Time of Flight (ToF): basics

1
Time of Flight (ToF) basics
Start counter
Stop counter

• TOF General consideration
• - early developments combining particle
  identifiers with TOF
• TOF for Beam Detectors or mass identification
• - TOF Constituents - based on the use of
  SEE effect
• - Thin Foils (SE generation)
• - SE transport
• ------------------------
  --------------------------------------------------
  --------------------------------------------------
  ---------------------------------
• - SE detection ( mainly MCP some
  basic set-up )
• Fast electronics
• - Fast preamplifiers and discriminators
  LE CFD ARC-CFD
• - Time walk and jitter basic
  consideration

2. part
2

• Timing measurements
• Pulse height measurements discussed up to now
  
• emphasize accurate measurement of signal
  charge
• Timing measurements optimize determination of
  time of
• occurrence ? timing output signal ( time
  stamp )
• For timing, the figure of merit is not the
  Signal / Noise ratio
• but the Slope / Noise ratio

3

• Fast Timing
• Timing measurements ? Detectors for
• Timing and their
  FEE
• - Scintillator Photomultiplier
  assembly
• - MCPs Fast Preamplifiers
• - Semiconductor detectors
  Preamplifiers
• 
  ( CSP vs. Current )
• b) Ultra-fast Timing Circuits and Signal
• - Time - stamp
• - Time - walk and Time - jitter

4
(No Transcript)
5
3rd layer 1st layer 2nd layer
300 ps 250 ps 350 ps
244Cm 241Am 239Pu
Time of Flight ns
( 200 keV energy loss )
Counts
IKP - TOF BPM Preliminary results - 250 /-
50 ps - coincidence with energy
measurements (SC DGF-4C-rev.F) - transparent
beam detector and tracking with 32x SC matrix
as Stop detector ? (real beam test is
requested!)
Counts
239Pu 241Am 244Cm
5.155 5.486 5.804
MeV
6

• a) Timing measurements ? Detectors for
• Timing and their
  FEE
• - Scintillator Photomultiplier
  assembly
• - MCPs MCP-PMT and Fast
  Preamplifiers
• ? and very briefly about other ultra-fast
  detectors
• - Semiconductor detectors (Si, Diamond)
  their Preamplifiers
• (
  CSP vs. Current )

7
Scintillators Photomultiplier tubes (PMT)
Detector
Photomultiplier tube - (PMT)
Gain 106 ? sec. secondary electrons /
photo-electron
Different geometries of PMT
Circular-cage type PMT
8
Box-and-grid type PMT
and the typical electron trajectories
Linear-focused type PMT
9
The transparent window material
commonly used in PMT - MgF2 crystal -
Sapphire - Synthetic silica - UV glass -
Borosilicate glass
Transmittance ()
Wavelength (nm)
Basic Photocathode commonly used in
PMT - Cs-I ? 100 M - Cs-Te ? 200 SM -
Bialkali (Sb-Rb-Cs, Sb-K-Cs) ? 400 USK -
Multialkali (Sb-Na-K-Cs ? 500 KUS - Ag-O-Cs ?
700KS-1 - GaAsP(Cs)
Photocathode Radiant sensitivity (mA/W)
Wavelength (nm)
10

• Photons
• transport
• WLS

WLS Wavelength shifter
Detector
11
BriLanCe Crystals - Properties (1)
12
BriLanCe Crystals - Properties (3)
s/n!
13
BriLanCe Crystals - Properties (2)
14
Signal output problem
and the solution

Countermeasures for very fast response
circuits (the miraculous (small) series
resistor and not parallel capacitor)
The effect of damping resistor on ringing (
remember the influence of resistor in the
quality factor of an oscillator or larger
capacitor value in a low-pass frequency filter
) C - filter-change the frequency ! Rs
oscillation damper !
15
The importance of Poles and Zeros
Pole-zero diagram
? Ideal oscillator
? real oscillator - R series
e.g. 10 pF 10 nH ? 500 MHz
Step 2
Step 1
Step 3
16

• Going from PMT ( Photomultiplier Tube)
• to MCP (Microchannel Plate)
• from a discrete dynode structure to a
• continuous distributed dynode structure
  but also
• more than 8 orders of magnitude scaled down
• design in volume
• ( 102 in length and gt 103 in diameter )

17
Multi-channel Plate Detectors








Electroding (on both face)
Channels
- e initial velocity ( 1eV) -
channel length/diameter ratio
Kc - constant
18
Metallized
Metallized

• Much stable operation vs. external high magnetic
  fields in comparison with PMT
• lower gain but in chevron configuration the gain
  106
• lower power consumption (gain vs. HV)

MCP assembly in chevron configuration
19
MCPs in Single, vs. Chevron and Z-stack
configurations
Gain 103-104 106 108
20








MCP gain dependence vs. - parameter and
stage configuration
MCP gain dependence vs. channel diameter
and technology
Comparison of gain characteristics of three
different types of 2-stage MCPs
Comparison of gain characteristics of various
single and multi stage MCPs
21
Parameter 6 µm Channel 12 µm Channel
Rise time ps 167 245
Fall time ps 721 716
Transit time ps 406 650
Transit time spread ps 67 81
Comparison of timing characteristics of
chevron 2-stage MCP-PMT, one with 6 µm and
another with 12 µm pore diameter
22
Time x10 -8 s
23
Mesh form anode ( e.g. X,Y delay
lines ? signal pulse amplitude only 15-20
compared to the solid anode standard version)
?

• Hamamatsu
• R-3809 U-50
• Photocathode
• diam. 10mm
• Price - ? !

HV 3 kV Rise time 150ps Fall time 350
ps FWHM 300 ps
Standard operating circuit for an MCP-PMT
24
Anode Return Path Problem
Current out of MCP is inherently fast- but return
path depends on where in the tube the signal is,
and it can be long and so rise-time is variable
Incoming Particle Trajectory
Signal
Would like to have return path be short, and
located right next to signal current crossing
MCP-OUT to Anode Gap
The Signal is a current and not a potential
Signal Return
10cm wire 0.2mm diam ? 150 nH Impedance _at_ 1Ghz
1 kOhm
10 pF impedance _at_ 1GHz 1.5 kOhm
Load
25
Detector Signal Collection
High Z
Circuit development
Low Z

• Impedance adaptation
• Amplitude resolution
• Time resolution
• Noise cut

Voltage source

Zo
Z
Rp
-
Low Z
T
Quo vadis ?

• Low Z output voltage source circuit can drive
  any load
• Output signal shape adapted to subsequent stage
  (ADC)
• Signal shaping is used to reduce noise
  (unwanted fluctuations) vs. signal

26
Front-end electronics overview
Detector as fast signal generator ?
electron-hole pairs collection
? only
electrons (or particles)

• if Z is high
• charge is kept on capacitor nodes and voltage
  builds up (until capacitor is discharged)
• Advantages
• - excellent E resolution
• - friendly pulse shape analysis
• Disadvantages
• - channel-to-channel crosstalk
• - pile up above 40 k c.p.s.
• - sensitivity to e.m.c.

Ci
FEE (Input stage)
27
Front-end electronics overview
Detector as fast signal generator ?
electron-hole pairs collection
? only
electrons (or particles)

• if Z is low
• charge flows as a current through the impedance
  in a short time.
• Advantages
• - limited signal pile up
• - limited channel-to-channel crosstalk
• - low sensitivity to parasitic signals
• - good timing resolution
• Disadvantages
• - pour signal/noise ratio
• - sensitive on return GND loop !

FEE (Input stage)
28
Capacitive Return Path Solution
Current from MCP-OUT
Return Current from anode
29
25mm
Ultra-fast detectors, extremely user-friendly
solutions, the only disadvantages - small area
of photocathode and extremely expensive
30
?
31
Chemical Vapour Deposition techniques
CERN - LHC experiment
CVD-Diamond
32
the CVD - Diamond Detectors
Two optical grade CVD and 100µm
thickness

• a 30 x 30 mm2 detector with 9 strips
• with a pitch of 3.1 mm and
• a 20 x 20 mm2 pixel detector with a
• pixel size of 4.5 x 4.5mm2 ?
• the first large-area CVD diamond detectors
• Installed at SIS

The largest diamond detector of 60 x 40 mm2 and
200µm thickness lt0gt in the focal plane detector
of a magnetic spectrometer
E. Berdermann et al, CVD-Diamond Detectors
Nucl. Phys. B 78 (1999), 533
E. Berdermann et al, The use of CVD-Diamond for
heavy ions Diamond and Related Materials
10(2010),1770
33
Charge Sensitive Preamplifier

• Active Integrator (Charge Sensitive
  Pre-Amplifier)
• Input impedance very high ( i.e. NO signal
  current flows into amplifier)
• Cf (Rf ) feedback capacitor (resistor)
  between output and input
• very large equivalent dynamic capacitance
• sensitivity A(?qi) q / Cf
• large open loop gain Ao 10,000 - 150,000

• very fast active integrator
• tr lt 1ns (sub-nanosecond CSP)
• A0 1,000-10,000
• Transconductance amplifier
• ASIC integrated structure

?Qi
Ci
E. Berdermann et al, The use of CVD-Diamond for
heavy ions Diamond and Related Materials
10(2010),1770
34
Ultra-fast branch of a CSP
35
Standard current amplifier solution
G2
G1
G1 gt G2 ? to minimize S/N ratio
HSMP 3862 series
tr 1.2 ns (10 to 90)
Imax (1µs) 1A Peak Inverse Voltage 50V Tj Max.
Junction Temperature 150C
(OK to be used in vacuum)
Simulation results of the amplifiers with
THS 3201 ultra-fast current feedback amplifier
36
Signal Output ? A1. A2. A3 . e
Noise Output ? A1.A2.A3. e1 A2.A3. e2
A3.e3 ? the gain of the first block
of amplification must be kept as high as
possible, in order to reduce the importance of
the noise contributions coming from the following
blocks i.e. the preamplifier gain has to
be as large as possible ? ! Ao gtgt10
4
37

• Ultra-fast Timing Circuits and Signal
• -Time-stamp
• - Time-walk Time-jitter as perturbation
  effects
• Timing time stamp but actually timing
  means measurement of time intervals (from fs to
  ms)
• Walk effect - variation of time stamp
  (timing) caused
• by signal variation in amplitude
  and/or rise time
• Jitter effect - timing fluctuations caused
  by noise and/or
• statistical fluctuations in the
  detector (intrinsic noise)
• ? two identical signal will not
  always trigger at the
• same point (time stamp) time
  variation dependent
• on the amplitude of fluctuation
  slope/noise ratio

38
(No Transcript)
39
Fast Pulse Shaping
40
MVP in fast time domain
tra tc
The noise bandwidth approaches the signal ?
bandwidth
? the timing jitter
41
- the Ortec 579 to slow for fast timing
New fast amplifiers - Ortec 9327 (1
GHz Amplifier and Timing
Discriminator) - Ortec 9309-4 ( Quad Ultra-Fast
Amplifier) - Ortec
9306 (1-GHz Preamplifier)
42
(No Transcript)
43

• this is the reason while only
• 1-2 amplifier stages
• this can be implemented only
• if the product gain x bandwidth
• of the amplifier is large enough ? !

10 cm wire 0.2 mm diam ? 150 nH Impedance _at_
1Ghz 1 kOhm
1 pF impedance _at_ 1GHz 150 Ohm
Simulation results of the amplifiers with
THS 3201 ultra-fast current feedback amplifier
44
Gain 7 (th. 10)
Gain 10 (th. 20)
Current Feedback Amplifier THS-3201
Main features - 1.8 GHz - 6700 V/µs _at_ G
2V/V RL 100 Ohm - 18mA _at_ /-
3.3V (120mW ? vacuum)
tr 1.2 ns (10 to 90)
Simulation results of the amplifiers with
THS 3201 ultra-fast current feedback amplifier
45
Wire impedance ? skin effect
(i.e. skin depth calculator)
R0 1 /pro2 s (DC low frequencies) - s
bulk conductivity - r0 wire radius L0 µ /
8p - µ permeability (µ0 4p.10-7
Henry/Meter) Rs 1/ (sd) q v2 r0 / d
d is the skin depth ? (pfµs)-1/2

• to remember about skin effect
• Material dependence
• (e.g. Ni vs. Cu skin effect
• depth one order of magnitude)
• - Frequency dependence

- this calculator only cover the range q lt
8 Which roughly correspond to r0/d lt 5 above
this value the Bessel functions become hard to
evaluate
46
Time walk
Time walk for a fixed trigger level ? time
stamp (time of threshold crossing) depends on
pulse amplitude

• Accuracy of timing
• measurement is limited
• by two factors
• time jitter ( to the slope/noise)
• time walk )
• (due to dependency on signal
• amplitude and rise time variations)
• ) - if the rise time is known and constant,
• the time walk can be compensated in
• software event-by-event by measuring the
• pulse height and correcting the timing
• - if rise time vary (e.g. HP-Ge Det.)
  this
• technique fails! ? PSA required

Hardware - threshold lowest practical level
(i.e. gt noise) or
compensation technique (e.g. CFD)
47

• LE method
• timing occurrence function of
• - amplitude
• - rise time
• - noise

Time Walk in LE discriminator due to -
amplitude and rise time variations -
charge sensitivity
Time jitter in LE discriminator due to
- noise on the Input Signal -
pulse high variation
48
going from LE to CFD
49
Constant Fraction Timing

Ideal comparator
--

• Implementing an active threshold, namely
• the threshold is derived from the signal passing
• it through an attenuator Vt f Vs (f lt 1)
• The signal applied to the comparator input ? the
• transition occurs after the threshold signal
  reached
• Its maximum value VT f V0

tr
50
The circuit compensates for amplitudes and rise
time if pulses have a sufficiently large range
that extrapolates to the same origin
delayed input signal
attenuated input signal
Timing occurrence at the output
51
(No Transcript)
52
(No Transcript)
53

• The condition for the delay must
• be met for the minimum rise time

and in this mode the fractional threshold
VT / V0 varies with rise time
For all amplitudes and rise times the
compensation range the comparator fires at the
time ? time stamp
t
54
Pulse Shaper, comprising the - delay
(td) - attenuator (f) -
subtraction followed by a zero cross trigger
Another view of CFD, namely the CFD can be
analyzed as a special pulse shaper
The new timing jitter depends on - the
slope at the zero- crossing (depends on
choice of f and td) - the noise at the output
of the shaper (which increases the noise
bandwidth)
Op. Amp /- 1.
55
Signal formation in a CFD ARC-CFD
T.J. Paulus - Timing Electronics and Fast Timing
Methods with scintillation detectors IEEE
Trans. NS NS-32, (1985), 1242
Ortec AN 42 Principles and Applications
of Timing Spectroscopy
56





Constant-Fraction Discrimination for TFC
Bipolar Signals

vs. Constant-Fraction Discrimination
for or ARC Timing
T.J. Paulus, Timing Electronics and Fast Timing
Methods with Scintillation Detectors,
EGG Ortec, IEEE Trans. on NS, Vol.NS-32 No-3
(1985), 1242
57





r.m.s. value of the input noise CFD attenuation
factor mean-square value of the input
noise autocorrelation function of the input
noise CFD shaping delay -for uncorrelated noise
/ signals
Timing uncertainty due to noise-
induced jitter for TFF timing
signal noise
58






For ARC timing with linear input signal the
slope of the CFD signal at zero crossing is
Combining former equations, we get the
expressions for noise-induced jitter with linear
input signals
- for TCF timing
- for ARC timing
59
CFD a realistic approximation In the case of
MCP real signals i.e. non-linear rise times
60
The development of MSCD method for picosecond
lifetime measurement. J.-M. Regis-
PhD work 2010

• the prompt curve determination
• ? energy dependent walk in CFD
• the prompt curve has to be
• calibrated for each branch but the
• timing asymmetry in the branch timing
• characteristic is canceled when a new
• physical quantity is defined, namely
• the Centroid Difference

Mirror Sensitive Centroid Method
61
(a) CFPHT

(b) ARC-Timing
M.A. El-Wahab et al, CFT with scintillation
detectors, IEEE Trans. On NS, Vol.36,
No.1,(1989) 401-406
62

• Variation of resolving time (W1/2) with
• the attenuation factor for three cases of
• CFD timing
• (1) - CFPHT, equivalent to LE timing
• (2) - ARC timing where ts tr
• td and tm from numerical solution
• (3) ARC timing where F(tm) A
• F(tm-td)A² and ts calculated
• from Eq.10
• Variation of resolving time (W1/2) with
• attenuation factor for different delay times

CFD ARC- CFD
(a) ARC- CFD (b)
Attenuation factor A
LE CFD
Attenuation factor A
M.A. El-Wahab et al, CFT with scintillation
detectors, IEEE Trans. On NS, Vol.36,
No.1,(1989) 401-406
63
LE Walk CFD
64
Ortec 583B , Ortec 584, Ortec 935, ESN-4000







Different design for walk adjustment, i.e.
monitor-inspect out.
65






Anode
Anode
Dynode
Dynode
Ortec 113 Preamplifier
Ortec 113 Preamplifier
T1
T2

• Anode vs Dynode
• as timing signal is still
• an open dispute ? ?!

Ortec 572 Filter Amplifier
Ortec 572 Filter Amplifier
E2
E1
Typical Fast / Slow Timing system for
gamma-gamma coincidence measurements with
scintillators and photomultiplier tubes
66

• Timing MCA
• Classical approach TPHC (TAC) ADC
• TDC
• - direct Time-Digitizer (TDC)
• - Time - Expansion (Time-to-Charge)
• - direct Digital Interpolation TDC

67
Principle of TPHC (TAC)

• ADC (13-14 bit)
• Dead Time 1-4 µs
• CC interface

68
Principle of Direct Time Digitizer
69
(No Transcript)
70
Time expanding (multihit) TDC
71
Principle of Interpolating in Direct Time
Digitizers


72
(No Transcript)
73
Measurement of 5.0 mm
5.1 mm
5.5 mm
Waveform diagram vernier like scale- TDC
74
An interpolating Time-to-Digital converter
implemented on an FPGA structure