the FRS<br />
β – from TOF<br />
A/Z<br />
2.66<br />
2.64<br />
2.62<br />
2.6<br />
132 Sn<br />
Z – from ∆E<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting<br />
2.58<br />
48 49 50 51 52 53<br />
Z [e]
Bρ-ΔE-TOF method: Requirements<br />
NO CHARGE STATES !<br />
Bρ = A/Z β γ<br />
TOF=L/β<br />
∆E~Z 2 /β 2<br />
A/Z, P<br />
Z<br />
Pos res. σ ≤ 1 mm<br />
Timing res. σ : 50 ps<br />
∆E resolution σ : 1-2 %<br />
• Position:<br />
• ΔE:<br />
• TOF:<br />
Wirechambers (single event readout)/Diamond<br />
MUSIC/TEGIC limit about 1MHz<br />
Plastic/Diamond<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Integration/Controls<br />
<br />
<br />
Challenge:<br />
Increasing complexity of setups.<br />
Typically few 100 channels <br />
up to several 100’000 channels.<br />
Prohibitive setup times.<br />
Solution:<br />
Integrated (frontend) electronics.<br />
VA64.hdr Si readout<br />
Modularity (distributed DAQ systems).<br />
Scalable Controls (e.g. EPICS + various frontends).<br />
“Exploder”<br />
MBS FEE<br />
combiner<br />
FEE projects: Tacquila, NXYTER, ASDQ, APV25<br />
integration into running setups (MBS).<br />
FPGA trigger implementations & diagnostics.<br />
EPICS implementation studies (also used for ITER). Logic Module<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting<br />
& softscope
Fast sampling & timing techniques<br />
<br />
<br />
Challenge:<br />
Beam identification at rates up to 1MHz.<br />
ToF over km distance with sub-ns resolution.<br />
∆E resolution 2-3%<br />
Solution:<br />
Fast sampling and FPGA based digital signal<br />
processing & pulse shape analysis.<br />
Campus wide Time Distribution System based<br />
on FAIR BuTiS timing system.<br />
TAC or DLL based Frontends.<br />
Tacquila<br />
σt 10ps electronic<br />
50 kHz events/s<br />
Online PSA @ 100MHz<br />
First studies using Tacquila@R³B/Cave-C.<br />
<br />
Digital Signal Processing (for PSP, MUSIC) in<br />
collaboration with KVI Groningen/JSI Lubljana<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting<br />
σt < 100ps/km<br />
BuTiS campus clock
NUSTAR DAQ/controls architecture<br />
(MBS[/DABC] & EPICS)<br />
• smooth transition from existing systems<br />
• triggered systems will be replaced whenever they cause<br />
excessive dead time or<br />
• … are difficult to handle<br />
Examples: (1) incoming tracker (deadtime)<br />
(2) calorimeter (100’000 channel or ;o)<br />
(3) delayed coincidences<br />
… so don’t forcefully leave out trigger capabilities<br />
and don’t force triggers to be there <br />
Building blocks<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Ian Lazarus<br />
R3B conceptual design<br />
for Si tracker FEE and DAQ<br />
Draft 1 April 2008<br />
Si Ladder<br />
DSSD<br />
ASIC<br />
ASIC<br />
NUSTAR<br />
DABC<br />
DAQ<br />
BUTIS<br />
R3B<br />
input<br />
code<br />
module<br />
fibre<br />
links<br />
Gbit Ethernet fibre(s)<br />
BUTIS<br />
Fan-out<br />
either fibre<br />
or HDMI links<br />
FEE<br />
Card<br />
Handles<br />
16 ASICs,<br />
each 64ch<br />
= 1024ch.<br />
Total approx<br />
100 FEE cards<br />
Fibre<br />
Ethernet<br />
Slow<br />
Control
Campus wide time distribution (BuTiS)<br />
P.Moritz/GSI<br />
• Campus wide time distribution via fibre optics / local TDS(s)<br />
• Synchronized local oscillators (100kHz, 10Mhz, and e.g. 200, 155 or 76 Mhz)<br />
with +/-100ps/km absolute uncertainty and
Precision timing (
BuTiS at work (20071015)<br />
P.Moritz/GSI<br />
• 10, 200 MHz sine<br />
waves (adj. phase)<br />
• T0 pulse for sync.<br />
every 100µs<br />
• very good phase<br />
stability<br />
• BuTiS oscillator<br />
can run standalone<br />
• about 10k€/system<br />
labeling of T0<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
White Rabbit Event- & TDS<br />
CERN/FAIR T. Fleck/GSI<br />
• Event and timing distribution with priorities<br />
Form factors:<br />
• GB<br />
copper/fibre<br />
P. Moritz<br />
• Piggy back<br />
VME<br />
COTS<br />
e.g. PC We !<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
H² Wörtche/<strong>Simon</strong> ● GET Meeting
First Steps: ADC coupled to Hades TRB2 at KVI<br />
Peter Schakel / Pim Lubberdink<br />
• Available hard/software<br />
environment:<br />
(1) ADC Piggy back / KVI<br />
100MS/14Bit<br />
50MHZ BW<br />
(2) Xilinx based board<br />
HADES TRB2<br />
(3) Base line follower/<br />
kσ trigger<br />
(J. Jungmann / M. Vencelj)<br />
Labview based readout system<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Test experiment S327 (16.-18.4.2008)<br />
12<br />
C: 550-700 MeV/u ; 2-50 kEv/s<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Results: Baseline<br />
Baseline follower works !<br />
(Bimodal Filter)<br />
Treatment of double hits !<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
“Intelligent sensors”<br />
N.Kurz, J.Hoffmann, W.Ott/GSI<br />
• NUSTAR uses MBS, a (in principle)<br />
triggered DAQ system …<br />
preproc./region of interest stored<br />
any<br />
LVDS<br />
input<br />
VME Readout<br />
or PCIe<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Pile up treatment (2)-1-x-1-(2): M. Vencelj<br />
shaping: 0.25µs<br />
raw<br />
shaping: 1.00µs<br />
ref<br />
corr<br />
2’x2’ NaI, 137 Cs,<br />
Heinrich Wörtche
Conclusion:<br />
Need Technical discussion<br />
based on physics requirements<br />
*Agata gets integrated ( 20090331 ;o)<br />
online analysis/controls<br />
transfer / storage<br />
event building<br />
frontend-readout<br />
MBS, DABC, Narval*, … EPICS<br />
MBS, user specific (Labview, FAIR-Contols, user spec.)<br />
Bottom up: systems defined, building blocks, …<br />
use existing experiments as testing ground ! manpower !<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
Systems:<br />
H² Wörtche/<strong>Simon</strong> ● GET Meeting
NUSTAR DAQ organisation<br />
Related<br />
NUSTAR Slow Control<br />
Heinrich Wörtche<br />
…<br />
‚GSI‘ DAQ<br />
Nikolaus Kurz<br />
NUSTAR COUNCIL<br />
NUSTAR DAQ<br />
Thomas Nilsson<br />
<strong>Haik</strong> <strong>Simon</strong> (coord.)<br />
Michael Böhmer<br />
Michael Block<br />
Christophor Kozhuharov<br />
Nikolaus Kurz<br />
Ian Lazarus<br />
Johan Nyberg<br />
Lolly Pollacco<br />
Heinrich Wörtche Alpi<br />
…<br />
Spiral2<br />
EXL<br />
Ian Lazarus<br />
R³B<br />
<strong>Haik</strong> <strong>Simon</strong><br />
DESPEC<br />
Johan Nyberg<br />
Experiments<br />
…
Change language