Analog and Digital Electronics for Detectors

Helmuth Spieler
Physics Division, Lawrence Berkeley National Laboratory∗
Berkeley, California 94720, U.S.A.

1 Introduction

Electronics are a key component of all modern detector systems. Although

experiments and their associated electronics can take very different forms, the
same basic principles of the electronic readout and optimization of signal-to-
noise ratio apply to all. This chapter provides a summary of front-end electron-
ics components and discusses signal processing with an emphasis on electronic
noise. Because of space limitations, this can only be a brief overview. A more
detailed discussion of electronics with emphasis on semiconductor detectors is
given elsewhere[1]. Tutorials on detectors, signal processing and electronics are
also available on the world wide web[2].
The purpose of front-end electronics and signal processing systems is to

1. Acquire an electrical signal from the sensor. Typically this is a short

current pulse.

2. Tailor the time response of the system to optimize

(a) the minimum detectable signal (detect hit/no hit),

(b) energy measurement,
(c) event rate,
(d) time of arrival (timing measurement),
(e) insensitivity to sensor pulse shape,
(f) or some combination of the above.

3. Digitize the signal and store for subsequent analysis.

Position-sensitive detectors utilize the presence of a hit, amplitude measurement

or timing, so these detectors pose the same set of requirements.
Generally, these properties cannot be optimized simultaneously, so com-
promises are necessary. In addition to these primary functions of an electronic
∗
This work was supported by the Director, Office of Science, Office of High Energy and
Nuclear Physics, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231
110 Helmuth Spieler

INCIDENT SENSOR PREAMPLIFIER PULSE ANALOG TO DIGITAL

RADIATION SHAPING DIGITAL DATA BUS
CONVERSION

Figure 1: Basic detector functions: Radiation is absorbed in the sensor and converted into
an electrical signal. This low-level signal is integrated in a preamplifier, fed to a pulse shaper,
and then digitized for subsequent storage and analysis.

readout system, other considerations can be equally or even more important.

Examples are radiation resistance, low power (portable systems, large detector
arrays, satellite systems), robustness, and – last, but not least – cost.

2 Example Systems

Figure 1 illustrates the components and functions of a radiation detector

system. The sensor converts the energy deposited by a particle (or photon)
to an electrical signal. This can be achieved in a variety of ways. In direct
detection – semiconductor detectors, wire chambers, or other types of ionization
chambers – energy is deposited in an absorber and converted into charge pairs,
whose number is proportional to the absorbed energy. The signal charge can be
quite small, in semiconductor sensors about 50 aC (5 · 10−17 C) for 1 keV x-rays
and 4 fC (4 · 10−15 C) in a typical high-energy tracking detector, so the sensor
signal must be amplified. The magnitude of the sensor signal is subject to
statistical fluctuations and electronic noise further “smears” the signal. These
fluctuations will be discussed below, but at this point we note that the sensor
and preamplifier must be designed carefully to minimize electronic noise. A
critical parameter is the total capacitance in parallel with the input, i.e. the
sensor capacitance and input capacitance of the amplifier. The signal-to-noise
ratio increases with decreasing capacitance. The contribution of electronic noise
also relies critically on the next stage, the pulse shaper, which determines the
bandwidth of the system and hence the overall electronic noise contribution.
The shaper also limits the duration of the pulse, which sets the maximum signal
rate that can be accommodated. The shaper feeds an analog-to-digital converter
(ADC), which converts the magnitude of the analog signal into a bit-pattern
suitable for subsequent digital storage and processing.
A scintillation detector (Figure 2) utilizes indirect detection, where the
absorbed energy is first converted into visible light. The number of scintilla-
Analog and Digital Electronics for Detectors 111

SCINTILLATOR PHOTOCATHODE ELECTRON THRESHOLD

MULTIPLIER DISCRIMINATOR

INCIDENT LIGHT ELECTRONS ELECTRICAL

RADIATION SIGNAL

PHOTOMULTIPLIER VTH

NUMBER OF NUMBER OF CHARGE IN PULSE LOGIC PULSE

SCINTILLATION PHOTONS PHOTO-ELECTRONS PROPORTIONAL TO
PROPORTIONAL TO PROPORTIONAL TO ABSORBED ENERGY
ABSORBED ENERGY ABSORBED ENERGY

Figure 2: In a scintillation detector absorbed energy is converted into visible light. The
scintillation photons are commonly detected by a photomultiplier, which can provide sufficient
gain to directly drive a threshold discriminator.

tion photons is proportional to the absorbed energy. The scintillation light is

detected by a photomultiplier (PMT), consisting of a photocathode and an elec-
tron multiplier. Photons absorbed in the photocathode release electrons, whose
number is proportional to the number of incident scintillation photons. At this
point energy absorbed in the scintillator has been converted into an electrical
signal whose charge is proportional to energy. Increased in magnitude by the
electron multiplier, the signal at the PMT output is a current pulse. Integrated
over time this pulse contains the signal charge, which is proportional to the ab-
sorbed energy. Figure 2 shows the PMT output pulse fed directly to a threshold
discriminator, which fires when the signal exceeds a predetermined threshold,
as in a counting or timing measurement. The electron multiplier can provide
sufficient gain, so no preamplifier is necessary. This is a typical arrangement
used with fast plastic scintillators. In an energy measurement, for example us-
ing a NaI(Tl) scintillator, the signal would feed a pulse shaper and ADC, as
shown in Figure 1.
If the pulse shape does not change with signal charge, the peak amplitude
– the pulse height – is a measure of the signal charge, so this measurement
is called pulse height analysis. The pulse shaper can serve multiple functions,
which are discussed below. One is to tailor the pulse shape to the ADC. Since
the ADC requires a finite time to acquire the signal, the input pulse may not be
too short and it should have a gradually rounded peak. In scintillation detector
systems the shaper is frequently an integrator and implemented as the first stage
of the ADC, so it is invisible to the casual observer. Then the system appears
112 Helmuth Spieler

PREAMPLIFIER SHAPER ANALOG PIPELINE ADC

TOKEN IN
ANALOG SIGNAL PROCESSING
SPARSIFICATION
ANALOG SIGNAL PROCESSING CONTROL
DIGITAL
ANALOG SIGNAL PROCESSING CONTROL DATA OUT
OUTPUT
ANALOG SIGNAL PROCESSING DRIVERS TOKEN OUT
ANALOG SIGNAL PROCESSING

TEST PULSE GENERATOR, DACs, R/ W POINTERS, etc.

Figure 3: Circuit blocks in a representative readout IC. The analog processing chain is
shown at the top. Control is passed from chip to chip by token passing.

very simple, as the PMT output is plugged directly into a charge-sensing ADC.
A detector array combines the sensor and the analog signal processing
circuitry together with a readout system. The electronic circuitry is often
monolithically integrated. Figure 3 shows the circuit blocks in a representa-
tive readout integrated circuit (IC). Individual sensor electrodes connect to
parallel channels of analog signal processing circuitry. Data are stored in an
analog pipeline pending a readout command. Variable write and read pointers
are used to allow simultaneous read and write. The signal in the time slot of
interest is digitized, compared with a digital threshold, and read out. Circuitry
is included to generate test pulses that are injected into the input to simulate
a detector signal. This is a very useful feature in setting up the system and
is also a key function in chip testing prior to assembly. Analog control levels
are set by digital-to-analog converters (DACs). Multiple ICs are connected to
a common control and data output bus, as shown in Figure 4. Each IC is as-
signed a unique address, which is used in issuing control commands for setup
and in situ testing. Sequential readout is controlled by token passing. IC1 is
the master, whose readout is initiated by a command (trigger) on the control
bus. When it has finished writing data it passes the token to IC2, which in
turn passes the token to IC3. When the last chip has completed its readout
Analog and Digital Electronics for Detectors 113

STRIP DETECTOR

IC3 IC2 IC1

TOKEN
PASSING
CONTROL BUS

DATA BUS

Figure 4: Multiple ICs are ganged to read out a silicon strip detector. The right-most chip
IC1 is the master. A command on the control bus initiates the readout. When IC1 has written
all of its data it passes the token to IC2. When IC2 has finished it passes the token to IC3,
which in turn returns the token to the master IC1.

the token is returned to the master IC, which is then ready for the next cycle.
The readout bit stream begins with a header, which uniquely identifies a new
frame. Data from individual ICs are labeled with a chip identifier and channel
identifiers. Many variations on this scheme are possible. As shown, the readout
is event oriented, i.e. all hits occurring within an externally set exposure time
(e.g. time slice in the analog buffer in Figure 3) are read out together. For a
concise discussion of data acquisition systems see ref. [3].
In colliding beam experiments only a small fraction of beam crossings
yields interesting events. The time required to assess whether an event is po-
tentially interesting is typically of order microseconds, so hits from multiple
beam crossings must be stored on-chip, identified by beam crossing or time-
stamp. Upon receipt of a trigger the interesting data are digitized and read
out. This allows use of a digitizer that is slower than the collision rate. It is
also possible to read out analog signals and digitize them externally. Then the
output stream is a sequence of digital headers and analog pulses. An alternative
scheme only records the presence of hit. The output of a threshold comparator
signifies the presence of a signal and is recorded in a digital pipeline that retains
the crossing number.
Figure 5 shows a closeup of ICs mounted on a hybrid using a flexible poly-
imide substrate[4]. The wire bonds connecting the IC to the hybrid are clearly
visible. Channels on the IC are laid out on a ∼ 50 µm pitch and pitch adapters
fan out to match the 80 µm pitch of the silicon strip detector. The space be-
114 Helmuth Spieler

tween chips accommodates bypass capacitors and connections for control busses
carrying signals from chip to chip.

3 Detection Limits and Resolution

The minimum detectable signal and the precision of the amplitude mea-
surement are limited by fluctuations. The signal formed in the sensor fluctu-
ates, even for a fixed energy absorption. In addition, electronic noise introduces
baseline fluctuations, which are superimposed on the signal and alter the peak
amplitude. Figure 6 (left) shows a typical noise waveform. Both the amplitude
and time distributions are random. When superimposed on a signal, the noise
alters both the amplitude and time dependence, as shown in Figure 6 (right).
As can be seen, the noise level determines the minimum signal whose presence
can be discerned.
In an optimized system, the time scale of the fluctuations is comparable

Figure 5: Closeup of ICs mounted on a hybrid utilizing a flexible polyimide substrate. The
high-density wire bonds at the upper edges connect via pitch adapters to the 80 µm pitch of
the silicon strip detector. The ground plane is patterned as a diamond grid to reduce material.
(Photograph courtesy of A. Ciocio.)
Analog and Digital Electronics for Detectors 115

TIME TIME

Figure 6: Waveforms of random noise (left) and signal + noise (right), where the peak signal
is equal to the rms noise level (S/N = 1). The noiseless signal is shown for comparison.

to that of the signal, so the peak amplitude fluctuates randomly above and
below the average value. This is illustrated in Figure 7, which shows the same
signal viewed at four different times. The fluctuations in peak amplitude are
obvious, but the effect of noise on timing measurements can also be seen. If
the timing signal is derived from a threshold discriminator, where the output
fires when the signal crosses a fixed threshold, amplitude fluctuations in the

TIME TIME

TIME TIME

Figure 7: Signal plus noise at four different times, shown for a signal-to-noise ratio of about
20. The noiseless signal is superimposed for comparison.
116 Helmuth Spieler

leading edge translate into time shifts. If one derives the time of arrival from
a centroid analysis, the timing signal also shifts (compare the top and bottom
right figures). From this one sees that signal-to-noise ratio is important for all
measurements – sensing the presence of a signal or the measurement of energy,
timing, or position.

4 Acquiring the Sensor Signal

The sensor signal is usually a short current pulse is (t). Typical durations
vary widely, from 100 ps for thin Si sensors to tens of µs for inorganic scintilla-
tors. However, the physical quantity of interest is the deposited energy, so one
has to integrate over the current pulse
�
E ∝ Qs = is (t)dt . (1)

This integration can be performed at any stage of a linear system, so one can

1. integrate on the sensor capacitance,

2. use an integrating preamplifier (“charge-sensitive” amplifier),

3. amplify the current pulse and use an integrating ADC (“charge sensing”
ADC),

4. rapidly sample and digitize the current pulse and integrate numerically.

In high-energy physics the first three options tend to be most efficient.

4.1 Signal integration

Figure 8 illustrates signal formation in an ionization chamber connected to
an amplifier with a very high input resistance. The ionization chamber volume
could be filled with gas or a solid, as in a silicon sensor. As mobile charge carriers
move towards their respective electrodes they change the induced charge on the
sensor electrodes, which form a capacitor Cd . If the amplifier has a very small
input resistance Ri , the time constant τ = Ri (Cd + Ci ) for discharging the
sensor is small, and the amplifier will sense the signal current. However, if the
input time constant is large compared to the duration of the current pulse, the
current pulse will be integrated on the capacitance and the resulting voltage at
the amplifier input
Qs
Vi = . (2)
Cd + Ci
Analog and Digital Electronics for Detectors 117

DETECTOR AMPLIFIER
tc

VELOCITY OF v
CHARGE CARRIERS
t
Cd Vi Ci Ri RATE OF INDUCED dqs
CHARGE ON SENSOR dt
ELECTRODES
t
Qs
SIGNAL CHARGE qs

Figure 8: Charge collection and signal integration in an ionization chamber.

The magnitude of the signal is dependent on the sensor capacitance. In a system

with varying sensor capacitances, a Si tracker with varying strip lengths, for
example, or a partially depleted semiconductor sensor, where the capacitance
varies with the applied bias voltage, one would have to deal with additional
calibrations. Although this is possible, it is awkward, so it is desirable to use a
system where the charge calibration is independent of sensor parameters. This
can be achieved rather simply with a charge-sensitive amplifier.
Figure 9 shows the principle of a feedback amplifier that performs integra-
tion. It consists of an inverting amplifier with voltage gain −A and a feedback
capacitor Cf connected from the output to the input. To simplify the calcula-
tion, let the amplifier have an infinite input impedance, so no current flows into
the amplifier input. If an input signal produces a voltage vi at the amplifier
input, the voltage at the amplifier output is −Avi . Thus, the voltage difference
across the feedback capacitor vf = (A + 1)vi and the charge deposited on Cf is

Cf

Qi
DETECTOR !A

Cd vi vo

Figure 9: Principle of a charge-sensitive amplifier

118 Helmuth Spieler

TEST
INPUT !V

CT

Q-AMP

Cd Ci

DYNAMIC INPUT
CAPACITANCE

Figure 10: Adding a test input to a charge-sensitive amplifier provides a simple means of
absolute charge calibration.

Qf = Cf vf = Cf (A + 1)vi . Since no current can flow into the amplifier, all of

the signal current must charge up the feedback capacitance, so Qf = Qi . The
amplifier input appears as a “dynamic” input capacitance
Qi
Ci = = Cf (A + 1) . (3)
vi
The voltage output per unit input charge
dvo Avi A A 1 1
AQ = = = = · ≈ (A � 1) , (4)
dQi Ci vi Ci A + 1 Cf Cf
so the charge gain is determined by a well-controlled component, the feedback
capacitor.
The signal charge Qs will be distributed between the sensor capacitance
Cd and the dynamic input capacitance Ci . The ratio of measured charge to
signal charge
Qi Qi Ci 1
= = = , (5)
Qs Qd + Qi Cd + Ci Cd
1+
Ci
so the dynamic input capacitance must be large compared to the sensor capac-
itance.
Another very useful byproduct of the integrating amplifier is the ease of
charge calibration. By adding a test capacitor as shown in Figure 10, a voltage
step injects a well-defined charge into the input node. If the dynamic input
capacitance Ci is much larger than the test capacitance CT , the voltage step at
the test input will be applied nearly completely across the test capacitance CT ,
thus injecting a charge CT ∆V into the input.
Analog and Digital Electronics for Detectors 119

DETECTOR AMPLIFIER

ii

is Cd vi Ri

Figure 11: Charge integration in a realistic charge-sensitive amplifier. First, charge is

integrated on the sensor capacitance and subsequently transferred to the charge-sensitive
loop, as it becomes active.

4.2 Realistic charge-sensitive amplifiers

The preceding discussion assumed that the amplifiers are infinitely fast,
so they respond instantaneously to the applied signal. In reality this is not the
case; charge-sensitive amplifiers often respond much more slowly than the time
duration of the current pulse from the sensor. However, as shown in Figure 11,
this does not obviate the basic principle. Initially, signal charge is integrated on
the sensor capacitance, as indicated by the left hand current loop. Subsequently,
as the amplifier responds the signal charge is transferred to the amplifier.
Nevertheless, the time response of the amplifier does affect the measured
pulse shape. First, consider a simple amplifier as shown in Figure 12.
The gain element shown is a bipolar transistor, but it could also be a field
effect transistor (JFET or MOSFET) or even a vacuum tube. The transistor’s

V+

RL

io Co
vo
vi

Figure 12: A simple amplifier demonstrating the general features of any single-stage gain
stage, whether it uses a bipolar transistor (shown) or an FET.
120 Helmuth Spieler

output current changes as the input voltage is varied. Thus, the voltage gain
dvo dio
Av = = · ZL ≡ gm ZL . (6)
dvi dvi
The parameter gm is the transconductance, a key parameter that determines
gain, bandwidth, and noise of transistors. The load impedance ZL is the parallel
combination of the load resistance RL and the output capacitance Co . This
capacitance is unavoidable; every gain device has an output capacitance, the
following stage has an input capacitance, and in addition the connections and
additional components introduce stray capacitance. The load impedance is
given by
1 1
= + iωCo , (7)
ZL RL
where the imaginary i indicates the phase shift associated with the capacitance.
The voltage gain
� �−1
1
Av = gm + iωCo . (8)
RL
At low frequencies where the second term is negligible, the gain is constant
Av = gm RL . However, at high frequencies the second term dominates and
the gain falls off linearly with frequency with a 90◦ phase shift, as illustrated
in Figure 13. The cutoff (corner) frequency, where the asymptotic low and
high frequency responses intersect, is determined by the output time constant
τ = RL Co , so the cutoff frequency
1 1
fu = = . (9)
2πτ 2πRL Co
In the regime where the gain drops linearly with frequency the product of gain
and frequency is constant, so the amplifier can be characterized by its gain–
bandwidth product, which is equal to the frequency where the gain is one, the
unity gain frequency ω0 = gm /Co .
The frequency response translates into a time response. If a voltage step
is applied to the input of the amplifier, the output does not respond instanta-
neously, as the output capacitance must first charge up. This is shown in the
second panel of Figure 13.
In practice, amplifiers utilize multiple stages, all of which contribute to
the frequency response. However, for use as a feedback amplifier, only one
time constant should dominate, so the other stages must have much higher
cutoff frequencies. Then the overall amplifier response is as shown in Figure 13,
except that at high frequencies additional corner frequencies appear.
Analog and Digital Electronics for Detectors 121

FREQUENCY DOMAIN TIME DOMAIN

log Av g m RL INPUT OUTPUT

gm
A v0 #i
!Co

V0

Av = 1
1 !0 log ! V = V0 (1 # exp(#t / $) )
RLCo
$ = RLCo

UPPER CUTOFF FREQUENCY 2" fu

Figure 13: The time constants of an amplifier affect both the frequency and the time
response. Both are fully equivalent representations.

We can now use the frequency response to calculate the input impedance
and time response of a charge-sensitive amplifier. Applying the same reasoning
as above, the input impedance of an inverting amplifier as shown in Figure 9,
but with a generalized feedback impedance Zf , is

Zf Zf
Zi = ≈ (A � 1) . (10)
A+1 A
At low frequencies the gain is constant and has a constant 180◦ phase shift,
so the input impedance is of the same nature as the feedback impedance, but
reduced by 1/A. At high frequencies well beyond the amplifier’s cutoff frequency
fu , the gain drops linearly with frequency with an additional 90◦ phase shift,
so the gain
ω0
A = −i . (11)
ω
In a charge-sensitive amplifier the feedback impedance
1
Zf = −i , (12)
ωCf

so the input impedance

i 1 1
Zi = − · ω = ≡ Ri . (13)
ωCf −i 0 ω0 Cf
ω
122 Helmuth Spieler

The imaginary component vanishes, so the input impedance is real. In other

words, it appears as a resistance Ri . Thus, at low frequencies f � fu the input
of a charge-sensitive amplifier appears capacitive, whereas at high frequencies
f � fu it appears resistive.
Suitable amplifiers invariably have corner frequencies well below the fre-
quencies of interest for radiation detectors, so the input impedance is resistive.
This allows a simple calculation of the time response. The sensor capacitance
is discharged by the resistive input impedance of the fedback amplifier with the
time constant
1
τi = Ri Cd = · Cd . (14)
ω0 Cf

From this we see that the rise time of the charge-sensitive amplifier increases
with sensor capacitance. As noted above, the amplifier response can be slower
than the duration of the current pulse from the sensor, but it should be much
faster than the peaking time of the subsequent pulse shaper. The feedback ca-
pacitance should be much smaller than the sensor capacitance. If Cf = Cd /100,
the amplifier’s gain–bandwidth product must be 100/τi , so for a rise time con-
stant of 10 ns the gain–bandwidth product must be 1010 radians = 1.6 GHz. The
same result can be obtained using conventional operational amplifier feedback
theory.
The mechanism of reducing the input impedance through shunt feedback
leads to the concept of the “virtual ground”. If the gain is infinite, the input
impedance is zero. Although very high gains (of order 105 to 106 ) are achievable
in the kHz range, at the frequencies relevant for detector signals the gain is
much smaller. The input impedance of typical charge-sensitive amplifiers in
strip detector systems is of order kΩ. Fast amplifiers designed to optimize
power dissipation achieve input impedances of 100 to 500 Ω [5]. None of these
qualify as a “virtual ground”, so this concept should be applied with caution.
Apart from determining the signal rise time, the input impedance is critical
in position-sensitive detectors. Figure 14 illustrates a silicon-strip sensor read
out by a bank of amplifiers. Each strip electrode has a capacitance Cb to
the backplane and a fringing capacitance Css to the neighboring strips. If
the amplifier has an infinite input impedance, charge induced on one strip will
capacitively couple to the neighbors and the signal will be distributed over many
strips (determined by Css /Cb ). If, on the other hand, the input impedance of
the amplifier is low compared to the interstrip impedance, practically all of the
charge will flow into the amplifier, as current seeks the path of least impedance,
and the neighbors will show only a small signal.
Analog and Digital Electronics for Detectors 123

Figure 14: To preserve the position resolution of strip detectors the readout amplifiers
must have a low input impedance to prevent spreading of signal charge to the neighboring
electrodes.

5 Signal Processing

As noted in the introduction, one of the purposes of signal processing

is to improve the signal-to-noise ratio by tailoring the spectral distributions
of the signal and the electronic noise. However, for many detectors electronic
noise does not determine the resolution. For example, in a NaI(Tl) scintillation
detector measuring 511 keV gamma rays, say in a positron-emission tomography
system, 25 000 scintillation photons are produced. Because of reflective losses,
about 15 000 reach the photocathode. This translates to about 3000 electrons
reaching the first dynode. The gain of the electron multiplier will yield about
3 · 109 electrons at the anode. The statistical spread of the signal is determined
by the smallest number of electrons in the chain,√ i.e. the 3000 electrons reaching
the first dynode, so the resolution ∆E/E = 1/ 3000 = 2%, which at the anode
corresponds to about 5 · 104 electrons. This is much larger than the electronic
noise in any reasonably designed system. This situation is illustrated in Figure
15 (top). In this case, signal acquisition and count rate capability may be the
prime objectives of the pulse processing system. The bottom illustration in
Figure 15 shows the situation for high resolution sensors with small signals, for
example semiconductor detectors, photodiodes or ionization chambers. In this
124 Helmuth Spieler

SIGNAL ! BASELINE NOISE SIGNAL + NOISE

BASELINE BASELINE BASELINE

SIGNAL ! BASELINE NOISE SIGNAL + NOISE

BASELINE BASELINE BASELINE

Figure 15: Signal and baseline fluctuations add in quadrature. For large signal variance
(top) as in scintillation detectors or proportional chambers, the baseline noise is usually neg-
ligible, whereas for small signal variance as in semiconductor detectors or liquid Ar ionization
chambers, baseline noise is critical.

case, low noise is critical. Baseline fluctuations can have many origins, external
interference, artifacts due to imperfect electronics, etc., but the fundamental
limit is electronic noise.

6 Electronic Noise

Consider a current flowing through a sample bounded by two electrodes,

i.e. n electrons moving with velocity v. The induced current depends on the
spacing l between the electrodes (following “Ramo’s theorem”[6], [1]), so
nev
i= . (15)
l
The fluctuation of this current is given by the total differential
� ne �2 � ev �2
2
�di� = �dv� + �dn� , (16)
l l
where the two terms add in quadrature, as they are statistically uncorrelated.
From this one sees that two mechanisms contribute to the total noise, velocity
and number fluctuations.
Analog and Digital Electronics for Detectors 125

1
NORMALIZED COUNT RATE
Qn
0.78

FWHM= 2.35 Q n
0.5

Q s /Q n

Figure 16: Repetitive measurements of the signal charge yield a Gaussian distribution whose
standard deviation equals the rms noise level Qn . Often the width is expressed as the full
width at half maximum (FWHM), which is 2.35 times the standard deviation.

Velocity fluctuations originate from thermal motion. Superimposed on

the average drift velocity are random velocity fluctuations due to thermal ex-
citations. This “thermal noise” is described by the long wavelength limit of
Planck’s black body spectrum where the spectral density, i.e. the power per
unit bandwidth, is constant (“white” noise).
Number fluctuations occur in many circumstances. One source is car-
rier flow that is limited by emission over a potential barrier. Examples are
thermionic emission or current flow in a semiconductor diode. The probability
of a carrier crossing the barrier is independent of any other carrier being emit-
ted, so the individual emissions are random and not correlated. This is called
“shot noise”, which also has a “white” spectrum. Another source of number
fluctuations is carrier trapping. Imperfections in a crystal lattice or impurities
in gases can trap charge carriers and release them after a characteristic lifetime.
This leads to a frequency-dependent spectrum dPn /df = 1/f α , where α is typ-
ically in the range of 0.5 – 2. Simple derivations of the spectral noise densities
are given in [1].
The amplitude distribution of the overall noise is Gaussian, so superim-
posing a constant amplitude signal on a noisy baseline will yield a Gaussian
amplitude distribution whose width equals the noise level (Figure 16). Inject-
ing a pulser signal and measuring the width of the amplitude distribution yields
the noise level.
126 Helmuth Spieler

6.1 Thermal (Johnson) noise

The most common example of noise due to velocity fluctuations is the
noise of resistors. The spectral noise power density vs. frequency
dPn
= 4kT , (17)
df
where k is the Boltzmann constant and T the absolute temperature. Since the
power in a resistance R can be expressed through either voltage or current,

V2
P = = I 2R , (18)
R
the spectral voltage and current noise densities

dVn2 dIn2 4kT

≡ e2n = 4kT R and ≡ i2n = . (19)
df df R
The total noise is obtained by integrating over the relevant frequency range
of the system, the bandwidth, so the total noise voltage at the output of an
amplifier with a frequency-dependent gain A(f ) is
�∞
2
von = e2n A2 (f )df . (20)
0

Since the spectral noise components are non-correlated (each black body ex-
citation mode is independent), one must integrate over the noise power, i.e.
the voltage squared. The total noise increases with bandwidth. Since small
bandwidth corresponds to large rise-times, increasing the speed of a pulse mea-
surement system will increase the noise.

6.2 Shot noise

The spectral density of shot noise is proportional to the average current
I:
i2n = 2eI , (21)
where e is the electronic charge. Note that the criterion for shot noise is that
carriers are injected independently of one another, as in thermionic emission or
semiconductor diodes. Current flowing through an ohmic conductor does not
carry shot noise, since the fields set up by any local fluctuation in charge density
can easily draw in additional carriers to equalize the disturbance.
Analog and Digital Electronics for Detectors 127

7 Signal-to-noise Ratio vs. Sensor Capacitance

The basic noise sources manifest themselves as either voltage or current

fluctuations. However, the desired signal is a charge, so to allow a comparison
we must express the signal as a voltage or current. This was illustrated for an
ionization chamber in Figure 8. As was noted, when the input time constant
Ri (Cd + Ci ) is large compared to the duration of the sensor current pulse, the
signal charge is integrated on the input capacitance, yielding the signal voltage
Vs = Qs /(Cd + Ci ). Assume that the amplifier has an input noise voltage Vn .
Then the signal-to-noise ratio
Vs Qs
= . (22)
Vn Vn (Cd + Ci )

This is a very important result – the signal-to-noise ratio for a given signal
charge is inversely proportional to the total capacitance at the input node.
Note that zero input capacitance does not yield an infinite signal-to-noise ratio.
As shown in ref. [1], this relationship only holds when the input time constant
is greater than about ten times the sensor current pulse width. The dependence
of signal-to-noise ratio on capacitance is a general feature that is independent of
amplifier type. Since feedback cannot improve signal-to-noise ratio, eqn 22 holds
for charge-sensitive amplifiers, although in that configuration the charge signal
is constant, but the noise increases with total input capacitance (see[1]). In the
noise analysis the feedback capacitance adds to the total input capacitance (the
passive capacitance, not the dynamic input capacitance), so Cf should be kept
small.

8 Pulse Shaping

Pulse shaping has two conflicting objectives. The first is to limit the
bandwidth to match the measurement time. Too large a bandwidth will increase
the noise without increasing the signal. Typically, the pulse shaper transforms
a narrow sensor pulse into a broader pulse with a gradually rounded maximum
at the peaking time. This is illustrated in Figure 17. The signal amplitude is
measured at the peaking time TP .
The second objective is to constrain the pulse width so that successive
signal pulses can be measured without overlap (pileup), as illustrated in Figure
18. Reducing the pulse duration increases the allowable signal rate, but at the
expense of electronic noise.
In designing the shaper it is necessary to balance these conflicting goals.
Usually, many different considerations lead to a “non-textbook” compromise;
128 Helmuth Spieler

SENSOR PULSE SHAPER OUTPUT

TP

Figure 17: In energy measurements a pulse processor typically transforms a short sensor
current pulse to a broader pulse with a peaking time TP .

“optimum shaping” depends on the application.

A simple shaper is shown in Figure 19. A high-pass filter sets the duration
of the pulse by introducing a decay time constant τd . Next a low-pass filter with
a time constant τi increases the rise time to limit the noise bandwidth. The
high-pass is often referred to as a “differentiator”, since for short pulses it forms
the derivative. Correspondingly, the low-pass is called an “integrator”. Since
the high-pass filter is implemented with a CR section and the low-pass with
an RC, this shaper is referred to as a CR-RC shaper. Although pulse shapers
are often more sophisticated and complicated, the CR-RC shaper contains the
essential features of all pulse shapers, a lower frequency bound and an upper
frequency bound.
After peaking the output of a simple CR-RC shaper returns to baseline
rather slowly. The pulse can be made more symmetrical, allowing higher signal
rates for the same peaking time. Very sophisticated circuits have been developed
towards this goal, but a conceptually simple way is to use multiple integrators, as
AMPLITUDE

AMPLITUDE

TIME TIME
Figure 18: Amplitude pileup occurs when two pulses overlap (left). Reducing the shaping
time allows the first pulse to return to the baseline before the second pulse arrives.
Analog and Digital Electronics for Detectors 129

SENSOR CURRENT INTEGRATOR HIGH-PASS FILTER LOW-PASS FILTER

“DIFFERENTIATOR” “INTEGRATOR”
!d !i
is
"A

e-t /!d

Figure 19: Components of a pulse shaping system. The signal current from the sensor is
integrated to form a step impulse with a long decay. A subsequent high-pass filter (“differ-
entiator”) limits the pulse width and the low-pass filter (“integrator”) increases the rise-time
to form a pulse with a smooth cusp.

1.0
SHAPER OUTPUT

n= 1

0.5 2

n= 8

0.0
0 1 2 3 4 5
TIME
Figure 20: Pulse shape vs. number of integrators in a CR-nRC shaper. The time constants
are scaled with the number of integrators to maintain the peaking time.

illustrated in Figure 20. The integration and differentiation time constants are
scaled to maintain the peaking time. Note that the peaking time is a key design
parameter, as it dominates the noise bandwidth and must also accommodate
the sensor response time.
Another type of shaper is the correlated double sampler, illustrated in
Figure 21. This type of shaper is widely used in monolithically integrated
circuits, as many CMOS processes (see Section 11.1) provide only capacitors
130 Helmuth Spieler

and switches, but no resistors. Input signals are superimposed on a slowly

fluctuating baseline. To remove the baseline fluctuations the baseline is sampled
prior to the signal. Next, the signal plus baseline is sampled and the previous
baseline sample subtracted to obtain the signal. The prefilter is critical to limit
the noise bandwidth of the system. Filtering after the sampler is useless, as noise
fluctuations on time scales shorter than the sample time will not be removed.
Here the sequence of filtering is critical, unlike a time-invariant linear filter (e.g.
a CR-RC filter as in Figure 19) where the sequence of filter functions can be
interchanged.
This is an example of a time-variant filter. The CR-nRC filter described
above acts continuously on the signal, whereas the correlated double sample
changes filter parameters vs. time.

9 Noise Analysis of a Detector and Front-end Amplifier

To determine how the pulse shaper affects the signal-to-noise ratio consider
the detector front-end in Figure 22. The detector is represented by the capac-
itance Cd , a relevant model for many radiation sensors. Sensor bias voltage is
applied through the resistor Rb . The bypass capacitor Cb shunts any external
interference coming through the bias supply line to ground. For high-frequency
signals this capacitor appears as a low impedance, so for sensor signals the “far
end” of the bias resistor is connected to ground. The coupling capacitor Cc
blocks the sensor bias voltage from the amplifier input, which is why a capaci-

SIGNAL

S1

S1
V1
V1 vn
Vo
S2
SIGNALS V2
S2
NOISE vs + vn
V2

!v= vs + vn vs
Vo
vn

Figure 21: Principle of a shaper using correlated double sampling.

Analog and Digital Electronics for Detectors 131

DETECTOR BIAS PREAMPLIFIER + DETECTOR BIAS SERIES PREAMPLIFIER +

PULSE SHAPER RESISTOR RESISTOR PULSE SHAPER
Cb
Rs ens
ena
Rb ind
OUTPUT Cd Rb inb ina
Cc Rs
Cd

Figure 22: A detector front-end circuit and its equivalent circuit for noise calculations.

tor serving this role is also called a “blocking capacitor”. The series resistor Rs
represents any resistance present in the connection from the sensor to the am-
plifier input. This includes the resistance of the sensor electrodes, the resistance
of the connecting wires or traces, any resistance used to protect the amplifier
against large voltage transients (“input protection”), and parasitic resistances
in the input transistor.
The following implicitly includes a constraint on the bias resistance, whose
role is often misunderstood. It is often thought that the signal current generated
in the sensor flows through Rb and the resulting voltage drop is measured. If
the time constant Rb Cd is small compared to the peaking time of the shaper TP ,
the sensor will have discharged through Rb and much of the signal will be lost.
Thus, we have the condition Rb Cd � TP , or Rb � TP /Cd . The bias resistor
must be sufficiently large to block the flow of signal charge, so that all of the
signal is available for the amplifier.
To analyze this circuit we’ll assume a voltage amplifier, so all noise con-
tributions will be calculated as a noise voltage appearing at the amplifier input.
Steps in the analysis are 1. determine the frequency distribution of all noise
voltages presented to the amplifier input from all individual noise sources, 2.
integrate over the frequency response of the shaper (for simplicity a CR-RC
shaper) and determine the total noise voltage at the shaper output, and 3. de-
termine the output signal for a known input signal charge. The equivalent noise
charge (ENC) is the signal charge for which S/N = 1.
The equivalent circuit for the noise analysis (second panel of Figure 22)
includes both current and voltage noise sources. The “shot noise” ind of the
sensor leakage current is represented by a current noise generator in parallel
with the sensor capacitance. As noted above, resistors can be modeled either
as a voltage or current generator. Generally, resistors shunting the input act as
noise current sources and resistors in series with the input act as noise voltage
132 Helmuth Spieler

sources (which is why some in the detector community refer to current and
voltage noise as “parallel” and “series” noise). Since the bias resistor effectively
shunts the input, as the capacitor Cb passes current fluctuations to ground, it
acts as a current generator inb and its noise current has the same effect as the
shot noise current from the detector. The shunt resistor can also be modeled
as a noise voltage source, yielding the result that it acts as a current source.
Choosing the appropriate model merely simplifies the calculation. Any other
shunt resistances can be incorporated in the same way. Conversely, the series
resistor Rs acts as a voltage generator. The electronic noise of the amplifier
is described fully by a combination of voltage and current sources at its input,
shown as ena and ina .
Thus, the noise sources are

sensor bias current : i2nd = 2eId

4kT
shunt resistance : i2nb =
Rb
series resistance : e2ns = 4kT Rs
amplifier : ena , ina ,

where e is the electronic charge, Id the sensor bias current, k the Boltzmann
constant and T the √ temperature.√Typical amplifier √noise parameters ena and
ina are of order nV/ Hz and fA/ Hz (FETs) – pA/ Hz (bipolar transistors).
Amplifiers tend to exhibit a “white” noise spectrum at high frequencies (greater
than order kHz), but at low frequencies show excess noise components with the
spectral density
Af
e2nf = , (23)
f
where the noise coefficient Af is device specific and of order 10−10 – 10−12 V2 .
The noise voltage generators are in series and simply add in quadrature.
White noise distributions remain white. However, a portion of the noise currents
flows through the detector capacitance, resulting in a frequency-dependent noise
voltage in /(ωCd ), so the originally white spectrum of the sensor shot noise and
the bias resistor now acquires a 1/f dependence. The frequency distribution of
all noise sources is further altered by the combined frequency response of the
amplifier chain A(f ). Integrating over the cumulative noise spectrum at the
amplifier output and comparing to the output voltage for a known input signal
yields the signal-to-noise ratio. In this example the shaper is a simple CR-RC
shaper, where for a given differentiation time constant the noise is minimized
when the differentiation and integration time constants are equal τi = τd ≡ τ .
Then the output pulse assumes its maximum amplitude at the time TP = τ .
Analog and Digital Electronics for Detectors 133

Although the basic noise sources are currents or voltages, since radiation
detectors are typically used to measure charge, the system’s noise level is con-
veniently expressed as an equivalent noise charge Qn . As noted previously, this
is equal to the detector signal that yields a signal-to-noise ratio of one. The
equivalent noise charge is commonly expressed in Coulombs, the corresponding
number of electrons, or the equivalent deposited energy (eV). For the above
circuit the equivalent noise charge
� 2 � �� � �
e 4kT � � C2
Q2n = 2eId + + i2na · τ + 4kT Rs + e2na · d + 4Af Cd2 .
8 Rb τ
(24)
The prefactor e2 /8 = exp(2)/8 = 0.924 normalizes the noise to the signal gain.
The first term combines all noise current sources and increases with shaping
time. The second term combines all noise voltage sources and decreases with
shaping time, but increases with sensor capacitance. The third term is the
contribution of amplifier 1/f noise and, as a voltage source, also increases with
sensor capacitance. The 1/f term is independent of shaping time, since for
a 1/f spectrum the total noise depends on the ratio of upper to lower cutoff
frequency, which depends only on shaper topology, but not on the shaping time.
Just as filter response can be described either in the frequency or time
domain, so can the noise performance. Detailed explanations are given in papers
by Goulding and Radeka[7] [8] [9] [10]. The key is Parseval’s theorem, which
relates the amplitude response A(f ) to the time response F (t).

�∞ �∞
2 2
|A(f )| df = [F (t)] dt . (25)
0 −∞

The left hand side is essentially integration over the noise bandwidth. The
output noise power scales linearly with the duration of the pulse, so the noise
contribution of the shaper can be split into a factor that is determined by the
shape of the response and a time factor that sets the shaping time. This leads
to a general formulation of the equivalent noise charge

C2
Q2n = i2n Fi TS + e2n Fv + Fvf Af C 2 , (26)
TS

where Fi , Fv , and Fvf depend on the shape of the pulse determined by the
shaper and TS is a characteristic time, for example the peaking time of a
CR-nRC shaped pulse or the prefilter time constant in a correlated double
sampler[1]. As before, C is the total parallel capacitance at the input. The
134 Helmuth Spieler

shape factors Fi , Fv are easily calculated:

�∞ �∞ � �2
1 2 TS dW (t)
Fi = [W (t)] dt , Fv = dt . (27)
2TS 2 dt
−∞ −∞

For time-invariant pulse shaping W (t) is simply the system’s impulse response
(the output signal seen on an oscilloscope) with the peak output signal normal-
ized to unity. For a time-variant shaper the same equations apply, but W (t) is
determined differently. See refs. [7], [8], [9], and [10] for more details.
A shaper formed by a single CR differentiator and RC integrator with
equal time constants has Fi = Fv = 0.9 and Fvf = 4, independent of the
shaping time constant, so for the circuit in Figure 19 eqn 24 becomes
� �
4kT � � C2
Q2n = 2qe Id + + i2na Fi TS + 4kT Rs + e2na Fv + Fvf Af C 2 . (28)
Rb TS

Pulse shapers can be designed to reduce the effect of current noise, e.g. mitigate
radiation damage. Increasing pulse symmetry tends to decrease Fi and increase
Fv , e.g. to Fi = 0.45 and Fv = 1.0 for a shaper with one CR differentiator and
four cascaded RC integrators.
Figure 23 shows how equivalent noise charge is affected by shaping time.
At short shaping times the voltage noise dominates, whereas at long shaping
times the current noise takes over. Minimum noise obtains where the current
and voltage contributions are equal. The noise minimum is flattened by the
presence of 1/f noise. Also shown is that increasing the detector capacitance
will increase the voltage noise contribution and shift the noise minimum to
longer shaping times, albeit with an increase in minimum noise.
For quick estimates one can use the following equation, which assumes an
FET amplifier (negligible ina ) and a simple CR-RC shaper with peaking time
τ . The noise is expressed in units of the electronic charge e and C is the total
parallel capacitance at the input, including Cd , all stray capacitances, and the
amplifier’s input capacitance.

� � � 2 � � �
e2 5 e kΩ τ e2 ns C2
Q2n = 12 Id τ + 6 · 10 + 3.6 · 10 4
e2n
nA · ns ns Rb 2 2
(pF) (nV) /Hz τ
(29)
The noise charge is improved by reducing the detector capacitance and
leakage current, judiciously selecting all resistances in the input circuit, and
Analog and Digital Electronics for Detectors 135

4
10
EQUIVALENT NOISE CHARGE (e)

3
TOTAL
10
TOTAL
1/f NOISE

CURRENT VOLTAGE
NOISE NOISE

102
0.01 0.1 1 10 100
SHAPING TIME (µs)

Figure 23: Equivalent noise charge vs. shaping time. At small shaping times (large band-
width) the equivalent noise charge is dominated by voltage noise, whereas at long shaping
times (large integration times) the current noise contributions dominate. The total noise as-
sumes a minimum where the current and voltage contributions are equal. The “1/f ” noise
contribution is independent of shaping time and flattens the noise minimum. Changing the
voltage or current noise contribution shifts the noise minimum. Increased voltage noise is
shown as an example.

choosing the optimum shaping time constant. The noise parameters of a well-
designed amplifier depend primarily on the input device. Fast, high-gain tran-
sistors are generally best.
In field effect transistors, both junction field effect transistors (JFETs) or
metal oxide semiconductor field effect transistors (MOSFETs), the noise cur-
rent contribution is very small, so reducing the detector leakage current and
increasing the bias resistance will allow long shaping times with correspond-
ingly lower noise. The equivalent input noise voltage e2n ≈ 4kT /gm , where gm
is the transconductance, which increases with operating current. For a given
current, the transconductance increases when the channel length is reduced,
so reductions in feature size with new process technologies are beneficial. At
a given channel length minimum noise obtains when a device is operated at
maximum transconductance. If lower noise is required, the width of the device
can be increased (equivalent to connecting multiple devices in parallel). This
increases the transconductance (and required current) with a corresponding de-
136 Helmuth Spieler

crease in noise voltage, but also increases the input capacitance. At some point
the reduction in noise voltage is outweighed by the increase in total input ca-
pacitance. The optimum obtains when the FET’s input capacitance equals the
external capacitance (sensor + stray capacitance). Note that this capacitive
matching criterion only applies when the input current noise contribution of
the amplifying device is negligible.
Capacitive matching comes at the expense of power dissipation. Since
the minimum is shallow, one can operate at significantly lower currents with
just a minor increase in noise. In large detector arrays power dissipation is
critical, so FETs are hardly ever operated at their minimum noise. Instead,
one seeks an acceptable compromise between noise and power dissipation (see
[1] for a detailed discussion). Similarly, the choice of input devices is frequently
driven by available fabrication processes. High-density integrated circuits tend
to include only MOSFETs, so this determines the input device, even where a
bipolar transistor would provide better performance.
In bipolar transistors the shot noise associated with the base current IB is
significant, i2nB = 2eIB . Since IB = IC /βDC , where IC is the collector current
and βDC the direct current gain, this contribution increases with device current.
On the other hand, the equivalent input noise voltage

2(kT )2
e2n = (30)
eIC
decreases with collector current, so the noise assumes a minimum at a specific
collector current
�
C � kT � Fv 1
Qn,min = 4kT √
2
Fi Fv at IC = C βDC . (31)
βDC e Fi TS

For a CR-RC shaper and βDC = 100,

� � � �
e √ µA · ns C
Qn,min ≈ 250 √ · C at IC = 260 · . (32)
pF pF TS

The minimum obtainable noise is independent of shaping time (unlike FETs),

but only at the optimum collector current IC , which does depend on shaping
time.
In bipolar transistors the input capacitance is usually
√ much smaller than
the sensor capacitance (of order 1 pF for en ≈ 1 nV/ Hz) and substantially
smaller than in FETs with comparable noise. Since the transistor input ca-
pacitance enters into the total input capacitance, this is an advantage. Note
that capacitive matching does not apply to bipolar transistors, because their
Analog and Digital Electronics for Detectors 137

AMPLITUDE

AMPLITUDE
VT 2!n

VT dV = max
dt

TIME TIME

2!t

Figure 24: Fluctuations in signal amplitude crossing a threshold translate into timing fluc-
tuations (left). With realistic pulses the slope changes with amplitude, so minimum timing
jitter occurs with the trigger level at the maximum slope.

noise current contribution is significant. Due to the base current noise bipolar
transistors are best at short shaping times, where they also require lower power
than FETs for a given noise level.
When the input noise current is negligible, the noise increases linearly with
sensor capacitance. The noise slope
�
dQn Fv
≈ 2en · (33)
dCd T

depends both on the preamplifier (en ) and the shaper (Fv , T ). The zero inter-
cept can be used to determine the amplifier input capacitance plus any addi-
tional capacitance at the input node.
Practical noise levels range from < 1 e for CCDs at long shaping times
to ∼ 104 e in high-capacitance liquid Ar calorimeters. Silicon strip detectors
typically operate at ∼ 103 electrons, whereas pixel detectors with fast readout
provide noise of 100 – 200 electrons. Transistor noise is discussed in more detail
in[1].

10 Timing Measurements

Pulse height measurements discussed up to now emphasize measurement

of signal charge. Timing measurements seek to optimize the determination of
the time of occurrence. Although, as in amplitude measurements, signal-to-
noise ratio is important, the determining parameter is not signal-to-noise, but
slope-to-noise ratio. This is illustrated in Figure 24, which shows the leading
138 Helmuth Spieler

edge of a pulse fed into a threshold discriminator (comparator), a “leading

edge trigger”. The instantaneous signal level is modulated by noise, where the
variations are indicated by the shaded band. Because of these fluctuations, the
time of threshold crossing fluctuates. By simple geometrical projection, the
timing variance, or “jitter”

σn tr
σt = ≈ , (34)
(dS/dt)ST S/N

where σn is the rms noise and the derivative of the signal dS/dt is evaluated
at the trigger level ST . To increase dS/dt without incurring excessive noise the
amplifier bandwidth should match the rise-time of the detector signal.The 10 –
90% rise time of an amplifier with bandwidth fu (see Figure 13) is

2.2 0.35
tr = 2.2τ = = . (35)
2πfu fu

For example, an oscilloscope with 350 MHz bandwidth has a 1 ns rise time.
When amplifiers are cascaded, which is invariably necessary, the individual rise
times add in quadrature
�
tr ≈ t2r1 + t2r2 + ... + t2rn . (36)

Increasing signal-to-noise ratio improves time resolution, so minimizing the

total capacitance at the input is also important. At high signal-to-noise ratios
the time jitter can be much smaller than the rise time.
The second contribution is time walk, where the timing signal shifts with
amplitude as shown in Figure 25. This can be corrected by various means, either
in hardware or software. For a more detailed tutorial on timing measurements
see ref. [11].

11 Digital Electronics

Analog signals utilize continuously variable properties of the pulse to

impart information, such as the pulse amplitude or pulse shape. Digital signals
have constant amplitude, but the presence of the signal at specific times is
evaluated, i.e. whether the signal is in one of two states, “low” or “high”.
However this still involves an analog process, as the presence of a signal is
determined by the signal level exceeding a threshold at the proper time.
Analog and Digital Electronics for Detectors 139

AMPLITUDE

VT

TIME

!T = “WALK”
Figure 25: The time at which a signal crosses a fixed threshold depends on the signal
amplitude, leading to “time walk”.

11.1 Logic elements

Figure 26 illustrates several functions utilized in digital circuits (“logic”
functions). An AND gate provides an output only when all inputs are high. An
OR gives an output when any input is high. An eXclusive OR (XOR) responds
when only one input is high. The same elements are commonly implemented
with inverted outputs, then called NAND and NOR gates, for example. The

A A
AND EXCLUSIVE
B OR B

A A

B B

D FLIP-FLOP D
A (LATCH)
OR
B CLK
D Q
A
Q
B CLK
?

Figure 26: Basic logic functions include gates (AND, OR, Exclusive OR) and flip flops. The
outputs of the AND and D flip flop show how small shifts in relative timing between inputs
can determine the output state.
140 Helmuth Spieler

AND NAND INVERTER R-S FLIP-FLOP

S Q

R Q

OR NOR EXCLUSIVE OR LATCH

D Q

CLK Q

Figure 27: Some common logic symbols. Inverted outputs are denoted by small circles or
by a superimposed bar, as for the latch output Q. Additional inputs can be added to gates
as needed. An R-S flip-flop sets the Q output high in response to an S input. An R input
resets the Q output to low.

D flip-flop is a bistable memory circuit that records the presence of a signal at

the data input D when a signal transition occurs at the clock input CLK. This
device is commonly called a latch. Inverted inputs and outputs are denoted by
small circles or by superimposed bars, e.g. Q is the inverted output of a flip
flop, as shown in Figure 27.
Logic circuits are fundamentally amplifiers, so they also suffer from band-
width limitations. The pulse train of the AND gate in Figure 26 illustrates a
common problem. The third pulse of input B is going low at the same time
that input A is going high. Depending on the time overlap, this can yield a
narrow output that may or may not be recognized by the following circuit. In
an EX-OR this can occur when two pulses arrive nearly at the same time. The
D flip-flop requires a minimum setup time for a level change at the D input to
be recognized, so changes in the data level may not be recognized at the correct
time. These marginal events may be extremely rare and perhaps go unnoticed.
However, in complex systems the combination of “glitches” can make the sys-
tem “hang up”, necessitating a system reset. Data transmission protocols have
been developed to detect such errors (parity checks, Hamming codes, etc.), so
corrupted data can be rejected.
Some key aspects of logic systems can be understood by inspecting the
circuit elements that are used to form logic functions. Figure 28 shows sim-
ple inverter circuits using MOS transistors. For this discussion it is sufficient to
know that in an NMOS transistor a conductive channel is formed when the input
electrode is biased positive with respect to the channel. The input, called the
“Gate” (G), is capacitively coupled to the output channel connected between
the “Drain” (D) and “Source” (S) electrodes. In the NMOS inverter applying
a positive voltage to the gate makes the output channel conduct, so the output
Analog and Digital Electronics for Detectors 141

VDD
VDD

VDD VDD S VDD

G
0 D 0 0
G D

VDD S VDD VDD

0 0 0

Figure 28: In an NMOS inverter the transistor conducts when the input is high (left),
whereas in a PMOS inverter the transistor conducts when the input is low (right). In both
circuits the input pulse is inverted, whether the input swings high or low.

level is low. A PMOS transistor is the complementary device, where a con-

ductive channel is formed when the gate is biased negative with respect to the
source. Since the source is at positive potential, a low level at the inverter input
yields a high level at the output. Regardless of the device and pulse polarity,
the output pulse is always the inverse of the input. NMOS and PMOS invert-
ers draw current when in their “active” state. Combining NMOS and PMOS
transistors in a complementary MOS (CMOS) circuit allows zero current draw
in both the high and low states with a substantial reduction in power consump-
tion. A CMOS inverter is shown in Figure 29, which also shows how devices
are combined to form a CMOS NAND gate. In the inverter the lower (NMOS)
transistor is turned off when the input is low, but the upper (PMOS) transistor
is turned on, so the output is connected to VDD , taking the output high. Since
the current path from VDD to ground is blocked by either the NMOS or PMOS
device being off, the power dissipation is zero in both the high and low states.
Current only flows during the level transition when both devices are on as the
input level is at approximately VDD /2. As a result, the power dissipation of
CMOS logic is significantly less than in NMOS or PMOS circuitry. This reduc-
tion in power only obtains in logic circuitry. CMOS analog amplifiers are not
fundamentally more power efficient than NMOS or PMOS circuits, although
CMOS provides greater flexibility in the choice of circuit topologies, which can
reduce overall power.

11.2 Propagation delays and power dissipation

Logic elements always operate in conjunction with other circuits, as

illustrated in Figure 30. The wiring resistance in conjunction with the total
142 Helmuth Spieler

load capacitance increases the rise time of the logic pulse and as a result delays
the time when the transition crosses the logic threshold. The energy dissipated
in the wiring resistance R is
�
E = i2 (t)R dt . (37)

The current flow during one transition

� �
V t
i(t) = exp − , (38)
R RC

so the dissipated energy per transition (either positive or negative)

�∞ � �
V2 2t 1
E= exp − dt = CV 2 . (39)
R RC 2
0

When pulses occur at a frequency f , the power dissipated in both the positive
and negative transitions
P = f CV 2 . (40)
Thus, the power dissipation increases with clock frequency and the square of
the logic swing.
Fast logic is time-critical. It relies on logic operations from multiple paths
coming together at the right time. Valid results depend on maintaining mini-
mum allowable overlaps and set-up times as illustrated in Figure 26. Each logic

VDD

VDD

VDD
VDD VDD
VDD 0
0 0
0
VDD

Figure 29: A CMOS inverter (left) and NAND gate (right).

Analog and Digital Electronics for Detectors 143

CASCADED CMOS STAGES EQUIVALENT CIRCUIT

V R i

0
VTH C VTH

T WIRING T+!t
RESISTANCE
SUM OF INPUT
CAPACITANCES

Figure 30: The wiring resistance together with the distributed load capacitance delays the
signal.

circuit has a finite propagation delay, which depends on circuit loading, i.e.
how many loads the circuit has to drive. In addition, as illustrated in Figure 30
the wiring resistance and capacitive loads introduce delay. This depends on the
number of circuits connected to a wire or trace, the length of the trace and the
dielectric constant of the substrate material. Relying on control of circuit and
wiring delays to maintain timing requires great care, as it depends on circuit
variations and temperature. In principle all of this can be simulated, but in
complex systems there are too many combinations to test every one. A more
robust solution is to use synchronous systems, where the timing of all tran-
sitions is determined by a master clock. Generally, this does not provide the
utmost speed and requires some additional circuitry, but increases reliability.
Nevertheless, clever designers frequently utilize asynchronous logic. Sometimes
it succeeds . . . and sometimes it doesn’t.

11.3 Logic arrays

Commodity integrated circuits with basic logic blocks are readily avail-
able, e.g. with four NAND gates or two flip-flops in one package. These can
be combined to form simple digital systems. However, complex logic systems
are no longer designed using individual gates. Instead, logic functions are de-
scribed in a high-level language (e.g. VHDL), synthesized using design libraries,
and implemented as custom ICs – “ASICs” (application specific ICs) – or pro-
grammable logic arrays. In these implementations the digital circuitry no longer
appears as an ensemble of inverters, gates, and flip-flops, but as an integrated
logic block that provides specific outputs in response to various input combina-
144 Helmuth Spieler

tions. This is illustrated in Figure 31. Field Programmable Gate or logic Arrays
(FPGAs) are a common example. A representative FPGA has 512 pads usable
for inputs and outputs, ∼ 106 gates, and ∼ 100K of memory. Modern design
tools also account for propagation delays, wiring lengths, loads, and tempera-
ture dependence. The design software also generates “test vectors” that can be
used to test finished parts. Properly implemented, complex digital designs can
succeed on the first pass, whether as ASICs or as logic or gate arrays.

12 Analog-to-digital Converters (ADCs)

For data storage and subsequent analysis the analog signal at the shaper
output must be digitized. Important parameters for analog-to-digital converters
(ADCs or A/Ds) used in detector systems are

1. Resolution: The “granularity” of the digitized output.

2. Differential non-linearity: How uniform are the digitization increments?
3. Integral non-linearity: Is the digital output proportional to the analog
input?
4. Conversion time: How much time is required to convert an analog signal
to a digital output?
5. Count-rate performance: How quickly can a new conversion commence
after completion of a prior one without introducing deleterious artifacts?
6. Stability: Do the conversion parameters change with time?

Instrumentation ADCs used in industrial data acquisition and control sys-

tems share most of these requirements. However, detector systems place greater

INPUTS LOGIC ARRAY OUTPUTS

Figure 31: Complex logic circuits are commonly implemented using logic arrays that as an
integrated block provide the desired outputs in response to specific input combinations.
Analog and Digital Electronics for Detectors 145

Vref COMPARATORS

INPUT ENCODER
R

R
DIGITIZED
OUTPUT
R

Figure 32: Block diagram of a flash ADC.

emphasis on differential non-linearity and count-rate performance. The latter

is important, as detector signals often occur randomly, in contrast to systems
where signals are sampled at regular intervals. As in amplifiers, if the DC gain
is not precisely equal to the high-frequency gain, the baseline will shift. Further-
more, following each pulse it takes some time for the baseline to return to its
quiescent level. For periodic signals of roughly equal amplitude these baseline
deviations will be the same for each pulse, but for a random sequence of pulse
with varying amplitudes, the instantaneous baseline level will be different for
each pulse and affect the peak amplitude.
Conceptually, the simplest technique is flash conversion, illustrated in Fig-
ure 32. The signal is fed in parallel to a bank of threshold comparators. The
individual threshold levels are set by a resistive divider. The comparator out-
puts are encoded such that the output of the highest level comparator that
fires yields the correct bit pattern. The threshold levels can be set to provide a
linear conversion characteristic where each bit corresponds to the same analog
increment, or a non-linear characteristic, to provide increments proportional to
the absolute level, which provides constant relative resolution over the range.
The big advantage of this scheme is speed; conversion proceeds in one
step and conversion times < 10 ns are readily achievable. The drawbacks are
component count and power consumption, as one comparator is required per
146 Helmuth Spieler

conversion bin. For example, an 8-bit converter requires 256 comparators. The
conversion is always monotonic and differential non-linearity is determined by
the matching of the resistors in the threshold divider. Only relative matching
is required, so this topology is a good match for monolithic integrated circuits.
Flash ADCs are available with conversion rates > 500 MS/s (megasamples per
second) at 8-bit resolution and a power dissipation of about 5 W.
The most commonly used technique is the successive approximation ADC,
shown in Figure 33. The input pulse is sent to a pulse stretcher, which follows
the signal until it reaches its cusp and then holds the peak value. The stretcher
output feeds a comparator, whose reference is provided by a digital-to-analog
converter (DAC). The DAC is cycled beginning with the most significant bits.
The corresponding bit is set when the comparator fires, i.e. the DAC output
becomes less than the pulse height. Then the DAC cycles through the less
significant bits, always setting the corresponding bit when the comparator fires.
Thus, n-bit resolution requires n steps and yields 2n bins. This technique makes
efficient use of circuitry and is fairly fast. High-resolution devices (16 – 20 bits)
with conversion times of order µs are readily available. Currently a 16-bit ADC
with a conversion time of 1 µs (1 MS/s) requires about 100 mW.
A common limitation is differential non-linearity, since the resistors that
set the DAC levels must be extremely accurate. For DNL < 1% the resistor
determining the 212 -level in a 13-bit ADC must be accurate to < 2.4 · 10−6 . As
a consequence, differential non-linearity in high-resolution successive approxi-
mation converters is typically 10 – 20% and often exceeds the 0.5 LSB (least
significant bit) required to ensure monotonic response.

PULSE CONTROL
COMPARATOR
STRETCHER LOGIC

ANALOG
INPUT DIGITIZED
OUTPUT

DAC
ADDRESS
DAC

Figure 33: Principle of a successive approximation ADC. The DAC is controlled to se-
quentially add levels proportional to 2n , 2n−1 , . . . 20 . The corresponding bit is set if the
comparator output is high (DAC output < pulse height).
Analog and Digital Electronics for Detectors 147

START STOP

VBL
PULSE
STRETCHER COMPARATOR

ANALOG
INPUT

IR

VBL
PEAK COUNTER
DETECTOR
OUTPUT

CLOCK

DIGITIZED
OUTPUT

Figure 34: Principle of a Wilkinson ADC. After the peak amplitude has been acquired,
the output of the peak detector initiates the conversion process. The memory capacitor is
discharged by a constant current while counting the clock pulses. When the capacitor is
discharged to the baseline level VBL the comparator output goes low and the conversion is
complete.

The Wilkinson ADC[12] has traditionally been the mainstay of precision

pulse digitization. The principle is shown in Figure 34. The peak signal ampli-
tude is acquired by a combined peak detector/pulse stretcher and transferred to
a memory capacitor. The output of the peak detector initiates the conversion
process:
1. The memory capacitor is disconnected from the stretcher,
2. a current source is switched on to linearly discharge the capacitor with
current IR , and simultaneously
3. a counter is enabled to determine the number of clock pulses until the
voltage on the capacitor reaches the baseline level VBL .
The time required to discharge the capacitor is a linear function of pulse height,
so the counter content provides the digitized pulse height. The clock pulses
are provided by a crystal oscillator, so the time between pulses is extremely
uniform and this circuit inherently provides excellent differential linearity. The
drawback is the relatively long conversion time TC , which for a given resolution
is proportional to the pulse height, TC = n × Tclk , where n is the channel
148 Helmuth Spieler

number corresponding to the pulse height. For example, a clock frequency of

100 MHz provides a clock period Tclk = 10 ns and a maximum conversion time
TC = 82 µs for 13 bits (n = 8192). Clock frequencies of 100 MHz are typical, but
> 400 MHz have been implemented with excellent performance (DNL < 10−3 ).
This scheme makes efficient use of circuitry and allows low power dissipation.
Wilkinson ADCs have been implemented in 128-channel readout ICs for silicon
strip detectors[13]. Each ADC added only 100 µm to the length of a channel
and a power of 300 µW per channel.

13 Time-to-digital Converters (TDCs)

The combination of a clock generator with a counter is the simplest

technique for time-to-digital conversion, as shown in Figure 35. The clock pulses
are counted between the start and stop signals, which yields a direct readout in
real time. The limitation is the speed of the counter, which in current technology
is limited to about 1 GHz, yielding a time resolution of 1 ns. Using the stop pulse
to strobe the instantaneous counter status into a register provides multi-hit
capability.
Analog techniques are commonly used in high-resolution digitizers to pro-
vide resolution in the range of ps to ns. The principle is to convert a time interval
into a voltage by charging a capacitor through a switchable current source. The
start pulse turns on the current source and the stop pulse turns it off. The re-
sulting voltage on the capacitor C is V = Q/C = IT (Tstop − Tstart )/C, which is
digitized by an ADC. A convenient implementation switches the current source
to a smaller discharge current IR and uses a Wilkinson ADC for digitization,

CLOCK
COUNTER

START S Q

DIGITIZED
STOP R OUTPUT
START STOP

Figure 35: The simplest form of time digitizer counts the number of clock pulses between
the start and stop signals.
Analog and Digital Electronics for Detectors 149

V+

START

STOP

IT COMPARATOR

IR C

VBL
COUNTER

CLOCK

DIGITIZED
OUTPUT

Figure 36: Combining a time-to-amplitude converter with an ADC forms a time digitizer
capable of ps resolution. The memory capacitor C is charged by the current IT for the
duration Tstart − Tstop and subsequently discharged by a Wilkinson ADC.

as illustrated in Figure 36. This technique provides high resolution, but at the
expense of dead time and multi-hit capability.

14 Signal Transmission

Signals are transmitted from one unit to another through transmission

lines, often coaxial cables or ribbon cables. When transmission lines are not
terminated with their characteristic impedance, the signals are reflected. As a
signal propagates along the cable, the ratio of instantaneous
� voltage to current
equals the cable’s characteristic impedance Z0 = L/C, where L and C are
the inductance and capacitance per unit length. Typical impedances are 50
or 75 Ω for coaxial cables and ∼ 100 Ω for ribbon cables. If at the receiving
end the cable is connected to a resistance different from the cable impedance, a
different ratio of voltage to current must be established. This occurs through a
reflected signal. If the termination is less than the line impedance, the voltage
150 Helmuth Spieler

TERMINATION: SHORT OPEN

PRIMARY PULSE

REFLECTED
PULSE

2td

PULSE SHARPE
AT ORIGIN

Figure 37: Voltage pulse reflections on a transmission line terminated either with a short
(left) or open circuit (right). Measured at the sending end, the reflection from a short at
the receiving end appears as a pulse of opposite sign delayed by the round trip delay of the
cable. If the total delay is less than the pulse width, the signal appears as a bipolar pulse.
Conversely, an open circuit at the receiving end causes a reflection of like polarity.

must be smaller and the reflected voltage wave has the opposite sign. If the
termination is greater than the line impedance, the voltage wave is reflected
with the same polarity. Conversely, the current in the reflected wave is of like
sign when the termination is less than the line impedance and of opposite sign
when the termination is greater. Voltage reflections are illustrated in Figure 37.
At the sending end the reflected pulse appears after twice the propagation delay
of the √cable. Since in the presence of a dielectric the velocity of propagation
v = c/ ε, in typical coaxial and ribbon cables the delay is 5 ns/m.
Cable drivers often have a low output impedance, so the reflected pulse is
reflected again towards the receiver, to be reflected again, etc. This is shown in
Figure 38, which shows the observed signal when the output of a low-impedance
pulse driver is connected to a high-impedance amplifier input through a 4 m
long 50 Ω coaxial cable. If feeding a counter, a single pulse will be registered
multiple times, depending on the threshold level. When the amplifier input is
terminated with 50 Ω, the reflections disappear and only the original 10 ns wide
pulse is seen.
There are two methods of terminating cables, which can be applied either
Analog and Digital Electronics for Detectors 151

individually or – in applications where pulse fidelity is critical – in combination.

As illustrated in Figure 39 the termination can be applied at the receiving or
the sending end. Receiving end termination absorbs the signal pulse when it
arrives at the receiver. With sending-end termination the pulse is reflected at the
receiver, but since the reflected pulse is absorbed at the sender, no additional
pulses are visible at the receiver. At the sending end the original pulse is
attenuated two-fold by the voltage divider formed by the series resistor and the
cable impedance. However, at the receiver the pulse is reflected with the same
polarity, so the superposition of the original and the reflected pulses provides
the original amplitude.
This example uses voltage amplifiers, which have low output and high
input impedances. It is also possible to use current amplifiers, although this
is less common. Then, the amplifier has a high output impedance and low
input impedance, so shunt termination is applied at the sending end and series
termination at the receiving end.
Terminations are never perfect, especially at high frequencies where stray
capacitance becomes significant. For example, the reactance of 10 pF at 100 MHz
is 160 Ω. Thus, critical applications often use both series and parallel termina-
tion, although this does incur a 50% reduction in pulse amplitude. In the µs
regime, amplifier inputs are usually designed as high impedance, whereas tim-
ing amplifiers tend to be internally terminated, but one should always check
if this is the case. As a rule of thumb, whenever the propagation delay of ca-
bles (or connections in general) exceeds a few percent of the signal risetime,
terminations are required.

1
2
VOLTAGE (V)

VOLTAGE (V)

1
0.5
0

-1
0
-2

0 200 400 600 800 1000 0 200 400 600 800 1000
TIME (ns) TIME (ns)

Figure 38: Left: Signal observed in an amplifier when a low-impedance driver is connected
to the amplifier through a 4 m long coaxial cable. The cable impedance is 50 Ω and the
amplifier input appears as 1 kΩ in parallel with 30 pF. When the receiving end is properly
terminated with 50 Ω, the reflections disappear (right).
152 Helmuth Spieler

Z0

RT = Z0

RT = Z0 Z0

Figure 39: Cables may be terminated at the receiving end (top, shunt termination) or
sending end (bottom, series termination).

15 Interference and Pickup

The previous discussion analyzed random noise sources inherent to the

sensor and front-end electronics. In practical systems external noise often lim-
its the obtainable detection threshold or energy resolution. As with random
noise, external pickup introduces baseline fluctuations. There are many possi-
ble sources, radio and television stations, local RF generators, system clocks,
transients associated with trigger signals and data readout, etc. Furthermore,
there are many ways through which these undesired signals can enter the sys-
tem. Again, a comprehensive review exceeds the allotted space, so only a few
key examples of pickup mechanisms will be shown. A more detailed discussion
is in refs [1] and [2]. Ott[14] gives a more general treatment and texts by John-
son and Graham[15][16] give useful details on signal transmission and design
practices.

15.1 Pickup mechanisms

The most sensitive node in a detector system is the input. Figure 40

shows how very small spurious signals coupled to the sensor backplane can
inject substantial charge. Any change in the bias voltage ∆V directly at the
sensor backplane will inject a charge ∆Q = Cd ∆V . Assume a silicon strip
sensor with 10 cm strip length. Then the capacitance Cd from the backplane to
a single strip is about 1 pF. If the noise level is 1000 electrons (1.6 · 10−16 C),
∆V must be much smaller than Qn /Cd = 160µV. This can be introduced as
noise from the bias supply (some voltage supplies are quite noisy; switching
power supplies can be clean, but most aren’t) or noise on the ground plane can
Analog and Digital Electronics for Detectors 153

Qi

Cd
R
VBIAS
!V C

Figure 40: Noise on the detector bias line is coupled through the detector capacitance to
the amplifier input.

V1
COMMON
GROUND BUS

!V

V2

Figure 41: Shared current paths introduce common voltage drops to different circuits.

couple through the capacitor C. Naively, one might assume the ground plane
to be “clean”, but it can carry significant interference for the following reason.
One of the most common mechanisms for cross-coupling is shared current
paths, often referred to as “ground loops”. However, this phenomenon is not
limited to grounding. Consider two systems. The first is transmitting large
currents from a source to a receiver. The second is similar, but is attempting
a low-level measurement. Following the prevailing lore, both systems are con-
nected to a massive ground bus, as shown in Figure 41. Current seeks the path
of least resistance, so the large current from source V1 will also flow through the
ground bus. Although the ground bus is massive, it does not have zero resis-
tance, so the large current flowing through the ground system causes a voltage
drop ∆V .
154 Helmuth Spieler

In system 2 (source V2 ) both signal source and receiver are also connected
to the ground system. Now the voltage drop ∆V from system 1 is in series
with the signal path, so the receiver measures V2 + ∆V . The cross-coupling has
nothing to do with grounding per se, but is due to the common return path.
However, the common ground caused the problem by establishing the shared
path. This mechanism is not limited to large systems with external ground
busses, but also occurs on the scale of printed circuit boards and micron-scale
integrated circuits. At high frequencies the impedance is increased due to skin
effect and inductance. Note that for high-frequency signals the connections can
be made capacitively, so even if there is no DC path, the parasitic capacitance
due to mounting structures or adjacent conductor planes can be sufficient to
close the loop.
The traditional way of dealing with this problem is to reduce the impedance
of the shared path, which leads to the “copper braid syndrome”. However,
changes in the system will often change the current paths, so this “fix” is not
very reliable. Furthermore, in many detector systems – tracking detectors, for
example – the additional material would be prohibitive. Instead, it is best to
avoid the root cause.

15.2 Remedial techniques

Figure 42 shows a sensor connected to a multistage amplifier. Signals are
transferred from stage to stage through definite current paths. It is critical to
maintain the integrity of the signal paths, but this does not depend on grounding
– indeed Figure 42 does not show any ground connection at all. The most

+VDET
+V

Q2
Q1 Q3
OUTPUT
DETECTOR
–V
–VDET

Figure 42: The signal is transferred from the sensor to the input stage and from stage to
stage via local current loops.
Analog and Digital Electronics for Detectors 155

critical parts of this chain are the input, which is the most sensitive node, and
the output driver, which tends to circulate the largest current. Circuit diagrams
usually are not drawn like Figure 42; the bottom common line is typically shown
as ground. For example, in Figure 40 the sensor signal current flows through
capacitor C and reaches the return node of the amplifier through “ground”.
Clearly, it is critical to control this path and keep deleterious currents from this
area.
SUPPORT /
COOLING STAVE

DETECTOR

SIGNAL OUTPUT

DETECTOR BIAS

ISOLATION
RESISTORS

DETECTOR

SIGNAL OUTPUT

DETECTOR BIAS

ISOLATION
RESISTORS

Figure 43: Capacitive coupling between detectors or detector modules and their environment
introduces interference when relative potentials and stray capacitance are not controlled.

However superfluous grounding may be, one cannot let circuit elements
simply float with respect to their environment. Capacitive coupling is always
present and any capacitive coupling between two points of different potential
will induce a signal. This is illustrated in Figure 43, which represents individual
detector modules mounted on a support/cooling structure. Interference can
couple through the parasitic capacitance of the mount, so it is crucial to reduce
156 Helmuth Spieler

this capacitance and control the potential of the support structure relative to the
detector module. Attaining this goal in reality is a challenge, which is not always
met successfully. Nevertheless, paying attention to signal paths and potential
references early on is much easier than attempting to correct a poor design after
it’s done. Troubleshooting is exacerbated by the fact that current paths interact,
so doing the “wrong” thing sometimes brings improvement. Furthermore, only
one mistake can ruin system performance, so if this has been designed into the
system from the outset, one is left with compromises. Nevertheless, although
this area is rife with myths, basic physics still applies.

16 Conclusion

Signal processing is a key part of modern detector systems. Proper design

is especially important when signals are small and electronic noise determines
detection thresholds or resolution. Optimization of noise is well understood
and predicted noise levels can be achieved in practical experiments within a few
percent of predicted values. However, systems must be designed very carefully
to avoid extraneous pickup.

References

[1] H. Spieler, Semiconductor Detector Systems, Oxford University Press, Ox-

ford, 2005. ISBN 0-19-852784-5

[2] http://www-physics.lbl.gov/∼spieler

[3] J. Butler, Triggering and Data Acquisition General Considerations, in In-

strumentation in Elementary Particle Physics, AIP Conf. Proc. 674 (2003)
101–129

[4] T. Kondo et al., Construction and performance of the ATLAS silicon mi-
crostrip barrel modules. Nucl. Instr. and Meth. A485 (2002) 27–42

[5] I. Kipnis, H. Spieler and T. Collins, A Bipolar Analog Front-End Integrated

Circuit for the SDC Silicon Tracker, IEEE Trans. Nucl. Sci. NS-41/4
(1994) 1095–1103

[6] S. Ramo, Currents Induced by Electron Motion. Proc. IRE 27 (1939) 584–
585

[7] F. S. Goulding, Pulse Shaping in Low-Noise Nuclear Amplifiers: A Physical

Approach to Noise Analysis, Nucl. Instr. Meth. 100 (1972) 493–504
Analog and Digital Electronics for Detectors 157

[8] F. S. Goulding and D.A. Landis, Signal Processing for Semiconductor De-
tectors, IEEE Trans. Nucl. Sci. NS-29/3 (1982) 1125–1141

[9] V. Radeka, Trapezoidal Filtering of Signals from Large Germanium Detec-

tors at High Rates, Nucl. Instr. Meth. 99 (1972) 525–539
[10] V. Radeka, Signal, Noise and Resolution in Position-Sensitive Detectors,
IEEE Trans. Nucl. Sci. NS-21 (1974) 51–64
[11] H. Spieler, Fast Timing Methods for Semiconductor Detectors, IEEE
Trans. Nucl. Sci. NS-29/3 (1982) 1142–1158
[12] D. H. Wilkinson, A Stable Ninety-Nine Channel Pulse Amplitude Analyser
for Slow Counting. Proc. Cambridge Phil. Soc. 46/3 (1950) 508–518
[13] M. Garcia-Sciveres et al., The SVX3D integrated circuit for dead-timeless
silicon strip readout, Nucl. Instr. and Meth. A435 (1999) 58–64
[14] H. W. Ott, Noise Reduction Techniques in Electronic Systems (2nd edn),
Wiley, New York, 1988, ISBN 0-471-85068-3, TK7867.5.087
[15] H. Johnson and M. Graham, High-Speed Digital Design, Prentice Hall PTR,
Upper Saddle River, 1993 ISBN, 0-13-395724-1, TK7868.D5J635
[16] H. Johnson and M. Graham, High-Speed Signal Propagation, Prentice Hall
PTR, Upper Saddle River, 2002, ISBN 0-13-084408-X, TK5103.15.J64
Review Talks
 The Electroweak Interactions in the Standard
Model and beyond
Guido Altarelli
CERN, Department of Physics, Theory Division
CH–1211 Geneva 23, Switzerland

ABSTRACT

We present a concise review of the status of the Standard Model and of the
models of new physics.

1 Precision Tests of the Standard Model

The results of the electroweak precision tests as well as of the searches for the
Higgs boson and for new particles performed at LEP and SLC are now available
in nearly final form. Taken together with the measurements of mt , mW and
the searches for new physics at the Tevatron, and with some other data from
low energy experiments, they form a very stringent set of precise constraints
[1] to compare with the Standard Model (SM) or with any of its conceivable
extensions. When confronted with these results, on the whole the SM performs
rather well, so that it is fair to say that no clear indication for new physics
emerges from the data [2].
All electroweak Z pole measurements, combining the results of the 5 ex-
periments, are summarised in Table 1. Information on the Z partial widths are
contained in the quantities:

12π Γee Γhad σh0 Γhad Γqq̄

σh0 = , R�0 = = , Rq0 = . (1)
m2Z Γ2Z σ�0 Γ�� Γhad

Here �� is the partial decay width for a pair of massless charged leptons. The
partial decay width for a given fermion species are related to the effective vector
and axial-vector coupling constants of the neutral weak current:

GF m3 � 2 �
Γf f̄ = NCf √ Z gAf CAf + gVf
2
CVf + ∆ew/QCD , (2)
6 2π

where NCf is the QCD colour factor, C{A,V}f are final-state QCD/QED cor-
rection factors also absorbing imaginary contributions to the effective coupling
162 Guido Altarelli

Observable Measurement SM fit

mZ [GeV] 91.1875 ± 0.0021 91.1873
ΓZ [GeV] 2.4952 ± 0.0023 2.4965
σh0 [nb] 41.540 ± 0.037 41.481
R�0 20.767 ± 0.025 20.739
A0,
FB
�
0.0171 ± 0.0010 0.0164
A� (SLD) 0.1513 ± 0.0021 0.1480
A� (Pτ ) 0.1465 ± 0.0033 0.1480
Rb0 0.21644 ± 0.00065 0.21566
Rc0 0.1718 ± 0.0031 0.1723
A0,
FB
b
0.0995 ± 0.0017 0.1037
A0,
FB
c
0.0713 ± 0.0036 0.0742
Ab 0.922 ± 0.020 0.935
Ac 0.670 ± 0.026 0.668
lept
sin 2
θeff (Qhad
FB ) 0.2324 ± 0.0012 0.23140
mW [GeV] 80.425 ± 0.034 80.398
ΓW [GeV] 2.133 ± 0.069 2.094
mt [GeV] (pp [5]) 178.0 ± 4.3 178.1
(5)
∆αhad (m2Z ) [6] 0.02761 ± 0.00036 0.02768

Table 1: Summary of electroweak precision measurements at high Q2 [3]. The first block
shows the Z-pole measurements. The second block shows additional results from other ex-
periments: the mass and the width of the W boson measured at the Tevatron and at LEP-2,
the mass of the top quark measured at the Tevatron, and the contribution to α(m2Z ) of the
hadronic vacuum polarisation.
The Electroweak Interactions in the Standard Model and beyond 163

constants, gAf and gVf are the real parts of the effective couplings, and ∆ con-
tains non-factorisable mixed corrections.
Besides total cross sections, various types of asymmetries have been mea-
sured. The results of all asymmetry measurements are quoted in terms of the
asymmetry parameter Af , defined in terms of the real parts of the effective
coupling constants, gVf and gAf , as:

gVf gAf gVf /gAf 3

Af = 2 = 2 , A0, f
FB = Ae Af . (3)
gVf + gAf
2 2 1 + (gVf /gAf )2 4

The measurements are: the forward-backward asymmetry (A0, f

FB = (3/4)Ae Af ),
the tau polarisation (Aτ ) and its forward backward asymmetry (Ae ) measured
at LEP, as well as the left-right and left-right forward-backward asymmetry
measured at SLC (Ae and Af , respectively). Hence the set of partial width and
asymmetry results allows the extraction of the effective coupling constants. In
particular, from the measurements at the Z, lepton universality of the neutral
weak current was established at the per-mille level.
Using the effective electroweak mixing angle, sin2 θeff
f
, and the ρ parameter,
the effective coupling constants are given by:
√ gVf
gAf = ρ T3f , = 1 − 4|qf | sin2 θeff
f
, (4)
gAf

where T3f is the third component of the weak iso-spin and qf the electric charge
of the fermion. The effective electroweak mixing angle is thus given indepen-
dently of the ρ parameter by the ratio gVf /gAf and hence in a one-to-one relation
by each asymmetry result.
The various asymmetries determine the effective electroweak mixing angle
lept
for leptons with highest sensitivity. The results on sin2 θeff are compared in
Figure 1. The weighted average of these six results, including small correlations,
is:
lept
sin2 θeff = 0.23150 ± 0.00016 . (5)

Note, however, that this average has a χ2 of 10.5 for 5 degrees of freedom,
corresponding to a probability of 6.2%. The χ2 is pushed up by the two most
lept
precise measurements of sin2 θeff , namely those derived from the measurements
of A� by SLD, dominated by the left-right asymmetry A0LR , and of the forward-
backward asymmetry measured in bb production at LEP, A0, b
FB , which differ by
about 2.9 standard deviations. No experimental effect in either measurement
has been identified to explain this, thus the difference is presumably either the
164 Guido Altarelli

Final
0,l
A fb 0.23099 0.00053
A l(SLD) 0.23098 0.00026
A l(Pt) 0.23159 0.00041
had
Q fb 0.2324 0.0012

Preliminary
0,b
A fb 0.23212 0.00029
0,c
A fb 0.23223 0.00081

Average 0.23150 0.00016

10
3 2
/d.o.f.: 10.5 / 5
mH [GeV]

(5)
2 Dahad = 0.02761 0.00036
10
mt = 178.0 4.3 GeV

0.23 0.232 0.234

lept
sin2 q
eff

Figure 1: Effective electroweak mixing angle sin2 θeff lept

derived from measurement results
depending on lepton couplings only (top) and also quark couplings (bottom) [1]. Also shown is
lept
the prediction of sin2 θeff in the SM as a function of mH , including its parametric uncertainty
(5)
dominated by the uncertainties in ∆αhad (m2Z ) and mt , shown as the bands.
The Electroweak Interactions in the Standard Model and beyond 165

effect of statistics or an unidentified systematics or a hint for new physics, as

further discussed below.
Also shown in Table 1 are the results on mW obtained at LEP-2 and at
the Tevatron, and the new world average of the top mass.
For the analysis of electroweak data in the SM one starts from the input
parameters: as in any renormalisable theory masses and couplings have to be
specified from outside. One can trade one parameter for another and this free-
dom is used to select the best measured ones as input parameters. As a result,
some of them, α, GF and mZ , are very precisely known [4], some other ones,
mflight , mt and αs (mZ ) are far less well determined while mH is largely un-
known. Note that the new combined CDF and DØ value for mt [5], as listed in
Table 1, is higher than the previous average by nearly one standard deviation.
Among the light fermions, the quark masses are badly known, but for-
tunately, for the calculation of radiative corrections, they can be replaced by
α(mZ ), the value of the QED running coupling at the Z mass scale. The value
(5)
of the hadronic contribution to the running, ∆αhad (m2Z ), reported in Table 1, is
obtained through dispersion relations from the data on e+ e− → hadrons at low
centre-of-mass energies [6]. From the input parameters one computes the ra-
diative corrections to a sufficient precision to match the experimental accuracy.
Then one compares the theoretical predictions and the data for the numerous
observables which have been measured, checks the consistency of the theory and
derives constraints on mt , αS (m2Z ) and mH .
The computed radiative corrections include the complete set of one-loop
diagrams, plus some selected large subsets of two-loop diagrams and some se-
quences of resummed large terms of all orders (large logarithms and Dyson re-
summations). In particular large logarithms, e.g., terms of the form
(α/π ln (mZ /mf� ))n where f� is a light fermion, are resummed by well-known
and consolidated techniques based on the renormalisation group. For example,
large logarithms dominate the running of α from me , the electron mass, up
to mZ , which is a 6% effect, much larger than the few per-mille contributions
of purely weak loops. Also, large logs from initial state radiation dramatically
distort the line shape of the Z resonance observed at LEP-1 and SLC and must
be accurately taken into account in the measurement of the Z mass and total
width.
Among the one loop EW radiative corrections, a remarkable class of con-
tributions are those terms that increase quadratically with the top mass. The
large sensitivity of radiative corrections to mt arises from the existence of these
terms. The quadratic dependence on mt (and possibly on other widely broken
isospin multiplets from new physics) arises because, in spontaneously broken
gauge theories, heavy loops do not decouple. On the contrary, in QED or QCD,
166 Guido Altarelli

the running of α and αs at a scale Q is not affected by heavy quarks with

mass M � Q. According to an intuitive decoupling theorem [7], diagrams with
heavy virtual particles of mass M can be ignored for Q � M provided that
the couplings do not grow with M and that the theory with no heavy particles
is still renormalizable. In the spontaneously broken EW gauge theories both
requirements are violated. First, one important difference with respect to un-
broken gauge theories is in the longitudinal modes of weak gauge bosons. These
modes are generated by the Higgs mechanism, and their couplings grow with
masses (as is also the case for the physical Higgs couplings). Second, the theory
without the top quark is no more renormalisable because the gauge symmetry is
broken if the b quark is left with no partner (while its measured couplings show
that the weak isospin is 1/2). Because of non decoupling, precision tests of the
electroweak theory may be sensitive to new physics even if the new particles are
too heavy for their direct production.
While radiative corrections are quite sensitive to the top mass, they are
unfortunately much less dependent on the Higgs mass. If they were suffi-
ciently sensitive, by now we would precisely know the mass of the Higgs. How-
ever, the dependence of one loop diagrams on mH is only logarithmic: ∼
GF m2W log(m2H /m2W ). Quadratic terms ∼ G2F m2H only appear at two loops and
are too small to be important. The difference with the top case is that m2t − m2b
is a direct breaking of the gauge symmetry that already affects the relevant one
loop diagrams, while the Higgs couplings to gauge bosons are ”custodial-SU(2)”
symmetric in lowest order.
We now discuss fitting the data in the SM. One can think of different types
of fit, depending on which experimental results are included or which answers
one wants to obtain. For example, in Table 2 we present in column 1 a fit of
all Z pole data plus mW and ΓW (this is interesting as it shows the value of mt
obtained indirectly from radiative corrections, to be compared with the value of
mt measured in production experiments), in column 2 a fit of all Z pole data plus
mt (here it is mW which is indirectly determined), and, finally, in column 3 a fit
of all the data listed in Table 1 (which is the most relevant fit for constraining
mH ). From the fit in column 1 of Table 2 we see that the extracted value of mt
is in perfect agreement with the direct measurement (see Table 1). Similarly we
see that the experimental measurement of mW in Table 1 is larger by about one
standard deviation with respect to the value from the fit in column 2. We have
seen that quantum corrections depend only logarithmically on mH . In spite
of this small sensitivity, the measurements are precise enough that one still
obtains a quantitative indication of the mass range. From the fit in column 3
we obtain: log10 mH (GeV) = 2.05 ± 0.20 (or mH = 113+62 −42 GeV). This result on
the Higgs mass is particularly remarkable. The value of log10 mH (GeV) is right
The Electroweak Interactions in the Standard Model and beyond 167

on top of the small window between ∼ 2 and ∼ 3 which is allowed, on the one
side, by the direct search limit (mH > ∼ 114 GeV from LEP-2 [8]), and, on the
other side, by the theoretical upper limit on the Higgs mass in the minimal SM,
mH <∼ 600 − 800 GeV [9].

Fit 1 2 3
Measurements mW , ΓW mt mt , mW , ΓW
mt (GeV) 178.5+11.0
−8.5 177.2 ± 4.1 178.1 ± 3.9
mH (GeV) 117+162
−62 129+76
−50 113+62
−42
log [mH (GeV)] 2.07+0.38
−0.33 2.11 ± 0.21 2.05 ± 0.20
αs (mZ ) 0.1187 ± 0.0027 0.1190 ± 0.0027 0.1186 ± 0.0027
2
χ /dof 16.3/12 15.0/11 16.3/13
mW (MeV) 80386 ± 23

Table 2: Standard Model fits of electroweak data. All fits use the Z pole results and
(5)
∆αhad (m2Z ) as listed in Table 1, also including constants such as the Fermi constant GF . In
addition, the measurements listed in each column are included as well. For fit 2, the expected
W mass is also shown. For details on the fit procedure see [3].

A different way of looking at the data is to consider the epsilon parameters.

As well known these parameters vanish in the limit of tree level SM plus pure
QED or pure QCD corrections. So they are a measure of the weak quantum
corrections. Their experimental values are given by [1]:

�1 103 = 5.4 ± 1.0 (6)

�2 103 = −8.9 ± 1.2 (7)
�3 103 = 5.25 ± 0.95 (8)
�b 103 = −4.7 ± 1.6 (9)

The experimental values are compared to the SM predictions as function of mt

and mH in Figure 2. We see that �3 points to a light Higgs, that �b is a bit too
large because of AbF B and �2 a bit too small because of mW .
Thus the whole picture of a perturbative theory with a fundamental Higgs
is well supported by the data on radiative corrections. It is important that there
is a clear indication for a particularly light Higgs: at 95% c.l. mH <∼ 237 GeV.
This is quite encouraging for the ongoing search for the Higgs particle. More
general, if the Higgs couplings are removed from the Lagrangian the resulting
168 Guido Altarelli

10
epsilon . 103
1000
3
5 1
3
1
1000

0
m H = 114 —1000 GeV CDF - D0

-5 b
b
2
114

-10 2

160 165 170 175 180 185 190

mt (GeV)

Figure 2: The epsilon variables: comparison of the data with the SM predictions. The data
should be horizontal bands but they are shown here near the central value of mt .

theory is non renormalisable. A cutoff Λ must be introduced. In the quan-

tum corrections log mH is then replaced by log Λ plus a constant. The precise
determination of the associated finite terms would be lost (that is, the value
of the mass in the denominator in the argument of the logarithm). A heavy
Higgs would need some unfortunate conspiracy: the finite terms, different in
the new theory from those of the SM, should accidentally compensate for the
heavy Higgs in a few key parameters of the radiative corrections (mainly �1 and
�3 , see, for example, [10]). Alternatively, additional new physics, for example in
the form of effective contact terms added to the minimal SM lagrangian, should
accidentally do the compensation, which again needs some sort of conspiracy.
In Table 3 we collect the results on low energy precision tests of the SM
obtained from neutrino and antineutrino deep inelastic scattering (NuTeV [11]),
parity violation in Cs atoms (APV [12]) and the recent measurement of the
parity-violating asymmetry in Moller scattering [13]. The experimental results
are compared with the predictions from the fit in column 3 of Table 2. We see the
agreement is good except for the NuTeV result that shows a deviation by three
standard deviations. The NuTeV measurement is quoted as a measurement of
sin2 θW = 1−m2W /m2Z from the ratio of neutral to charged current deep inelastic
cross-sections from νµ and ν̄µ using the Fermilab beams. There is growing
evidence that the NuTeV anomaly could simply arise from an underestimation
The Electroweak Interactions in the Standard Model and beyond 169

of the theoretical uncertainty in the QCD analysis needed to extract sin2 θW .

In fact, the lowest order QCD parton formalism on which the analysis has been
based is too crude to match the experimental accuracy. In particular a small
asymmetry in the momentum carried by the strange and antistrange quarks,
s − s̄, could have a large effect [14]. A tiny violation of isospin symmetry in
parton distributions, too small to be seen elsewhere, can similarly be of some
importance. In conclusion we believe the discrepancy has more to teach about
the QCD parton densities than about the electroweak theory.

Observable Measurement SM fit

sin2 θW (νN [11]) 0.2277 ± 0.0016 0.2226
QW (Cs) (APV [12]) −72.83 ± 0.49 −72.91
lept
sin2
θeff (e e
− −
[13]) 0.2296 ± 0.0023 0.2314

Table 3: Summary of other electroweak precision measurements, namely the measurements

of the on-shell electroweak mixing angle in neutrino-nucleon scattering, the weak charge of
cesium measured in an atomic parity violation experiment, and the effective weak mixing angle
measured in Moller scattering, all performed in processes at low Q2 . The SM predictions are
derived from fit 3 of Table 2. Good agreement of the prediction with the measurement is
found except for νN.

When confronted with these results, on the whole the SM performs rather
well, so that it is fair to say that no clear indication for new physics emerges
from the data. However, as already mentioned, one problem is that the two
lept
most precise measurements of sin2 θeff from ALR and A0, b
FB differ nearly three
standard deviations. In general, there appears to be a discrepancy between
lept
sin2 θeff measured from leptonic asymmetries ((sin2 θeff )l ) and from hadronic
asymmetries ((sin2 θeff )h ), see also Figure 1. In fact, the result from ALR is
in good agreement with the leptonic asymmetries measured at LEP, while all
hadronic asymmetries, though their errors are large, are better compatible with
the result of A0, b
FB .
The situation is shown in Figure 3 [15]. The values of (sin2 θeff )l , (sin2 θeff )h
and their formal combination are shown each at the mH value that would corre-
spond to it given the central value of mt . Of course, the value for mH indicated
lept
by each sin2 θeff has an horizontal ambiguity determined by the measurement
error and the width of the ±1σ band for mt . Even taking this spread into ac-
count it is clear that the implications on mH are sizably different. One might
imagine that some new physics effect could be hidden in the Zbb̄ vertex. Like
for the top quark mass there could be other non decoupling effects from new
170 Guido Altarelli

0.2325 hadr. asymm

0.232 mt=173.7 GeV

178.0 GeV
lept

eff
sin2 q

182.3 GeV

0.2315 world av.

lept.asymm
0.231

50 100 200 300 400 500

MH [GeV]

Figure 3: lept
The data for sin2 θeff are plotted vs mH . For presentation purposes the measured
points are shown each at the mH value that would ideally correspond to it given the central
value of mt (updated from [15]).

heavy states or a mixing of the b quark with some other heavy quark. However,
it is well known that this discrepancy is not easily explained in terms of some
new physics effect in the Zbb̄ vertex. In fact, A0, b
FB is the product of lepton-
and b-asymmetry factors: A0, b 0, b
FB = (3/4)Ae Ab . The sensitivity of AFB to Ab
is limited, because the Ae factor is small, so that a rather large change of the
b-quark couplings with respect to the SM is needed in order to reproduce the
measured discrepancy (precisely a ∼ 30% change in the right-handed coupling,
an effect too large to be a loop effect but which could be produced at the tree
level, e.g., by mixing of the b quark with a new heavy vectorlike quark [16]).
But then this effect should normally also appear in the direct measurement of
Ab performed at SLD using the left-right polarized b asymmetry, even within
the moderate precision of this result, and it should also be manifest in the accu-
rate measurement of Rb ∝ gRb 2
+ gLb
2
. The measurements of neither Ab nor Rb
confirm the need of a new effect. Even introducing an ad hoc mixing the overall
The Electroweak Interactions in the Standard Model and beyond 171

fit is not terribly good, but we cannot exclude this possibility completely. Alter-
natively, the observed discrepancy could be due to a large statistical fluctuation
or an unknown experimental problem. The ambiguity in the measured value of
lept
sin2 θeff could thus be larger than the nominal error, reported in Equation 5,
obtained from averaging all the existing determinations.
We have already observed that the experimental value of mW (with good
agreement between LEP and the Tevatron) is a bit high compared to the SM
prediction (see Figure 4). The value of mH indicated by mW is on the low side,
lept
just in the same interval as for sin2 θeff measured from leptonic asymmetries. It
is interesting that the new value of mt considerably relaxes the previous tension
lept
between the experimental values of mW and sin2 θeff measured from leptonic
asymmetries on one side and the lower limit on mH from direct searches on the
other side [17, 18]. This is also apparent from Figure 4.

80.5
M W [GeV]

80.45

MW word average

80.4

Mt=182.3 GeV

178 GeV
80.35
173.7 GeV

80.3

50 100 200 300 400 500

MH [GeV]

Figure 4:The world average for mW is compared with the SM prediction as a function of
mH (updated from [15]).

The main lesson of precision tests of the standard electroweak theory can
be summarised as follows. The couplings of quark and leptons to the weak gauge
bosons W± and Z are indeed precisely those prescribed by the gauge symmetry.
The accuracy of a few per-mille for these tests implies that, not only the tree
level, but also the structure of quantum corrections has been verified. To a lesser
172 Guido Altarelli

accuracy the triple gauge vertices γW+ W− and ZW+ W−�have also been found
in agreement with the specific prediction of the SU (2) U (1) gauge theory.
This means that it has been verified that the gauge symmetry is unbroken in
the vertices of the theory: the currents are indeed conserved. Yet there is
obvious evidence that the symmetry is otherwise badly broken in the masses.
Thus the currents are conserved but the spectrum of particle states is not at
all symmetric. This is a clear signal of spontaneous symmetry breaking. The
practical implementation of spontaneous symmetry breaking in a gauge theory
is via the Higgs mechanism. The Higgs sector of the SM is still very much
untested. What has been tested is the relation m2W = m2Z cos2 θW , modified by
computable radiative corrections. This relation means that the effective Higgs
(be it fundamental or composite) is indeed a weak isospin doublet. The Higgs
particle has not been found but in the SM its mass can well be larger than the
present direct lower limit mH >
∼ 114 GeV obtained from direct searches at LEP-
2. The radiative corrections computed in the SM when compared to the data
on precision electroweak tests lead to a clear indication for a light Higgs, not
too far from the present lower bound. No signal of new physics has been found.
However, to make a light Higgs natural in presence of quantum fluctuations
new physics should not be too far. This is encouraging for the LHC that should
experimentally clarify the problem of the electroweak symmetry breaking sector
and search for physics beyond the SM.

2 Outlook on Avenues beyond the Standard Model

Given the success of the SM why are we not satisfied with that theory?
Why not just find the Higgs particle, for completeness, and declare that particle
physics is closed? The reason is that there are both conceptual problems and
phenomenological indications for physics beyond the SM. On the conceptual
side the most obvious problems are that quantum gravity is not included in
the SM and the related hierarchy problem. Among the main phenomenological
hints for new physics we can list coupling unification, dark matter, neutrino
masses, baryogenesis and the cosmological vacuum energy.
The computed evolution with energy of the effective SM gauge couplings
clearly points towards the unification of the electro-weak and strong forces
(Grand Unified Theories: GUT’s) at scales of energy MGU T ∼ 1015 − 1016 GeV
which are close to the scale of quantum gravity, MP l ∼ 1019 GeV . One is led
to imagine a unified theory of all interactions also including gravity (at present
superstrings provide the best attempt at such a theory). Thus GUT’s and the
realm of quantum gravity set a very distant energy horizon that modern parti-
cle theory cannot ignore. Can the SM without new physics be valid up to such
The Electroweak Interactions in the Standard Model and beyond 173

large energies? This appears unlikely because the structure of the SM could
not naturally explain the relative√smallness of the weak scale of mass, set by
the Higgs mechanism at µ ∼ 1/ GF ∼ 250 GeV with GF being the Fermi
coupling constant. This so-called hierarchy problem is related to the presence
of fundamental scalar fields in the theory with quadratic mass divergences and
no protective extra symmetry at µ = 0. For fermion masses, first, the�diver-
gences are logarithmic and, second, they are forbidden by the SU (2) U (1)
gauge symmetry plus the fact that at m = 0 an additional symmetry, i.e. chiral
symmetry, is restored. Here, when talking of divergences, we are not worried of
actual infinities. The theory is renormalisable and finite once the dependence
on the cut off is absorbed in a redefinition of masses and couplings. Rather
the hierarchy problem is one of naturalness. We should see the cut off as a pa-
rameterization of our ignorance on the new physics that will modify the theory
at large energy scales. Then it is relevant to look at the dependence of phys-
ical quantities on the cut off and to demand that no unexplained enormously
accurate cancellations arise.
The hierarchy problem can be put in very practical terms: loop corrections
to the higgs mass squared are quadratic in Λ. The most pressing problem is
from the top loop. With m2h = m2bare + δm2h the top loop gives

3GF 2 2
δm2h|top ∼ √ mt Λ ∼ (0.3Λ)2 (10)
2π 2
If we demand that the correction does not exceed the light Higgs mass
indicated by the precision tests, Λ must be close, Λ ∼ o(1 T eV ). Similar
constraints arise from the quadratic Λ dependence of loops with gauge bosons
and scalars, which, however, lead to less pressing bounds. So the hierarchy
problem demands new physics to be very close (in particular the mechanism
that quenches the top loop). Actually, this new physics must be rather special,
because it must be very close, yet its effects are not clearly visible (the ”LEP
Paradox” [19]). Examples of proposed classes of solutions for the hierarchy
problem are:
Supersymmetry. In the limit of exact boson-fermion symmetry the
quadratic divergences of bosons cancel so that only log divergences remain.
However, exact SUSY is clearly unrealistic. For approximate SUSY (with soft
breaking terms), which is the basis for all practical models, Λ is replaced by the
splitting of SUSY multiplets, Λ ∼ mSU SY − mord . In particular, the top loop
is quenched by partial cancellation with s-top exchange.
Technicolor. The Higgs system is a condensate of new fermions. There
are no fundamental scalar Higgs sector, hence no quadratic devergences asso-
ciated to the µ2 mass in the scalar potential. This mechanism needs a very
174 Guido Altarelli

strong binding force, ΛT C ∼ 103 ΛQCD . It is difficult to arrange that such

nearby strong force is not showing up in precision tests. Hence this class of
models has been disfavoured by LEP, although some special class of models
have been devised aposteriori, like walking TC, top-color assisted TC etc (for
recent reviews, see, for example, [20]).
Large compactified extra dimensions. The idea is that MP L appears
very large, that is gravity seems very weak because we are fooled by hidden
extra dimensions so that the real gravity scale is reduced down to o(1 T eV ).
This possibility is very exciting in itself and it is really remarkable that it is
compatible with experiment.
”Little Higgs”models. In these models extra symmetries allow mh �= 0
only at two-loop level, so that Λ can be as large as o(10 T eV ) with the Higgs
within present bounds (the top loop is quenched by exchange of heavy vectorlike
new charge-2/3 quarks).
We now briefly comment in turn on these possibilities.
SUSY models are the most developed and most widely accepted. Many
theorists consider SUSY as established at the Planck scale MP l . So why not
to use it also at low energy to fix the hierarchy problem, if at all possible? It
is interesting that viable models exist. The necessary SUSY breaking can be
introduced through soft terms that do not spoil the good convergence prop-
erties of the theory. Precisely those terms arise from supergravity when it is
spontaneoulsly broken in a hidden sector. This is the case of the MSSM [21].
Of course, minimality is only a simplicity assumption that could possibly be re-
laxed. The MSSM is a completely specified, consistent and computable theory
which is compatible with all precision electroweak tests. In this most traditional
approach SUSY is broken � in a hidden sector and the scale of SUSY breaking is
−1/2
very large of order Λ ∼ GF MP l . But since the hidden sector only commu-
nicates with the visible sector through gravitational interactions the splitting of
the SUSY multiplets is much smaller, in the TeV energy domain, and the Gold-
stino is practically decoupled. But alternative mechanisms of SUSY breaking
are also being considered. In one alternative scenario [22] the (not so much)
hidden sector is connected to the visible one by ordinary gauge interactions.
As these are much stronger than the gravitational interactions, Λ can be much
smaller, as low as 10-100 TeV. It follows that the Goldstino is very light in these
models (with mass of order or below 1 eV typically) and is the lightest, stable
SUSY particle, but its couplings are observably large. The radiative decay of
the lightest neutralino into the Goldstino leads to detectable photons. The sig-
nature of photons comes out naturally in this SUSY breaking pattern: with
respect to the MSSM, in the gauge mediated model there are typically more
photons and less missing energy. The main appeal of gauge mediated models
The Electroweak Interactions in the Standard Model and beyond 175

is a better protection against flavour changing neutral currents but naturality

problems tend to increase. As another possibility it has been pointed out that
there are pure gravity contributions to soft masses that arise from gravity theory
anomalies[23]. In the assumption that these terms are dominant the associated
spectrum and phenomenology have been studied. In this case gaugino masses
are proportional to gauge coupling beta functions, so that the gluino is much
heavier than the electroweak gauginos, and the wino is most often the lightest
SUSY particle.
What is really unique to SUSY with respect to all other extensions of the
SM listed above is that the MSSM or similar models are well defined and com-
putable up to MP l and, moreover, are not only compatible but actually quan-
titatively supported by coupling unification and GUT’s. At present the most
direct phenomenological evidence in favour of supersymmetry is obtained from
the unification of couplings in GUTs. Precise LEP data on αs (mZ ) and sin2 θW
show that standard one-scale GUTs fail in predicting sin2 θW given αs (mZ )
(and α(mZ )) while SUSY GUTs are in agreement with the present, very pre-
cise, experimental results. If one starts from the known values of sin2 θW and
α(mZ ), one finds [24] for αs (mZ ) the results: αs (mZ ) = 0.073 ± 0.002 for Stan-
dard GUTs and αs (mZ ) = 0.129 ± 0.010 for SUSY GUTs to be compared with
the world average experimental value αs (mZ ) = 0.119 ± 0.003. Another great
asset of SUSY GUT’s is that proton decay is much slowed down with respect
to the non SUSY case. First, the unification mass MGU T ∼ few 1016 GeV,
in typical SUSY GUT’s, is about 20-30 times larger than for ordinary GUT’s.
This makes p decay via gauge boson exchange negligible and the main decay
amplitude arises from dim-5 operators with higgsino exchange, leading to a
rate close but still compatible with existing bounds (see, for example,[25]). It
is also important that SUSY provides an excellent dark matter candidate, the
neutralino. We finally recall that the range of neutrino masses as indicated by
oscillation experiments, when interpreted in the see-saw mechanism, point to
MGU T and give additional support to GUTs [26].
In spite of all these virtues it is true that the lack of SUSY signals at LEP
and the lower limit on mH pose problems for the MSSM. The lightest Higgs par-
ticle is predicted in the MSSM to be below mh < ∼ 135 GeV (the recent increase
of mt helps in this respect). The limit on the SM Higgs mH > ∼ 114 GeV consid-
erably restricts the available parameter space of the MSSM requiring relatively
large tan β (tan β > ∼ 2 − 3: at tree level mh = mZ cos 2β) and rather heavy s-
2 2 2

top (the loop corrections increase with log m̃t ). Stringent naturality constraints
2

also follow from imposing that the electroweak symmetry breaking occurs at the
right place: in SUSY models the breaking is induced by the running of the Hu
mass starting from a common scalar mass m0 at MGU T . The squared Z mass
176 Guido Altarelli

m2Z can be expressed as a linear combination of the SUSY parameters m20 , m21/2 ,
A2t , µ2 ,... with known coefficients. Barring cancellations that need fine tuning,
the SUSY parameters, hence the SUSY s-partners cannot be too heavy. The
LEP limits, in particular the chargino lower bound mχ+ > ∼ 100 GeV , are suffi-
cient to eliminate an important region of the parameter space, depending on the
amount of allowed fine tuning. For example, models based on gaugino univer-
sality at the GUT scale are discarded unless a fine tuning by at least a factor of
20 is not allowed. Without gaugino universality [27] the strongest limit remains
on the gluino mass: m2Z ∼ 0.7 m2gluino + . . . which is still compatible with the
present limit mgluino > ∼ 200 GeV .
The non discovery of SUSY at LEP has given further impulse to the quest
for new ideas on physics beyond the SM. Large extra dimensions [28] and ”little
Higgs” [29] models are the most interesting new directions in model building.
Large extra dimension models propose to solve the hierarchy problem by bring-
ing gravity down from MP l to m ∼ o(1 T eV ) where m is the string scale.
Inspired by string theory one assumes that some compactified extra dimensions
are sufficiently large and that the SM fields are confined to a 4-dimensional
brane immersed in a d-dimensional bulk while gravity, which feels the whole
geometry, propagates in the bulk. We know that the Planck mass is large be-
cause gravity is weak: in fact GN ∼ 1/MP2 l , where GN is Newton constant.
The idea is that gravity appears so weak because a lot of lines of force escape
in extra dimensions. Assume you have n = d − 4 extra dimensions with com-
pactification radius R. For large distances, r >> R, the ordinary Newton law
applies for gravity: in natural units F ∼ GN /r2 ∼ 1/(MP2 l r2 ). At short dis-
tances, r < ∼ R, the flow of lines of force in extra dimensions modifies Gauss
law and F −1 ∼ m2 (mr)d−4 r2 . By matching the two formulas at r = R one
obtains (MP l /m)2 = (Rm)d−4 . For m ∼ 1 T eV and n = d − 4 one finds that
n = 1 is excluded (R ∼ 1015 cm), for n = 2 R is at the edge of present bounds
R ∼ 1 mm, while for n = 4, 6, R ∼ 10−9 , 10−12 cm. In all these models a
generic feature is the occurrence of Kaluza-Klein (KK) modes. Compactified di-
mensions with periodic boundary conditions, as for quantization in a box, imply
a discrete spectrum with momentum p = n/R and mass squared m2 = n2 /R2 .
There are many versions of these models. The SM brane can itself have a thick-
ness r with r <∼ 10−17 cm or 1/r >∼ 1 T eV , because we know that quarks
and leptons are pointlike down to these distances, while for gravity there is no
experimental counter-evidence down to R <∼ 0.1 mm or 1/R >∼ 10−3 eV .
In case of a thickness for the SM brane there would be KK recurrences for SM
fields, like Wn , Zn and so on in the T eV region and above. There are models
with factorized metric (ds2 = ηµν dxµ dxν + hij (y)dy i dy j , where y (i,j) denotes
the extra dimension coordinates (and indices), or models with warped metric
The Electroweak Interactions in the Standard Model and beyond 177

(ds2 = e − 2kR|φ|ηµν dxµ dxν − R2 φ2 [30]. In any case there are the towers of
KK recurrences of the graviton. They are gravitationally coupled but there are
a lot of them that sizably couple, so that the net result is a modification of
cross-sections and the presence of missing energy.
Large extra dimensions provide a very exciting scenario [31]. Already it is
remarkable that this possibility is compatible with experiment. However, there
are a number of criticisms that can be brought up. First, the hierarchy problem
is more translated in new terms rather than solved. In fact, the basic relation
Rm = (MP l /m)2/n shows that Rm, which one would apriori expect to be 0(1),
is instead ad hoc related to the large ratio MP l /m. In this respect the Randall-
Sundrum variety is more appealing because the hierarchy suppression mW /MP l
could arise from the warping factor e−2kR|φ| , with not too large values of kR.
Also it is not clear how extra dimensions can by themselves solve the LEP
paradox (the large top loop corrections should be controlled by the opening of
the new dimensions and the onset of gravity): since mH is light Λ ∼ 1/R must
be relatively close. But precision tests put very strong limits on Λ. In fact, in
typical models of this class, there is no mechanism to sufficiently quench the
corrections. No simple, realistic model has yet emerged as a benchmark. But
it is attractive to imagine that large extra dimensions could be a part of the
truth, perhaps coupled with some additional symmetry or even SUSY.
In the extra dimension general context an interesting direction of devel-
opment is the study of symmetry breaking by orbifolding and/or boundary
conditions. These are models where a larger gauge symmetry (with or without
SUSY) holds in the bulk. The symmetry is reduced in the 4 dimensional brane,
where the physics that we observe is located, as an effect of symmetry breaking
induced geometrically by suitable boundary conditions. There are models where
SUSY, valid in n > 4 dimensions is broken by boundary conditions [32], in par-
ticular the model of Ref.[33], where the mass of the Higgs is computable and can
be extimated with good accuracy. Then there are ”Higgsless models” where it
is the SM electroweak gauge symmetry which is broken at the boundaries [34].
Or models where the Higgs is the 5th component of a gauge boson of an ex-
tended symmetry valid in n > 4 [35]. In general all these alternative models
for the Higgs mechanism face severe problems and constraints from electroweak
precision tests [36]. At the GUT scale, symmetry breaking by orbifolding can
be applied to obtain a reformulation of SUSY GUT’s where many problematic
features of ordinary GUT’s (e.g. a baroque Higgs sector, the doublet-triplet
splitting problem, fast proton decay etc) are improved [31], [37].
In ”little Higgs” models, the symmetry of the SM is extended to a � suitable
global group G that also � contains some gauge � enlargement of SU (2) U (1),
for example G ⊃ [SU (2) U (1)]2 ⊃ SU (2) U (1). The Higgs particle is a
178 Guido Altarelli

pseudo-Goldstone boson of G that only takes mass at 2-loop level, because two
distinct symmetries must be simultaneously broken for it to take mass, which
requires the action of two different couplings in the same diagram. Then in the
relation between δm2h and Λ2 there is an additional coupling and an additional
loop factor that allow for a bigger separation between the Higgs mass and the
cut-off. Typically, in these models one has one or more Higgs doublets at
mh ∼ 0.2 T eV , and a cut-off at Λ ∼ 10 T eV . The top loop quadratic cut-off
dependence is partially canceled, in a natural way guaranteed by the symmetries
of the model, by a new coloured, charge-2/3, vectorial quark χ of mass around
1 T eV (a fermion not a scalar like the s-top of SUSY models). Certainly these
models involve a remarkable level of group theoretic virtuosity. However, in the
simplest versions one is faced with problems with precision tests of the SM [38].
Even with vectorlike new fermions large corrections to the epsilon parameters
arise from exchanges of the new gauge bosons W � and Z � (due to lack of custodial
SU (2) symmetry). In order to comply with these constraints the cut-off must
be pushed towards large energy and the amount of fine tuning needed to keep
the Higgs light is still quite large. Probably these bad features can be fixed by
some suitable complication of the model (see for example, [39]). But, in my
opinion, the real limit of this approach is that it only offers a postponement
of the main problem by a few TeV, paid by a complete loss of predictivity at
higher energies. In particular all connections to GUT’s are lost.
Finally, we stress the importance of the cosmological constant or vacuum
energy problem [40]. The exciting recent results on cosmological parameters,
culminating with the precise WMAP measurements [41], have shown that vac-
uum energy accounts for about 2/3 of the critical density: ΩΛ ∼ 0.65, Trans-
lated into familiar units this means for the energy density ρΛ ∼ (2 10−3 eV )4
1/4
or (0.1 mm)−4 . It is really interesting (and not at all understood) that ρΛ ∼
Λ2EW /MP l (close to the range of neutrino masses). It is well known that in field
theory we expect ρΛ ∼ Λ4cutof f . If the cut off is set at MP l or even at 0(1 T eV )
there would an enormous mismatch. In exact SUSY ρΛ = 0, but SUSY is bro-
1/4
ken and in presence of breaking ρΛ is in general not smaller than the typical
SUSY multiplet splitting. Another closely related problem is ”why now?”: the
time evolution of the matter or radiation density is quite rapid, while the den-
sity for a cosmological constant term would be flat. If so, them how comes that
precisely now the two density sources are comparable? This suggests that the
vacuum energy is not a cosmological constant term, buth rather the vacuum ex-
pectation value of some field (quintessence) and that the ”why now?” problem
is solved by some dynamical mechanism.
Clearly the cosmological constant problem poses a big question mark on
the relevance of naturalness as a relevant criterion also for the hierarchy prob-
The Electroweak Interactions in the Standard Model and beyond 179

lem: how we can trust that we need new physics close to the weak scale out
of naturalness if we have no idea on the solution of the cosmological constant
huge naturalness problem? The common answer is that the hierarchy problem
is formulated within a well defined field theory context while the cosmological
constant problem makes only sense within a theory of quantum gravity, that
there could be modification of gravity at the sub-eV scale, that the vacuum en-
ergy could flow in extra dimensions or in different Universes and so on. At the
other extreme is the possibility that naturalness is misleading. Weinberg [42]
has pointed out that the observed order of magnitude of Λ can be successfully
reproduced as the one necessary to allow galaxy formation in the Universe. In a
scenario where new Universes are continuously produced we might be living in a
very special one (largely fine-tuned) but the only one to allow the development
of an observer. One might then argue that the same could in principle be true
also for the Higgs sector. Recently it was suggested [43] to abandon the no-
fine-tuning assumption for the electro-weak theory, but require correct coupling
unification, presence of dark matter with weak couplings and a single scale of
evolution from the EW to the GUT scale. A ”split SUSY” model arises as a
solution with a fine-tuned light Higgs and all SUSY particles heavy except for
gauginos, higgsinos and neutralinos, protected by chiral symmetry. Or we can
have a two-scale non-SUSY GUT with axions as dark matter. In conclusion, it
is clear that naturalness can be a good heuristic principle but you cannot prove
its necessity.

3 Summary and Conclusion

Supersymmetry remains the standard way beyond the SM. What is unique to
SUSY, beyond leading to a set of consistent and completely formulated models,
as, for example, the MSSM, is that this theory can potentially work up to the
GUT energy scale. In this respect it is the most ambitious model because it
describes a computable framework that could be valid all the way up to the
vicinity of the Planck mass. The SUSY models are perfectly compatible with
GUT’s and are actually quantitatively supported by coupling unification and
also by what we have recently learned on neutrino masses. All other main ideas
for going beyond the SM do not share this synthesis with GUT’s. The SUSY
way is testable, for example at the LHC, and the issue of its validity will be
decided by experiment. It is true that we could have expected the first signals
of SUSY already at LEP, based on naturality arguments applied to the most
minimal models (for example, those with gaugino universality at asymptotic
scales). The absence of signals has stimulated the development of new ideas
like those of large extra dimensions and ”little Higgs” models. These ideas are
180 Guido Altarelli

very interesting and provide an important reference for the preparation of LHC
experiments. Models along these new ideas are not so completely formulated
and studied as for SUSY and no well defined and realistic baseline has so far
emerged. But it is well possible that they might represent at least a part of the
truth and it is very important to continue the exploration of new ways beyond
the SM.
I would like to express my gratitude to the Organisers of the ICFA ’03
School for their invitation and their magnificent hospitality in Itacuruca. In
particular I would like to thank Bernard Marechal.

References

[1] The LEP EW Working Group, hep-ex/0212036.

[2] G. Altarelli and M. Grunewald, hep-ph/0404165.

[3] The ALEPH, DELPHI, L3, OPAL, SLD Collaborations and the LEP Elec-
troweak Working Group, A Combination of Preliminary Electroweak Mea-
surements and Constraints on the Standard Model, hep-ex/0312023, and
references therein.

[4] The Particle Data Group, Phys. Rev. D66 (2002) 1.

[5] The CDF Collaboration, the DØ Collaboration, and the Tevatron Elec-
troweak Working Group, Combination of CDF and DØ Results on the
Top-Quark Mass, hep-ex/0404010.

[6] H. Burkhardt, B. Pietrzyk, Update of the Hadronic Contribution to the

QED Vacuum Polarization, Phys. Lett. B 513 (2001) 46.

[7] Th. Appelquist, J. Carazzone, Infrared Singularities and Massive Fields,

Phys. Rev. D11 (1975) 2856.

[8] The ALEPH, DELPHI, L3 and OPAL Collaborations, and the LEP Work-
ing Group for Higgs Boson Searches, Search for the Standard Model Higgs
Boson at LEP, hep-ex/0306033, Phys. Lett. B565 (2003) 61–75.

[9] Th. Hambye, K. Riesselmann, Matching Conditions and Higgs Mass Upper
Bounds Revisited, hep-ph/9610272, Phys. Rev. D55 (1997) 7255–7262.

[10] G. Altarelli, R. Barbieri, F. Caravaglios, Electroweak Precision Tests: A

Concise Review, hep-ph/9712368, Int. Jour. Mod. Phys.A13 (1998) 1031–
1058.
The Electroweak Interactions in the Standard Model and beyond 181

[11] The NuTeV Collaboration, G.P. Zeller et al., Phys. Rev. Lett. 88 (2002)
091802.
[12] M. Yu. Kuchiev and V. V. Flambaum, Radiative Corrections to Parity
Non-Conservation in Atoms, hep-ph/0305053.
[13] The SLAC E158 Collaboration, P.L. Anthony et al., Observation of Parity
Non-Conservation in Moller scattering, hep-ex/0312035, A New Measure-
ment of the Weak Mixing Angle, hep-ex/0403010. We have added 0.0003 to
the value of sin2 θ(mZ ) quoted by E158 in order to convert from the MSbar
scheme to the effective electroweak mixing angle [4].
[14] S. Davidson et al, hep-ph/0112302.
[15] P. Gambino, The Top Priority: Precision Electroweak Physics from Low-
Energy to High-Energy, hep-ph/0311257.
[16] D. Choudhury, T.M.P. Tait, C.E.M. Wagner, Beautiful Mirrors and Pre-
cision Electroweak Data, hep-ph/0109097, Phys. Rev. D65 (2002) 053002.
[17] M. S. Chanowitz, Electroweak Data and the Higgs Boson Mass: A Case
for New Physics, hep-ph/0207123, Phys. Rev. D66 (2002) 073002.
[18] G. Altarelli, F. Caravaglios, G.F. Giudice, P. Gambino, G. Ridolfi, Indica-
tion for Light Sneutrinos and Gauginos from Precision Electroweak Data,
hep-ph 0106029, JHEP 0106:018, 2001.
[19] R. Barbieri and A. Strumia, hep-ph/0007265.
[20] K. Lane, hep-ph/0202255,
R. S. Chivukula, hep-ph/0011264.
[21] For a recent introduction see, for example, S. P. Martin, hep-ph/9709356.
[22] M. Dine and A. E. Nelson, Phys. Rev. D48 (1993) 1277;
M. Dine, A. E. Nelson and Y. Shirman, Phys. Rev. D51 (1995) 1362;
G. F. Giudice and R. Rattazzi, Phys. Rept. 322 (1999) 419.
[23] L. Randall and R. Sundrum, Nucl. Phys. B557 (1999) 79;
G.F. Giudice et al, JHEP 9812 (1998) 027.
[24] P. Langacker and N. Polonsky, Phys. Rev. D52 (1995) 3081.
[25] G. Altarelli, F. Feruglio and I. Masina, JHEP 0011:040,2000.
[26] For a review see, for example, G. Altarelli and F. Feruglio, hep-ph/0405048.
182 Guido Altarelli

[27] G. Kane et al, Phys.Lett. B551 (2003) 146.

[28] For a review and a list of refs., see, for example, J. Hewett and M. Spiropulu,
hep-ph/0205196.
[29] For a review and a list of refs., see, for example, M. Schmaltz, hep-
ph/0210415.
[30] L. Randall and R. Sundrum, Phys. Rev. Lett. 83 (1999) 3370, 83 (1999)
4690.
[31] For a review, see F. Feruglio, hep-ph/0401033.
[32] I. Antoniadis, C. Munoz and M. Quiros, Nucl. Phys. B397 (1993) 515;
A. Pomarol and M. Quiros, Phys. Lett. B438 (1998) 255.
[33] R. Barbieri, L. Hall and Y. Nomura, Nucl. Phys. B624 (2002) 63;
R.Barbieri, G. Marandella and M. Papucci, hep-ph/0205280, hep-
ph/0305044, and refs therein.
[34] see for example, C. Csaki et al, hep-ph/0305237 , hep-ph/0308038, hep-
ph/0310355; S. Gabriel, S. Nandi and G. Seidl, hep-ph/0406020 and refs
therein.
[35] see for example, C. A. Scrucca, M. Serone and L. Silvestrini, hep-
ph/0304220 and refs therein.
[36] R. Barbieri, A. Pomarol and R. Rattazzi, hep-ph/0310285.
[37] Y. Kawamura, Progr. Theor. Phys. 105 (2001) 999.
[38] J. L. Hewett, F. J. Petriello, T. G. Rizzo, hep-ph/0211218;
C. Csaki et al, hep-ph/0211124, hep-ph/0303236. H-C. Cheng and I. Low,
hep-ph/0405243.
[39] H-C. Cheng and I. Low, hep-ph/0405243.
[40] For orientation, see, for example, M. Turner, astro-ph/0207297.
[41] The WMAP Collaboration, D. N. Spergel et al, astro-ph/0302209.
[42] S. Weinberg, Phys. Rev. Lett. 59 (1987) 2607.
[43] N.Arkani-Hamed and S. Dimopoulos, hep-th/0405159; G. Giudice and A.
Romanino, hep-ph/0406088.
 Near beam detectors∗
Hélio da Motta
Centro Brasileiro de Pesquisas Fı́sicas – CBPF/CLAFEX
Rua Dr. Xavier Sigaud 150
22280-190 Rio de Janeiro – RJ -Brazil
E-mail: [email protected]

ABSTRACT

There are instances of high energy particle interactions where one, or both, of
the interacting particles survives the proccess experiencing only a slight scatter-
ing that deviates it from the original beam. The identification of the particles
scattered at small angles is of uppermost importance in the study of such events.
Typical 4π detectors usually miss these particles. We describe here a series of
detectors specially built to operate very close to the beam enlarging the capa-
bility of the experiment to study the physics of this kind of event.

1 Introduction

In high energy physics, diffraction is a proccess in which no quantum num-

ber is exchanged between interacting particles which remain intact. Diffraction
encompasses events in which one or both incoming particles undergo diffractive
dissociation with any surviving particle having a small angle with respect to the
beam axis. It can account for 40% of the inclusive cross section of a process.
The interest in diffractive scattering has been growing since the obser-
vation of diffractive production of jets by the UA8 collaboration[1, 2]. The
experimental difficulty in observing these events is that the scattered proton
tends to remain in the beam pipe and can not be detected using typical collider
central detector as DØ and CDF. It is necessary to add special forward particle
detectors to the central assembly, close to the beam and at large distances from
the collision point.
Although rapidity gap (absence of particles in a region of the detector)
signatures can also be used to tag diffractive events, only the addition of a
forward particle detector allow access to the full kinematics of the scattered
particle.
∗
ICFA SCHOOL – Itacuruçá – Rio de Janeiro, Brazil – 8 – 20 December 2003
184 Hélio da Motta

By detecting the scattered p or p̄ one can measure its momentum |� p| and

derive the variables: xp = |� pbeam |; t = (pbeam − p)2 ; ξ = 1 − xp where t is
p|/|�
the four-momentum transfer of the p (or p̄)and xp is the fractional longitudinal
momentum of the scattered particle.

2 Near the beam

The detection of the particles scattered at small angles requires the posi-
tioning of special detectors very close to the beam, usually inside the accelerator
beampipe. That must be done with consideration to the very critical conditions
existing in such environment, specially the ultra high vacuum (UHV) that may
be present in the area. The detectors may have to be moved close to beam or
removed to clear the area during accelerator insertion periods. Its location must
be known within some precision and accuracy and a high degree of repeatability
must be achieved.
Accelerator vacuum may reach up to 10−10 Torr, or even more, and any
apparatus built to operate in such environment must not disturd this vacuum.
We can use a simple description to help understanding the meaning of such
vacuum.
Consider a cube whose sides are 10.0 cm long (a 1 L volume). Imagine the
inside of this cube completely void of particles. That would mean a 0.0 Torr
vacuum. If we cover the cube surface with a single layer of N2 molecules we will
need 6.0×1017 N2 molecules. Throwing all these molecules into the cube volume
the vacuum will correspond to 1.7 × 10−2 Torr. This is 8 orders of magnitude
bigger than 10−10 Torr. To achieve UHV, we must remove N2 molecules from
inside the cube, leaving only 1 out of every 100, 000, 000 molecules in there !
This is not simple to do. The tinyiest impurity on the material surface
would make our efforts vain. Even when completely clean of impurities, the
material would degass (liberating particles that are embeded in the material,
like water, nitrogen). This degassing will also ruin the attempts to achieve
UHV.
A UHV chamber is built following a series of steps, from which we shall
mention:

• Use of 360L stainless steel.

• TIG (Tungsten Inert Gas) welding (a proccess where a cloud of inert gas
is set around the welding point). Welding is made in the interior of the
chamber, so there are no air pockets left inside.

• Cleaning of all parts with demineralized water and neutral detergent.

Near beam detectors 185

• Baking the parts (this speeds up degassing, cleaning up the parts)

• Once the chamber is finished, it is sealed, and a ion pump is attached to

it.

• Vacuum is done by means of a conventional pump. Then, the chamber

is baked and the ion pump turned on. During the baking stage, that
may last a few days, the vacuum increases until UHV. Baking is then
interruped but the ion pump is left on.

Another issue to be considered when putting a detector close to the beam

is the need to have a precise knowledge of its position and the ability to move it
precisely and accuretely with a high level of repeatibility. All those are required
if one does not want to risk the beam and also to have a reliable measurement
of the quantities described above. These mechanical conditions demand high
precision and reliable mechanical devices.
The area close to the beam may also present high levels of radiation, in
which case all detectors and components must be radiation hard.
All these conditions make it a very complex task the building of any system
designed to operate close to the beam. We describe now a set of detectors that
have been built to such use.

3 Roman Pots

Roman pots are stainless containers that allow the position detector to
function outside of the machine ultra high vacuum, but close to the beam. The
scattered particle traverses a thin steel window at the entrance of the pot, goes
through the detector that rests in the pot and exit the pot through another thin
window. The pots are remotely controlled and can be moved close to the beam
during stable conditions.
The first ever built Roman pot[3] was used in 1970-1972 at the ISR by
the CERN-Rome group. It did not have a thin window and housed a small
hodoscope of scintillating counters. Figure1 shows the first Roman pot and a
later Roman pot that was used by UA4 at SPS. The UA4 Roman pot section
facing the beam is concave in shape, allowing a closer aproach to the beam. A
thin window is used to reduce the material between the beam and the detector
inside the pot.
More recently, CDF installed a set of three Roman pots with a scintillating
fiber detector, as viewed in Figure2. When hit by the scattered particle, the fiber
scintillates and the light is guided to a multianode photomultiplier that collects
the light. Figure2 also shows details of the CDF scintillating fiber detector.
186 Hélio da Motta

Figure 1: First Roman pot (left). UA4 Roman pot (right). The concave bottom of the UA4
Roman pot allows a closer aproach to the beam and the thin window reduces the material
the particle has to traverses before hiting the detector that rests inside the pot.

4 The Forward Proton Detector

The DØ collaboration has built a very comprehensive system of Roman

pot detectors that is know as Forward Proton Detector (FPD)[4]. It consists of
a series of momentum spectrometers that make use of accelerator magnets in
conjunction with position detectors along the beam line in order to determine
the kinematic variables (t and ξ) of the scattered p and p̄.
The detector itself consists of an array of square scintillating fibers or-
ganized in three different directions (x,u and v). Clear fibers transport the
light signal to multi anode photomultipliers (MAPMT), whose outputs feed an
electronic shaping and amplifying system and are read out by electronics. The
detectors are placed in Roman Pot structures.
The FPD has 18 Roman pots arranged in 6 stainless steel chambers called
castles located at different distances with respect to the DØ interaction point
and in locations that do not interfere with any accelerator element. The exper-
imental arrangement of the FPD is shown in Figure3. Four castles are located
downstream of the low beta quadrupole magnets on each side of the colliding
point: two on the p side (P1 and P2)and two on the p̄ side (A1 and A2). Each of
these quadrupole castles contains 4 Roman pots arranged to cover most of the
area around the beam. Two castles (D2 and D1) are located on the outgoing
p̄ side after the dipole magnet. Each of these dipole castles contain only one
Roman pot.
Figures 4,5 and 6 show the driving system that makes it possible for the pot
Near beam detectors 187

Roman Pot Arrangement

Top View -56m distance from interaction point
266.51cm
98.5 98.5

7.62
20

30

BO
Bellows
Recoil Track To MCPMT Fiber
Tracker
Detail
To MCPMT

CDF

P X
Y

Figure 2: CDF Roman pot system. Schematics of the detector is shown. It consists of an
array of scintillating fibers oriented in the X and Y direction. The scattered particle leaves a
signal in a pair of XY channel that is used to determine its position.

to move perpendicularly to the beam in a very precise, safe and accurate way.
The system is operated by a step motor and a set of reduction gears allows pot
motion with a precision of aproximately 5 µm. A set of cylindrical and conical
bearings allows adjustment of the pot alignment and a linear variable differential
transducer (LVDT) monitors the pot position. A steel bellows guaranties the
movement of the pot whithout affecting the vacuum.
The FPD driving assembly is basically a cylindrical tube threaded on
its outer surface that is actuated by a worm-gear system. The system has
a reduction of 120 times which accounts both for a precision of about 5µm
movement of the detector and for compensation for the inward force of about
2, 000 N resulting from the vacuum inside the castle. This arrangement makes
possible the use of small low torque motors and makes unecessary the use of any
vacuum compensation system. An exploded view of the driving system is shown
in Figure5. The compacteness of the whole piece makes it easy to employ in as
many units as needed. Details of the driving assembly structure is presented in
Figure6. The moving parts get a thin layer of a molibidenium lubricant.
188 Hélio da Motta

DO
Interaction
P
A2U A1U Region P1U P2U
P A2O A10 P10 P20 P

D21 D1I A2I A1I P1I P2I

A2D A1D P1D P2D
y
x P
z
-59m -58m -31m -23m 0m 23m 31m

Figure 3: FPD layout at DØ experimental area. Quadrupole castles are named P or A when
placed on the p side or the p̄, respectively.

Figure 4: FPD driving assembly.

Near beam detectors 189

Figure 5: FPD driving assembly (exploded view).

279.5
145.6

EXTENDED POSITION VIEW

Figure 6: FPD driving assemby (cut view).

190 Hélio da Motta

Each position detector is made of 0.8 mm thick double-clad square scintil-

lating fibers (Bycron BCF10) bundled in groups of four forming a scintillating
structure measuring 0.8 mm × 3.2 mm. One end of the detector element is
aluminized (aproximately 3 µm thick layer) to increase the light yield and the
other end of each scintillating fiber is spliced to a double-clad clear fiber of
square cross section (Bycron BCF98) with the same dimensions† . The scat-
tered p or p̄ goes through 3.2 mm of scintillating material yielding aproximately
10 photoelectrons. The 4 clear fibers then take the light of one element to a
single channel of a Hamamatsu H6568 16-channel multi anode photomultiplier
(MAPMT).
The use of clear fibers, spliced to the scintillating fibers that constitute
the active part of the detector, is conceived as a way to reduce the effect of
halo background in the fibers, the optical cross talk in the fibers, and to mini-
mize light attenuation since clear fibers have an attenuation length longer than
scintillating fibers.
Each detector consists of six planes in three views (U , V and X) in order
to minimize ghost hit problems and to reduce the reconstruction ambiguities.
Each view is made of two planes (U −U � ,V −V � and X −X � ), the unprimed layer
being offset by 2/3 of a fiber with respect to primed ones. This arrangement
yields a theoretical detector point resolution of 80 µm. U and V planes are
oriented at ±45◦ with respect to the horizontal bottom of the detector, while
the X plane is at 90◦ . There are 20 channels in each layer of the U and V
planes and 16 channels in each of the X layers. There are 112 channels (each
with four fibers) per detector, giving a total of 2016 channels in the 18 Roman
pots. Each detector needs 7 MAPMT and also includes a trigger scintillator
read out by a fast photomultiplier (Phillips XP2282). Figure7 shows the fiber
arrangement and the way they are connected to the MAPMT. Figure8 shows
the FPD position detector.

The use of square fibers gives an increase of about 20 % in light output

†

compared to round fibers.

Near beam detectors 191

1234
MAPMT
Sci Fibers

4 mm

4.5 mm
Clear Fibers

30 mm
1 2
3 4
4 mm

2 mm

Fibers

Figure 7: FPD detector fibers and the MAPMT.

5 Near the beam without a pot

Fermilab experiments E710 and E811 used a detector made of a bundle

of scintillating fibers placed directly into the accelerator vacuum. No pot or
any other structure was used and so there was no residual dead space between
the detector edge and the beam. The bundle of scintillating fibers was placed
just above the beam, with the fibers aligned along the beamline. A scattered
particle hits one of the fibers originating a light signal that os guided to a CCD
camera attached to the other end of the fibers bundle (a photomultiplier could
also be used). This would give information on the scattered particle position.
Figure9 shows a scheme of this detector.
A potless device is now under development possibly to be used at the LHC.
This device, known as microstation and whose schematic is shown in Figure10
will use a silicon detector arranged in two moving structures placed around the
beam. Small actuators will move the detectors to place them in the desired
position around the beam. The interior of the chamber is kept under vacuum.
192 Hélio da Motta

Figure 8: FPD position detector.

II

II

PM

SCINTILLANTING
COUNTER

SCINTILLANTING FIBRES BUNDLE

Figure 9: E710/E811 near beam detector.

Near beam detectors 193

Figure 10: Microstation.

6 Conclusions

Near beam detectors have been used for over 30 years in different acceler-
ators with different beams. They allow the measurement of several proccesses
(like elastic scattering and diffraction dissociation at high energy) that can not
be as well studied without these detectors.
They are as more important as higher the energy. Consequently, they
shall be of great interest in the new LHC era that is to start in a few years. The
technology to design and built such detectors is already of common knowledge
and has been tested successfully. They are safe to operate and can become an
integrated part of the accelerator structure.

7 Acknowledgments

The author thanks the Organizers of the ICFA School for the invitation
to present this talk in Itacuruçá, and thanks Marcelo Juni (LNLS) and Regis
Neuenschwander (LNLS) for their very useful information on vacuum and me-
chanics. Special thanks goes to Alberto Santoro (UERJ) whose support and
determination has made the FPD possible.

References

[1] R. Bonino et al.(UA8 Collaboration), Phys. Lett. B 211 (1988) 239.

194 Hélio da Motta

[2] A. Brandt et al.(UA8 Collaboration), Phys. Lett. B 297 (1992) 417.

[3] U. Amaldi et al., Phys. Lett. B 43 (1973) 231.

[4] A. Brandt et al., FERMILAB-Pub-97/377
[5] J. Barreto and J. Montanha, DØ Note 3790 (2000)
[6] G. Alves et al., DØ Note 4054 (2002)
 High Energy Physics GRID
Alberto Santoro
DFNAE-IF-UERJ
Rua São Francisco Xavier,524
20550-013 Rio de Janeiro – RJ – Brazil
E-mail: [email protected]

ABSTRACT
Computing is one of the technologies associated to High Energy Physics (HEP).
The needs of the next generation CERN HEP experiments push the development
of new computing architectures. The scale of data will be orders of magnitude
larger than the one we currently have. In this talk we will present the current
status of GRID computing development.

1 Introduction

The Grid will permeate all science, allowing groups around the world to
collaborate objectively and giving different regions the oportunity to take part
in the science frontier. This technology will be present in our professional life
for a long time. New chips and computer memmories, new hard disks, the Raid
technology, are developments that help to advance the proposed computing
architecture for HEP even faster.

1.1 Origin
HEP physicists have always been involved with technology and, since
the begining, computing has been present as one of the main tools. Computing
is present in the accelerator, in the proposal, during the development, in each
detector project, in the data acquisition, in the analysis. Putting it simple,
computing is in all parts of a high energy physics experiment. Each experiment
builds its own instrumentation and new tools according to its needs. Grid
is a natural step forward. Frontiers of science imply frontier of technologies.
The World Wide Web was developed by Tim Bernes Lee team at CERN as
a need of HEP because collaborations were becoming more participative and
comunication was a priority.
The arrival of New Technologies (CPU, storage, networks, new lan-
guages) will be very useful in the very high energy collisions (14 TeV) environ-
ment with a much larger amount of events (tens of PetaBytes).
196 Alberto Santoro

The four experiments (ATLAS, ALICE, CMS, LHCb) at LHC (Large

Hadron Collider) at CERN will collect, in the first year, an amount about 20
Petabytes of data. Each experiment will involve an average of 1000 physicists.
This is an important parameter since all collaborators must have access to the
data.
New fast eletronics and/or photonics are being developed for use in the
triggers that will have to deal with higher rates.
A micro-society must be created and organized for the sake of efficiency.
This means Virtual Labs and Virtual Organizations in addition to the real
HEP.

1.2 Definitions
Before going ahead, I would like to give a few useful definitions acronyms
frequently used in the field.

Petabyte: 1 Petabyte = 103 Terabyte = 106 Gigabyte = 109 Megabyte

= 1012 = Kilobyte = 1015 Bytes. It is the unity generally used in the
LHC environment.

GRID: is the best computing combination of distributed and shared

CPUs and storage, added by higher bandwidths. No bottleneck is ac-
cepted in the network.

EGEE[1]: Enabling Grids for E-science in Europe. This is the European

organization for Grid in Science. It is a strong collaboration with high
energy physics basic software.

OSG[2]: Open Science Grid. This is the simmilar organization for United
States Sciences.

LCG[3]: LHC Computing Grid middleware. This is the organization that

takes care of the Grid for the four LHC experiments.

GRID3: Is the organization that coordinates the HEPGRID in United

States.

All these organizations are consequence of the development of the GRID

in HEP and the interest for this technology by other sciences. Many computing
professionals and industries are now involved with Grid Technology meaning
that many improvements will soon appear.
High Energy Physics GRID 197

1.3 New World

HEP has one of the most advanced and suitable structures for GRID de-
velopment. A GRID has to take into acount each aspect of one HEP experiment
as, (i) Project for Particle Accelerators and Detectors; (ii) Data Acquisition Sys-
tems and Data Storage; (iii) Development of software and languages; (iv) Data
reconstruction using parallel processing; (v)Monitoring, control, simmulation,
security, networking; (vi) Data Analysis (one of the most important activities
in GRID computing for LHC era); and (vii) Video Conferencing (already used
at global level). All this defines a new world or a new way to work. The devel-
opment of GRID and applications will certainly represent a new revolution in
the internet. Several branches of society are being re-organized in view of these
new possibilities.

2 Projects

HEP GRID has a number of projects based on the planning of new collidder
experiments. These experiments and related Grid projects are summarized here.

2.1 iVDGL + GryPhyN + PPDG = Trilling

Trilling (the name of a beautiful flower) is a coordination of the three

main USA HEP Grid projects:iVDGL[4]; GriPhyN[5] and PPDG[6]. It was
noticed that there was a considerable overlap of software developments, people
and experiments. This coordination has brought benefits to all of them and has
started to produce software like VDT (Virtual Data Toolkit) + PACKMAN
(package management and distribution tools). Each project is associated to a
set of physics experiments:
iVDGL, GriPhyN is associated to LIGO (an experiment for detect-
ing Einstein’ s Gravitational Waves) + NVO (National Virtual Observatory)
+ SDSS (Sloan Digital Sky Survey) nnd PPDG is associated to BaBar and
Nuclear Physics.
We can name these projects as General Grid Projects for HEP. The only
common specific purpose is to be dedicated to one or more HEP experiment.
The choice of HEP is clear: it is one of the best ”laboratories” to develop an
idea like GRID. We encourage the reader to go through the web pages listed at
the end of this text to have a complete view of each one of these projects.
198 Alberto Santoro

2.2 GRID for Alice, Atlas, LHCb, CMS

We would like to summarize the LHC experiments and direct the reader
to the web pages for more details. They can be used by anyone who would like
to be associate to one of these experiments and, eventually, by referees. I use
practically the same web pages and information from GRID groups for each
LHC experiment that I used in my talk in Lima (SILAFAE).[7] Figure 1 shows
the four detectors to be installed at LHC.

Figure 1: The four detectors under construction for the LHC.

ALICE[8] is dedicated mainly to plasma physics, heavy ions colliding in

the center of mass with very high energies. The collaboration count about 1000
physicists from almost one hundred institutions all over the world. The ALIEN
(ALIce ENvironment) is a system of softwares created as a grid-like system for
job submission and data management. This system intends to help ALICE to
build a computing mode to be defined based on functionatility, interoperability,
performance, scalability and standards. These properties do not differ much
from the ones of the other LHC experiments.
ATLAS[9] is a collaboration of approximately 2000 physicists and engi-
neers from about 150 Institutes from 34 Countries. The Atlas detector weights
7000 Tons, is 25 m in diameter and 46 m long. For more details the reader
High Energy Physics GRID 199

should go to the Atlas[9] web page. Many technologies have been developed
by ATLAS collaboration to build one of the most interesting general purpose
detector. The goals of the experiment are: Detect the Standard Model Higgs
Boson, detect Supersymmetric states, study Standard Model QCD (Chromod-
inamics), EW (Electroweek), HQ (Heavy Quark) Physics, and new physics (to
be defined by each collaboration but, in general, we can say that new physics
is everything beyond the Standard Model). The Collaboration has been in
intensive collaborative development of GRID software. Some of the tools de-
veloped in collaboration with LHCb are: GridView ( Simple tool to monitor
status of testbed), Gripe (unified user accounts), Magda (Manager for Grid
Data), Pacman (package management and distribution tool), Grappa (web por-
tal using active notebook technology), GRAT (Grid Application Toolkit), Grd-
searcher (browser), GridExpert (Knowledge Database), VOToolkit (Site Au-
thentication,Authorization).
LHCb[10], is a detector dedicated mainly to b physics. The LHCb (Large
Hadron Collider Beauty Experiment) experiment has 563 physicists from 50 In-
stitutes from 12 Countries. The experiment expects to get 1012 bb pairs per year,
a much higher statistics than the current B factories. Another set of numbers ex-
pected for LHCb experiments is: (i) 200,000 reconstructed B 0 → J/ψKs events
per year; (ii) 26,000 reconstructed B 0 → π + π − ; (iii) all B Mesons and Barions.
The LHCb collaboration has produced many useful software for analysis in the
near future in cooperation with ATLAS. Two examples, in addition to those
pointed out above on ATLAS subsection are GANGA ( Gaudi ANd Grid), an
user interface for Grid, and DIRAC ( Distributed Infrastructure with Remote
Agent Control) for Monte Carlo event production. Details of this experiment
can be found on the LHCb[10] web page.
CMS[11] collaboration has approximately 2000 physicists from 160 Insti-
tutions from about 40 countries. It weights 12,500 Tons, is 15 m in diameter
and 22 m in length. It will use a Magnet field of 4 Tesla.
The Detector is composed by Silicon Microstrips as a central Tracker;
Electromagnetic and Hadronic Calorimeters; a Superconducting Coil; Iron Yoke;
Muon Barrel and Muon Endcaps. CMS will explore a big number of physics
topics. In the case of Higgs, the collaboration intends to explore the full range
of 100 - 1000 GeV as the allowed region for Higgs mass. Topics as QCD,
Heavy Flavor Physics, SUSY, New Phenomena are in the list of future analysis.
Diffractive Physics will be explored by the Totem Group (for elastic and total
cross section only) and by a large number of physicists interested in exploring
all innelastic diffraction and respective topologies. Hard diffraction will be the
central part of these studies.
200 Alberto Santoro

2.3 Other GRID Projects

The idea of the Grid computing architecture can be generalized in science
as an useful tool for all data intensive Sciences.
Each area is developing its own framework, with similar characteristics
of that used in HEP, many times using almost the same software, as it is the
case of the Mamogrid project. Security Services are also common developments
using several generic mechanisms (maintaining the data Integrity, confientiality,
authentication, availability, and several types of appropriated applications.
The interested reader can easily find examples in several fields like medicine
(Mammogrid[12]), astronomy (SLOAN-Digital Sky Survey), Gravitation (LIGO,
the gravity wave experiment), Biology (the Genoma project), Fusion Physics
(project ITER and Gloria D). It is important to keep in mind that all areas
with a data intensive system is a good field for the development of a computing
Grid.

3 Digital Divide and GRID

Networking is one important component of the GRID systems. In a Grid

architecture, all available CPUs are useful only if they are connected by a good
bandwidth. The existence of a good bandwidth is the fundamental condition
for the stablishing of a good Grid. ICFA (International Committee for Future
Accelerators) has set a subcommittee known as SCIC (Standing Committee on
Interregional Connectivity) to investigate this question in the HEP field. This
committee issued a report[13] that evaluates the world situation of networks and
points many regions with strong connectivity problems. The bottleneck may be
present in different parts of a network. There are problems like poor last mile
connection, long mile connection, poor connectivity in the local institutions
network. The idea of Grid for the next generation of experiments will not
achieve its purposes if the bottlenecks are not eliminated. See details in the
reports of ICFA/SCIC/DD[13].

4 News and Conclusions

After the School HEPGRID has experienced considerable progress in

several different projects. HEPGRID BRAZIL[14], that was inaugurated in
November at UERJ/BRAZIL with 200 CPUs, is a good example. Progress was
also done in the Networks[15] (see the ICFA/SCIC report). Under the leader-
ship of CALTECH, a consortium of institutions (in which Brazilian institutions
- UNESP, ANSP, UERJ, RNP - have actively participated) has set a demonstra-
High Energy Physics GRID 201

tion that consisted in transmitting a record of 101 Gbps. This was one of the
activities of the Bandwidth Challenge of the Supercomputing 2005 conference.
We expect that the T2-HEPGRID BRAZIL will soon be connected to the full
circuit of CMS Grid. Basic information about Grid systems can be find on the
web[16, 17, 18, 19]

5 Acknowledgments

The author thanks the Organizers of the ICFA School for the invitation to
present this talk in Itacuruçá, where we found a very good ambient created by
the participants. I would like to thank FAPERJ and CNPq by partial financial
support. Finally, I thank my colleagues P. Avery, H. Newman, D. Barberis, J.
Bunn, R. Gardness, R. Mount, S. Bgnaco, P. Cerello, R. Barbera, P. Buncic,
F. Caminati, P. Satz, G. Pulard, N. Brook, C.Eck, J. Marco, F. Gagliardi,
T. Wenaus and F. Harri for all the information about their projects. Finally
I would like to thank my colleague Helio da Mota for interesting discussions
during the preparation of this talk.

References

[1] http://public.eu-egee.org/

[2] http://www.opensciencegrid.org/

[3] http://lcg.web.cern.ch/LCG/

[4] http://www.ivdgl.org/

[5] http://www.griphyn.org/

[6] http://www.ppdg.net/

[7] Future Experiments - GRID and LHC Proceedings of the Simpósio Latino
Americano de Fı́sica de Altas Energias -Lima, Peru, July 12-17,2004.

[8] http://www.alice.cern.ch/

[9] http://www.atlas.cern.ch/

[10] http://www.lhcb.cern.ch/

[11] http://www.cms.cern.ch/
202 Alberto Santoro

[12] See talk ”Mamogrid: applying Grid Technology to Health Care (in
breast cancer diagnose)” by Salvator Roberto Amendolia (CERN) in
http://www.lishep.uerj.br/
[13] http://icfa-scic.web.cern.ch/ICFA-SCIC/;
http://www.weforum.org/site/homepublic.nsf/Content/Global+Digital+
Divide+Initiative; and the book ”The Digital Divide”, Edited by Benjamin
M. Campaine, 2003.
[14] http://www.hepgridbrasil.uerj.br/
[15] http://ultralight.caltech.edu/gaeweb/
[16] http://www.cs.wisc.edu/condor/
[17] http://www.globus.org
[18] (i) The Grid Blueprint for a New computing Infrastructure, Edited by Ian
Foster and Carl Kesselman, and the second volume too; (ii) Grid Comput-
ing Making the Global Infrastructure a Reality - Fram Berman , Anthony
J.G. Hey and Geoffrey C. Fox
[19] The Digital Divide - Facing a Crisis or Creating a Myth? - Edited by
Benjamin M. Compaine
Laboratory Courses
 Laboratory Course on Silicon Sensors
Elisabetta Crescio, Marek Idzik
INFN, Torino, Italy

Danielle Moraes and Alan Rudge

CERN, CH-1211, Geneva, Switzerland

ABSTRACT
This laboratory course consists of five different mini sessions, in order to give
the student some hands-on experience on various aspects of silicon sensors and
related integrated electronics. The five experiments are:
• Characterization of silicon detectors (VCI measurements).
• Double light spot (measurement of the collection time of electrons and
holes separately versus voltage in a silicon diode).
• Understanding of the VA read-out chip operation, and measurement of its
noise versus detector capacitance characteristics.
• Measurement of the position resolution of a microstrip detector.
• Observation of spectra in a silicon diode.

1 Characterization of silicon diodes for particle detection

1.1 Introduction
Near intrinsic n-type silicon with a metallised p-doped region is the most
frequently used semiconductor structure for detecting charged tracks in high-
energy physics experiments. A polarisation voltage is applied across the diode
structure, which depletes the silicon from charge carriers. Charged particles
or photons interacting with the silicon will create electron-hole pairs that drift
along the electric field lines to the contacts located on the silicon surface. A
schematic picture of a silicon sensor diode is shown in Figure 1.
A first step in constructing a particle detector based upon a silicon sensor
is to characterise the sensor without readout electronics attached. The static
characteristics of a sensor are usually adequate to determine if the sensor can
be used for particle detection. The leakage current behaviour as a function of
206 Elisabetta Crescio et al.

metal contact
charged particle −V
p−type implant

guard ring
n−type bulk
n+ type contact

electron GND
hole

Figure 1: A schematic picture of a silicon sensor diode.

voltage and the voltage needed to fully deplete the sensor are two important
parameters. The voltage needed to fully deplete the sensor can be determined
by measuring the capacitance between the diode implant and the backplane
of the sensor. In the final particle detector system, both the capacitance and
the leakage current will influence the performance of the readout electronics.
The capacitance and leakage current depend on the geometry of the sensor and
the quality of the material and manufactoring process. In a well controlled
and uniform process sensors with the same geometrical layout, processed on
the same substrate, should have the same behaviour. In reality, there may be
variation both in the process and in material and therefore there may be sensors
which differ largely from what we naively would expect. When constructing
an experiment consisting of many sensors we have to measure them in the
laboratory to find the good sensors that can be assembled into the experiment.
This session requires some knowledge of the basic principles of diodes and
will give experience handling unprotected diodes, operating the microscope and
probe manipulators. Because of the short time available we restrict ourselves
to simple DC-coupled diodes.

1.2 Laboratory setup

In this experiment we use the following equipment:
• probe station;
• digital multimeter;
• capacitance meter and power supply;
Laboratory Course on Silicon Sensors 207

• vacuum pump;
• microscope;
In Figure 2 a schematic picture of the setup is shown.

CV-meter

probe
station

microscope

conductive
rubber
Vacuum
diode pump

Figure 2: A schematic picture of the setup of CV and IV measurement.

1.3 Measurement of IV-properties of silicon diodes at room temperature

The main sources of leakage current in silicon sensors are:
1. Diffusion of charge carriers from undepleted regions of the detector to the
depleted region.
2. Thermal generation of electron-hole pairs in the depleted region.
3. Surface currents depending on contamination, surface defects from pro-
cessing and edge effect from dicing, etc.
208 Elisabetta Crescio et al.

Contribution (1) is generally well controlled and small giving a few nA/cm2 .
The contribution from (2) depends largely on the purity of the material since
recombination centres and trapping centres increase the creation of electron-
hole pairs. The magnitude is higher than that from (1), giving a few µA/cm2 .
The leakage current originating from thermal generation is of course tempera-
ture dependent. By lowering the temperature of the sensor we may reduce the
contribution. By decreasing the temperature by 10o C the leakage current will
typically be reduced to a third. In some cases the contribution from (3) may be
the dominant source of leakage current. The surface current may be caused by
effects on the non-depleted edge region or by a bad processing environment. The
leakage current originating from the surface may vary extensively from sensor
to sensor. To reduce the effects from surface current a guard ring structure is
processed on the silicon. The guard ring can be anything, from a single implant
around the diode to a complex structure of alternating implants and floating
metal rings around the silicon diode. We will now study the IV-characteristic
of a silicon diode sensor with 300 µ m thickness. Execute the steps below:

1. Place the silicon diode on the vacuum chuck with conductive rubber under
it.

2. Connect the diode with the probe needle to the negative pole of the battery
pack.

3. Connect the chuck (and thus, the silicon backplane) to the positive pole
of the battery.

4. Cover the probe station to prevent light from generating current in the
diodes, note down the voltage and current at 0 V.

5. Ramp up the voltage and note down the voltage and the corresponding
current.

1.4 Measurement of CV-properties of silicon diodes at room temperature

At low reverse bias voltage the capacitance will fall such that 1/C 2 is
proportional to V. When the sensor has reached full depletion the capacitance
will not change anymore. This is clearly visible when plotting 1/C 2 vs. V. The
point where the curve shows a kink and does not reduce further gives the point
for full depletion. For large single diodes the capacitance to the backplane is
the dominant contribution to the total capacitance. For small and segmented
sensors such as pixel and microstrip sensors the inter pixel/strip capacitance will
dominate over the backplane capacitance when the sensor is fully depleted. The
Laboratory Course on Silicon Sensors 209

inter pixel/strip capacitance do not change in the same way as the backplane
capacitance when applying reverse bias to the sensor. In this experiment, the
capacitance of the silicon diode is measured between the backplane and the
p-implant. Execute the steps below:
1. Find the capacitance C0 of the setup by leaving the circuit open.
2. Follow the procedure outlined for IV-measurements. Measure the full
capacitance (capacitance of the diode and capacitance of the setup), cal-
culate the capacitance of the diode by subtracting the value C0 from the
measured one, and note down the measurements.

1.5 Results from the measurements

The results obtained from the measurements are shown in Table 1.5. The
corresponding curves are shown in Figure 3. The silicon diode shows a strong
increase in leakage current after 40V. This may indicate a breakdown in the
structure. The capacitance decreases with higher depletion voltage, and the
silicon diode shows the expected behaviour of capacitance vs. depletion voltage.
At 60V the silicon diode shows full depletion.

Table 1: IV and CV measurements.

Voltage[V] Current[pA] Capacitance[pF] C0

0 -12 51 38 pF
10 27 28
20 56 18
30 6 16
40 86 13
50 114 11
60 132 10
70 215 10
80 360
210 Elisabetta Crescio et al.

102
I (pA)

C (pF)
2
10

10

10

0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80
Voltage (V) Voltage (V)

Figure 3: The current of the diode as a function of the applied voltage (left) and the
logarithm of the corrected capacitance vs. depletion voltage in V (right).

2 Study of the Viking architecture and noise performance

2.1 The Viking architecture

A number of readout methodologies exist. Among these, the principal ones

are the MX series which use double correlated sampling, successfully used in
the DELPHI vertex detector, and the Viking ”time continuous” shaping. The
Amplex/Viking ”time continuous” shaping is shown in Figure 4.
A voltage corresponding to the input charge is stored on the sample and
hold capacitor and read out sequentially via the output multiplexer. It is im-
portant to understand the shift in/out concept to daisy chain many (up to 20)
chips. The signals “clockb” and “shift-in” activate the start and stop unit in
the chip, which creates an internal clock (“clocki”), a “start” signal for the shift
registers and activate signal (“ero”) for the output buffer. “Ero” is necessary
to allow daisy chaining of chips. The “start” goes into the output multiplexer
and “clocki” shifts it through the 128 channels so that each channel is for one
clock cycle connected to the output buffer to read the channels out. After 128
clock cycles the outcoming “shift-in” from channel 128 stops the internal clock,
disables the output buffer and creates a shift-out which can be used as a shift-in
for the next readout chip. Figure 5 shows a timing diagram for the readout se-
quence of a Viking chip. Corresponding screen shots from a digital oscilloscope
are shown in Figure 7(a) and Figure 7(b). Corresponding screen shots from a
digital oscilloscope are shown in Figure ??(a) and ??(b).
An important mode of operation is the single channel mode, where the
output from one amplifier is seen. This is shown in Figure ??(b) for a signal
from an Americium source.
Laboratory Course on Silicon Sensors 211

test_on
ckb shift_in_b

pad preamp shaper S&H

128 Analogue multiplexer

128 Analogue multiplexer

128 Cell bit register

pad preamp shaper S&H
128 Cell bit register

pad preamp shaper S&H

pad preamp shaper S&H

pad preamp shaper S&H

shift_out_b
holdb
Analogue bias analogue out
& adjustments

Figure 4: The Amplex/Viking architecture.

2.2 Study of noise performance of the Viking readout circuit

An essential part in operating silicon sensors is a low noise amplifier circuit.

The energy needed to create an electron-hole pair at room temperature in silicon
is around 3.6 eV. A minimum ionising particle traversing 300 µ m of silicon will
on average create 25000 electron-hole pairs. For low energy X-ray applications
the requirements for low noise is even more demanding. A 10 KeV X-ray will
only produce 2800 electron-hole pairs in silicon. The main contribution of the
silicon sensor to the total noise of the assembly comes from:

• the load capacitance of the silicon sensor;

• the leakage current in the silicon sensor;

• possible resistance between the active element of the sensor and ground,
or the bias supply.
212 Elisabetta Crescio et al.

Holdb
T=1us
Shift_in_b

Ckb

Shift_out−b
Signal?
Analogue out
#1 #2 #3 #4 #5 #6 #127
First (=top) Last (=bottom)
channel channel
Time of
physics event

Figure 5: Timing diagram for the readout sequence of a Viking chip.

16-Jun-99 HARDCOPY
16-Jun-99 CHANNEL 1
10:22:45 output to
14:49:53 Input 2 Int. Printer
2 Normal 1 s Disk
GPIB
1 s H1BN
5.0 V R$ 232
Centronicos
2.00 V Coupling
2 D50 1 page feed
Grounded
1 DC1M 1 s
1 s Grounded
AC1M 100 mV
Off On

2.00 V protocol
V/div OFFset
Deskjet b/w
NORMAL
HP 7470
ECL TTL HP 7550
TIFF
1 Global BWL
TIFF compr.

Off On

( 30 MHz )

Probe sensed (x10)

2
1
1 s BWL
1 .2 V DC 10 Ext DC - 360 mV STOPPED 1 s BWL 9.6 s
100 Ms/s
2 .2 V DC 10 1 .1 V AC
Ext DC - 360 mV STOPPED
2 .5 V DC 100 Ms/s

(a) (b)

Figure 6: (a)Screen shots showing hold and shift signals. (b)Screen shots showing
clocking and analogue output pulse.

The main sources of noise for silicon detector assemblies are: The Equiv-
alent Noise Charge (ENC) of the input FET of the pre-amplifiers proportional
to the load capacitance:
�
Ce 4kT
EN CF ET = (1)
q 3gm τ
Laboratory Course on Silicon Sensors 213

16-Jun-99 CHANNEL 1
TRIGGER SETUP 15:10:20
16-Jun-99 Input
Normal
15:02:08 Edge SMART
HIBU

Coupling
DC50
Grounded

1 1 Ocino
Grounded
2 s
trigger or
1 2 Ext
2 s AC1

1.00 V Extlo Lino 0.50 V V/div Off set

coupling 1 NORMAL
DC AC LFREJ ECL TTL
HIFREJ HF

clope 1
Pos Neg Global DNL
Window Off On
(38 MHz)

Probe Atten
holdoff x1
2 s x2
SHL OFF Time Evts x5
x10
x20
1.1 V DC 2 s BNL
2.2 V DC 1 C 0.60 V STOPPED
108 Ms/s 1.5 V AC
2.2 V OC 1 C 0.33 V STOPPED
100 Ms/s

(a) (b)

Figure 7: (a)Screen shots showing output waveform showing hits in two adjacent channels.
(b)Output from a single channel.

The detector (and input FET) leakage noise:

�
e qIl τ
EN Cl = (2)
q 4

The bias and feedback resistor noise:

�
e τ kT
EN CR = (3)
q 2R

where C is the total load capacitance, Il is the leakage current, τ is the peaking
time of the amplifier, k the Boltzmann constant, T the absolute temperature in
Kelvin and R is parallel of the bias resistor and feedback resistor. The parameter
gm is the trasconductance of the MOS device. The total noise contribution is
the squared sum of the components listed above:
�
EN Ctotal = EN CF2 ET + EN Cl2 + EN CR
2 (4)

For the Viking chip the noise is typically 70 e− + 12 e− Cload [pF ]. It is obvious
that we want to keep the leakage current small by choosing a sensor that has
low leakage current at a voltage that fully depletes the sensor. The contribution
to the noise from the detector capacitance and series resistance will be studied
in this experiment.
214 Elisabetta Crescio et al.

2.3 Laboratory setup

In this session we use the following equipment:

1. Timing unit for readout circuit, VIKING TIMING;

2. Pulse generator for trigger;

3. Current limited power supply for readout circuit;

4. NIM crate;

5. Digital oscilloscope capable of measuring RMS.

We have an assembly with a Viking chip connected to capacitors of various

values.

2.4 Measurement
In order to measure the noise, the setup has to be calibrated. This is done
by applying a known charge pulse Qcal on the input of the Viking chip and
measuring the output from the chip Voutl . The input charge is generated by
applying a known voltage pulse Vcal over a known calibration capacitor Ccal :

Ccal V cal
Qcal = (5)
e
where e is here the magnitude of the electron charge (e = 1.6 × 10−19 C. The
calibration capacitance Ccal is on our case 1.8 pF. The RMS of the noise Vnoise
from the Viking chip without test pulse in mV can be calculated using a modern
digital oscilloscope. Since the calibration is known, the measured value in mV
can be converted to ENC by the relation:

Vnoise Qcal
EN C = RM S e (6)
Vout
Six channels have been pre-wired with different load capacitances to the input
pads of the Viking chip. Determine the calibration and measure the correspond-
ing ENC noise for the channels, and note down the values.

2.5 Results from the measurement

The measured values are reported in Table 2.5.
The plot of noise vs. capacitance is shown in Figure 8.
Laboratory Course on Silicon Sensors 215

Table 2: Values of the measured noise.

Capacitance[pF] Vcal [mV] Vout [mV] Vnoise [RMS mV] ENC[RMS e− ]

0 2 2500 10 90
3 2 2500 13 115
13 2 2500 28 250
25 2 2300 43 420
36 2 2200 60 600
51 2 2000 78 880

1000
Noise RMS (e-)

χ2 / ndf 3017 / 4
900
p0 15.4 ± 0.7146
800
p1 63.89 ± 20
700

600

500

400

300

200

100

0 10 20 30 40 50
Capacitance (pF)

Figure 8: VA2 noise vs. detector capacitance.

3 Study of the position resolution of a silicon microstrip detector

3.1 Introduction
Silicon microstrip sensors are the most commonly used device for high reso-
lution tracking in particle physics. The strip design allows a large sensitive area
with relatively few readout channels. The basic strip detector is read out on
one side giving information of the track position only in one dimension. Vari-
ous solutions to measure track position in two dimensions exist. The simplest
solution is to glue two single-sided sensors back-to-back, but a more demanding
design is to process strips on both sides of the sensor. In this experiment we
216 Elisabetta Crescio et al.

will study a single sided sensor which is illuminated by a pulsed laser source on
the strip side (because the back of the sensor is fully aluminised).

3.2 The laboratory setup

In this experiment we use the following equipment:

1. Timing unit for readout circuit, VIKING TIMING;

2. Pulse generator for trigger and laser;

3. Current limited power supply for readout circuit;

4. NIM crate;

5. Oscilloscope;

6. Laser diode driver;

7. Battery pack for biasing the sensor;

The silicon microstrip sensor is wire bonded to the Viking readout circuit, which
has been placed on a readout PCB (Printed Circuit Board). The assembly has
been mounted on a slide which can be precisely moved such that the translation
direction is orthogonal to the strips. An optical fibre has been mounted about
100 µ m away from the silicon surface. Figure 9 shows a photograph of the PCB
card and the detector mounted on the slide.

Figure 9: Picture of the PCB board and the detector mounted on the slide.
Laboratory Course on Silicon Sensors 217

3.3 Measurement

We will now try to determine the position resolution of the silicon microstrip
sensor. The peak of the light source is a few strips wide. We require the
information from a number of strips in order to accurately determine the peak
position. By moving the fibre closer to the silicon sensor the laser spot size
will reduce, but on the other hand we risk mechanically damaging the sensor.
Proceed as folows:

1. move the slide by turning the micrometer screw on the right hand side
of the box (seen from the repeater electronics board). You will now see
the signal from the light moving from one strip to another. You may also
notice that this translation is not very smooth. The reason for this is the
aluminium on top of the implanted strips which reflect a fraction of the
light and reduces the signal locally. This effect would be larger if we had
a better focused light spot.

2. Place the sensor in a region with a nicely distributed signal.

3. Determine and write down the amplitudes of the channels in the peak by
moving the cursor on the oscilloscope.

4. Four complete turns of the screw correspond to 1 mm. Therefore one

major division equals 10 µ m. Move the slide by 10 µ m and repeat 3.

5. Repeat 4. and 3.

3.4 Results from the measurement

Table 3.4 shows the measured amplitudes of the signal for different positions
of the sensor. Figure 10 shows the oscilloscope output without a laser spot (a),
and with the laser spot in two different positions (b and c).
One horizontal division corresponds to about two readout channels. Now
perform a gaussian fit for each position and calculate the peak positions. Table
3.4 shows the results from a previous measurement.
Since we are averaging the signals from the Viking in order to minimize
the electronics noise, and taking 10 points to determine the peak in the gaussian
fit, it can be seen that position accuracy down to a few µ m can be achieved.
What are the main error sources?
218 Elisabetta Crescio et al.

Table 3: Raw data from the setup.

Strip n 0 µm 10 µ m 20 µ m 30 µ m 40 µ m 50 µ m
1 304 292 268 240 232 212
2 424 400 368 340 320 288
3 612 584 544 504 472 436
4 888 844 808 760 716 680
5 1030 1000 984 948 912 884
6 1150 1140 1130 1100 1080 1050
7 1140 1160 1170 1150 1130 1120
8 1010 1070 1100 1130 1160 1200
9 736 772 820 864 916 968

CHANNEL 1 CHANNEL 1 CHANNEL 1

9-Jul-99 9-Jul-99 9-Jul-99
8:31:27 Coupling 8:31:27 Coupling 8:31:27 Coupling
1 1 1
DC50" DC50" DC50"
2 !S Grounded 2 !S Grounded 2 !S Grounded
200 mV 200 mV 200 mV
DC1M" DC1M" DC1M"
Grounded Grounded Grounded
ACM" ACM" ACM"

V/div Offset V/div Offset V/div Offset

NORMAL NORMAL NORMAL
ECL TTL ECL TTL ECL TTL

Global BWL Global BWL Global BWL

Off 30MHz Off 30MHz Off 30MHz

Probe Atten Probe Atten Probe Atten

x1 x1 x1

x2 x2 x2
x5 x5 x5
1 x10 1 x10 1 x10
x20 x20 x20
2 !s 2.6 !s 2 !s 2.6 !s 2 !s 2.6 !s
1.2 V 50" 1.2 V 50" 1.2 V 50"
100 MS/s 100 MS/s 100 MS/s
2.5 V AC 2.5 V AC 2.5 V AC
2 DC -0.11 V 2 DC -0.11 V 2 DC -0.11 V
AUTO AUTO AUTO

(a) (b) (c)

Figure 10: Oscilloscope output without a laser spot (a), and with the laser spot in two
different positions (b and c).

Table 4: Results from the position resolution measurement (in µ m).

0 µm 10 µ m 20 µ m 30 µ m 40 µ m 50 µ m
Calculated peak 264 274 282 292 302 312
Difference 10 8 10 10 10
Laboratory Course on Silicon Sensors 219

4 Study of charge transport in silicon with a fast amplifier

4.1 Introduction
In gaseous detectors the mobility for electrons is several orders of magni-
tude higher than for positive ions. In semiconductors the mobility for holes
is only slightly lower than for electrons. In general the signal propagation in
semiconductors is a few nanoseconds while the signal propagation in gaseous
detectors typically varies from microseconds to milliseconds. In order to study
the drifting of electrons and holes in silicon, very fast electronics is required.
The drift velocity for electrons and holes in silicon at low electric field strength
is given by:
ve = µe E (7)

vh = µh E (8)
where µe and µh are the mobilities for electrons and holes respectively, and E is
the electric field. The mobility in silicon at room temperature is 1350 cm2 /V s
for electrons, and 480 cm2 /V s for holes. At high field, the velocity saturates
with velocities of the order of 107 cm/s.
When the sensor is fully depleted, the signals from holes and electrons will
arrive almost at the same time. We can try to study the slower transit of holes
by shining a short laser pulse on the back side of a non-depleted n-type silicon
diode. By choosing a wavelength which does not penetrate far in the silicon
the charge can be generated close to the surface of the silicon. The holes have
now to drift to the other side of the sensor, while the electrons are formed at
the interface. If the electric field is low and the detector has a reasonably large
depleted region we will now see a difference in the total time between the signal
arising from electrons and holes. It is important to realise that the signal itself
arises immediately as the charge carriers start moving, and lasts until the last
charge carriers are collected.
By starting with a high depletion voltage the signal from the setup will have
the same shape as the light pulse from the laser which is shown in Figure 11.
The current signal which is induced on the diode is due to the movement
of both charge carriers in the electric field. The transit time for the carriers
can be calculated by integrating Equations 7 and , remembering that the field
varies as a function of position:

qND
E(x) = x + Emin (9)
�
where � is the dielectric constant, Emin is a function of the bias voltage, and is
220 Elisabetta Crescio et al.

4-Jun-99 MEASURE
10:30:31
A : Average (1) OFF Cursors
20 ns Parameters
176 mV mode
-573.4 mV Time
208 swps Amplitude
type
Relative
Absolute
show
Diff - Ref
1 Diff & Ref
20 ns
200 mV
-570 mV
Reference
cursor
Track OFF On

Difference
20 ns cursor
1 .2 V 50 t 7.3 ns 1 137 MHz
2 .2 V AC t 500 MS/s
3 .5 V 50 1 DC -0.344 V
4 .10 mV 50 NORMAL

Figure 11: The amplifier output from the laser pulse with a fully depleted silicon diode.

a minimum of zero for a barely depleted detector, then the transit times are:

� � w + (�/qN )E �
D min
th = ln (10)
µh ND x0 + (�/qND )Emin

� � x + (�/qN )E �
0 D min
te = ln (11)
µh ND (�/qND )Emin
where x0 is the depth of production, measured from the ohmic contact side.
The induced current is given by Ramos theorem, which states that the current
on the electrode of interest is equal to the charge value multiplied by the dot
product of the ”weighting field” and the charge velocity. The weighting field
is an hypothetical field calculated by putting unit potential on the electrode of
interest, zero on all other electrodes, and ignoring any static charges (ND in
our case). For a simple diode, this reduces to a constant 1/w, and the velocities
Laboratory Course on Silicon Sensors 221

are aligned with the electric field, so calculating the scalar velocities gives the
induced current:
1 � x0 qND ) �� � ND ��
ih = q µh Emin + exp µh q t (12)
w � �

1 � x0 qND ) �� � ND ��
ie = q µe Emin + exp − µe q t (13)
w � �
In fact these pulses become smeared by the amplifier bandwidth, and the re-
sulting pulse is what is seen on the oscilloscope.

4.2 The setup

In this experiment we use a fast amplifier chain connected to a diode of
n-type silicon with the backplane not covered by aluminium. Two light fibres
lead the laser light to the two opposite sides of the diodes. A picture of the
sensor setup, with a zoom of the sensor region, is shown in Figure 12.

Figure 12: Picture of the sensor setup.Top:box containing the detector with the two optic
fibers and cables for the detector polarization and read-out. Bottom: zoom on the detector
and electronics.

Other equipment used are:

222 Elisabetta Crescio et al.

1. Oscilloscope;

2. Laser diode driver;

3. Battery pack for biasing the sensor;

4. Pulse generator for trigger and laser;

5. Low voltage power supply;

6. NIM crate.

4.3 Signals when shining laser on the junction side (p-side)

With the laser pulse on the p-side, you get the picture on oscilloscope shown
in Figure 13.

2-Jun-99 MEASURE
22:23:54
B : Average (A) Off Cursors
20 ns Parameters
49.5 mV mode
-107.26 mV
Time
139 swps
Amplitude
B : 1-4 type
20 ns Relative
44.0 mV Absolute
-105.5 mV
show
Diff - Ref
Diff & Ref

Reference
cursos
Track Off On

Dofference
cursor
20 ns
1. 20 mV 50 t -12.2 ns 1 -82.0 Mhz
2. 2 V AC t 500 MS/s
3. 5 V 50
Ext AC -240 mV 1M
4. 20 mV 50 NORMAL

Figure 13: The signal from an unbiased sensor (read out from the p-side) when laser is
shined on the p-side.

Turn on the sensor bias and ramp slowly up the voltage. The amplitude
of the signal gets larger but the shape stays approximately unchanged. The
sensor is depleted from the p-side and the n-side is conducting transporting the
electrons to the amplifier.

4.4 Signals when shining laser on the backplane (n-side)

When the pulse is incident on the n-side and the detector is off, there is
a very tiny signal. The holes are trapped close to the backplane. As the bias
is increased, one sees an increasing signal (see Figure 14), with a double peak
Laboratory Course on Silicon Sensors 223

structure. In the double peak structure are visible the contribution coming from
fast arriving electrons and the contribution from holes with a lower mobility
which have to travel through the depleted region of the sensor.

4-Jun-99 MEASURE
10:36:27
A : Average (1) Off Cursors
20 ns Parameters
9.6 mV
-9.963 mV mode
451 swps 1 Time
Amplitude
type
Relative
Absolute
show
1 Diff - Ref
Diff & Ref
20 ns
10.0 mV
-8.55 mV

Reference
cursor
Track Off On

Difference
cursor
20 ns
1 10 mV 50 t 11.3 ns 1/!t 88.5 Mhz
2 .2 V AC 500 MS/s
3 5 V 50
4 10 mV 50 1 DC -30.6 mV
NORMAL

Figure 14: The signal amplitude at low bias voltage when laser is shined on n-side of the
silicon sensor (The signal is read out from the n-side).

The double peak structure disappears when the bias voltage gets higher.
At full depletion the amplitude of the signal is comparable to that obtained
when laser is shined on p and n sides (see Figure 15).

4-Jun-99 MEASURE
10:28:50
A : Average(1) Off Cursors
20 ns Parameters
44.0 mV mode
-161.95 mV Time
150 swps 1 Amplitude
type
Relative
Absolute
show
Diff – Ref
1 Absolute
20 ns
44.0 mV
-155.8 mV

Reference
cursor
Track Off On

Difference
cursor
20 ns
"t 13.4 ns 1/"t 74.6 Mhz
1 50 mV 50! 500 MS/s
2 .2 V AC 1 DC – 86 mV
3 5 V 50! NORMAL
4 10 mV 50!

Figure 15: The signal amplitude at full depletion when laser is shined on n-side (read out
from n-side).

The optical fibre cables are not labeled; can you discover, by
224 Elisabetta Crescio et al.

using one or the other cable, which cable points on the n-side, and
which on the p-side?

5 Spectroscopy with the Viking chip and pad detector

5.1 Introduction
During this experiment we will use a silicon sensor to see the spectrum form
a γ source and study its energy resolution.

5.2 The setup

The VIKING chip is used in single channel mode and the inputs bonded to
a small silicon detector with 36 pads of 1 mm square. Channel 12 is bonded to
1 pad, channel 36 to 2 pads, channel 58 to 4 pads, channel 78 to 8 pads and
channel 102 to 21 pads. The other equipment used in this experiment is the
following:

1. Oscilloscope;

2. 241
Am source;

3. Battery pack for biasing the sensor;

4. Viking Timing unit;

5. Low voltage power supply;

6. NIM crate.

7. Multi Channel Analyzer and PC.

A picture of the sensor setup is shown in Figure 16.

5.3 The measurement

A spectrum of 241 Am is taken with channel 12 (only one pad) with the
Am source placed under the printed circuit board hosting the detector. The
spectrum is shown in Figure 17. The peak corresponding to the 59.95 X-ray
line is clearly visible at channel 54. Since the detector is supported on the
board with copper and the contact to the backplane of the sensor is done with
silver loaded glue, other peaks at lower energies are observed, corresponding to
florescence lines excited by the 59.95 keV line from copper and silver. The Kα
Laboratory Course on Silicon Sensors 225

Figure 16: Photograph of the setup. The sensor and the readout chip are mounted on a
printed circuit board which is placed into a aluminium box.
counts

90 241 Am
80

70

60

50

40

30

20

10

0
10 20 30 40 50 60 70 80
Channel

Figure 17: Spectrum from a 241 Am source. The source is placed under the printed circuit
board hosting the detector.

and Kβ lines for copper and silver are: Cu (Kα ) 8.03 keV, Cu (Kβ ) 8.90 keV,
Ag (Kα ) 21.99 keV and Ag (Kβ ) 24.94 keV.
Successively, other measurement are taken with the source placed into the
aluminium box on a “bridge” above the sensor. In this case only the peak cor-
responding to the 59.95 keV line is visible. In the following, the peaks obtained
reading out channels 12,36,58,78,102 are shown.
From Figures 18,19 and 20 it can be seen that the energy resolution de-
teriorates increasing the number of the readout pads, due to the increase of
the sensor capacitance. If it assumed that the energy resolution is limited by
226 Elisabetta Crescio et al.

counts
30
counts

graph12 graph12
Entries 9 25 Entries 10
Mean 54.15 Mean 54.77
25
RMS 1.323 RMS 1.998
20
20

15
channel 12 - 1 pad
15 channel 36 - 2 pads
10
10

5 5

0 0
40 45 50 55 60 65 70 75 80 40 45 50 55 60 65 70 75 80
Channel Channel

(a) (b)

Figure 18: Spectrum from a 241 Am source obtained from channel 12, with only one pad
connected (a), and from channel 36 with two pads connected (b).
counts

graph12
counts

graph12
25 Entries 10 11
25 Entries
Mean 54.74
RMS 1.556 Mean 54.78
RMS 2.463
20
20

15 15
channel 58 - 4 pads
channel 78 - 8 pads
10 10

5 5

0 0
40 45 50 55 60 65 70 75 80 40 45 50 55 60 65 70 75 80
Channel Channel

(a) (b)

Figure 19: Spectrum from a 241 Am source obtained from channel 58, with 4 pads connected
(a), and from channel 78 with 8 pads connected (b).
counts

graph12
25 Entries 16
Mean 56
RMS 2.741
20

15 channel 102 - 21 pds

10

0
40 45 50 55 60 65 70 75 80
Channel

Figure 20: Spectrum from a 241 Am source obtained from channel 102, with 21 pads con-
nected.

the electronic noise, then the noise can be calculated knowing the RMS of the
energy distribution:

RM S(eV )
N oise (RM S e−) = (14)
3.6eV
Laboratory Course on Silicon Sensors 227

where 3.6 eV represent the energy required to create an electron-hole pair from
and incident particle.