EG&G ORTEC 118A User manual

100 MIDLAND ROAD. OAK RIDGE, TENNESSEE, 37830
AREA CODE [phone] TWX 810-572-1078
INSTRUCTION
MANUAL
118A
PREAMPLIFIER
•
I ii ^ ^ Li"k is ^ i y il tL' is ^
QUEEN'S UNIVERSITY
EBsCS COMPANY

®
INSTRUCTION MANUAL
118A
PREAMPLIFIER
Serial No.
Purchaser
Date Issued
r ;
INCORPORATED
P. 0. BOX C
OAK RIDGE, TENNESSEE 37830
Telephone [phone]
TWX 810-572-1078
ORTEC IncorporatecJ 1966
Printed In UaS«A«

TABLE OF CONTENTS
Page
WARRANTY
PHOTOGRAPH
1. DESCRIPTION 1-1
2. SPECIFICATIONS 2-1
3. INSTALLATION INSTRUCTIONS 3-1
3.1 Connection to Detector 3-1
3.2 Operation Above lOOOV Detector Bias 3-1
3.3 Connection to a Shaping Main Amplifier 3-1
3.4 Input Power 3-2
3.5 Test Pulse 3-2
4. OPERATING INSTRUCTIONS 4-1
4.1 Detector Bias 4-1
4.2 Linear Output 4-1
5. CIRCUIT DESCRIPTION 5-1
5.1 Charge-Sensitive Loop 5-1
5.2 Voltage Amplifier 5-1
5.3 Cable Driver 5-1
5.4 Performance Data 5-1
6. MAINTENANCE INSTRUCTIONS 6-1
6.1 Testing Performance 6-1
6.2 Suggestions for Troubleshooting 6-3
BLOCK DIAGRAM and SCHEMATIC
118A-0400-B1 118A-0000-S1

STANDARD WARRANTY FOR ORTEC ELECTRONIC INSTRUMENTS
DAMAGE IN TRANSIT
Shipments should be examined immediately upon receipt for evidence of external or con
cealed damage. The carrier making delivery should be notified immediately of any such damage,
since the carrier is normally liable for damage in shipment. Packing materials, v^aybills, and other
such documentation should be preserved in order to establish claims. After such notification to the
carrier, notify ORTEC of the circumstances so that we may assist in damage claims and in providing
replacement equipment when necessary.
WARRANTY
ORTEC warrants its electronic products to be free from defects in workmanship and materials,
other than vacuum tubes and semiconductors, for a period of twelve months from date of ship
ment, provided that the equipment has been used in a proper manner and not subjected to abuse.
Repairs or replacement, at ORTEC option, will be made without charge at the ORTEC factory.
Shipping expense will be to the account of the customer except in cases of defects discovered upon
initial operation. Warranties of vacuum tubes and semiconductors, as made by their manufacturers,
will be extended to our customers only to the extent of the manufacturers' liability to ORTEC.
Specially selected vacuum tubes or semiconductors cannot be warranted. ORTEC reserves the right
to modify the design of its products without incurring responsibility for modification of previously
manufactured units. Since installation conditions are beyond our control, ORTEC does not assume
an^ risks or liabilities associated with the methods of installation, or installation results.
QUALITY CONTROL
Before being approved for shipment, each ORTEC instrument must pass a stringent set of
quality control tests designed to expose any flaws in materials or workmanship. Permanent records
of these tests are maintained for use in warranty repair and as a source of statistical information
for design improvements.
REPAIR SERVICE
ORTEC instruments not in warranty may be returned to the factory for repairs or checkout at
modest expense to the customer. Standard procedure requires that returned instruments pass the
same quality control tests as those used for new production instruments. Please contact the factory
for instructions before shipping equipment.

■■y^L "■
DET. INPUT
mM.
m
PULSE
POWER

WARRANTY VOID
The warranty is voided unless the following precautions are taken:
DO NOT connect a detector, coble, capacitor, or other copocitive device or low
impedance to the DET. INPUT connector, shown in the Front View on page ii of this
manual, unless the detector bias circuitry is COMPLETELY DISCHARGED. (Refer to
explanation below.)
The input transistor wil I be destroyed if the DET. INPUT connector is shorted, i.e.,
by connecting a detector, cable, capacitor, or other capacitive device such as a
voltmeter probe, etc., while the detector bios components are charged. A short
circuit, short-term or continuous, will cause the applied bias voltage (stored on C3)
to be coupled via C3 directly to the input transistor, causing catastrophic breakdown.
If this happens, the only recourse is to replace Q1 and perhaps Q2 also, depending on
the failure mode of Q1 .
To discharge the detector bias circuitry requires that either a low impedance (short cir
cuit preferable) be connected across the DET. BIAS connector, shown in the Rear View
on page ii, for at least one minute, or the bias supply voltage be reduced to zero for a
time adequate to discharge the input coupling capacitor. A shorting cap is provided at
the DET. BIAS connector for the purpose of discharging the bias circuitry and keeping it
discharged when preamp is not in use.
If a variable bias supply is used, merely turning down the voltage control to zero and
leaving it for at least one minute will suffice, since the bias circuitry can discharge
itself through the output impedance of the bios supply.
Sometimes it is necessary to simply disconnect the bias supply, such as when using
batteries for bios. This situation leaves no discharge path, so a path must be provided
by placing a short circuit or low impedance across the DET. BIAS connector on the rear
panel of the unit. (Use shorting cop provided.)
DO NOT SHORT the DET. INPUT connector on the front panel.
This problem has been studied extensively, and any attempt to protect the input tran
sistor by limiting the voltage swing by the use of diodes, zeners, etc., degrades the
resolution to approximately 5.5 keV (Si) and degrades the slope to approximately
0.15 keV/pF.

1-1
ORTEC 118A PREAMPLIFIER
1. DESCRIPTION (See Block Diagram, 118-0400-B)
The ORTEC 118A Preamplifier is intended for use with cooled, low-leakage semi
conductor detectors having relatively low (0 to 100 pF) capacity. In order to
achieve the high performance specified, the detector bios resistor is necessarily
large (2000 megohms), and therefore the voltage drop across this resistor can be
significant with detectors having high leakage.
The 118A is an all-transistor preamplifier using a field-effect transistor in the
input stage. The preamplifier contains a charge-sensitive stage having an open
loop gain greater than 20,000 followed by a voltage amplifier with switch-selectable
gains of 0.5 and 5. The voltage amplifier is followed by a cable driver capable of
driving a 93-ohm cable (series terminated at the preamplifier end) to ±7V.
When the preamp is operated at high counting rotes, there is a possibility that the
input pulses will pile up and cause the preamplifier to saturate. In order to opti
mize the preamplifier counting rote capabilities, a SO-psec clip (RC differentiation)
has been inserted between the charge-sensitive stage and the voltage amplifier.
With the addition of this clip, both the charge-sensitive stage and the voltage ampli
fier saturate at approximately the same level. The counting rate capabilities of the
118A for various energies are shown in the following table:
Input Energy
Count Rate With «1%
(silicon equivalent)
counting losses (cts/sec)
100 KeV
«2 X 10^
1 MeV
«2 X 10^
10 MeV
ft .2 X 104

2-1
2. SPECIFICATIONS
Basis of Warranty: <.03 keV/pF fwhm Ge slope, si .5 keV fwhm Ge at 0 pF
input capacity. Typical performance is shown below:
TYPICAL PERFORMANCE
Noise with 2p
sec RC Main Amp
Ext
Cap.
(pF)
Resolution
keV (Ge)
Resolution
keV (Si)
Eq. Noise Chg.
rms electrons
Relative
Amplitude
Rise Time
(nsec)
0
1.19
1.48
175
1.00
48
5
1.26 1.57
185 1.00
52
10
1.40
1.74
205 1.00
56
15 1.53 1.90
224 1.00
65
25 1.80
2.24
264 1.00
72
50 2.50
3.11
367
1.00
103
100
4.0
4.98
587
.995 168
200
6.9 8.6
1015
.985 280
500
15.0
18.6
2200
.950
580
1000
27.5 34.2
4040
.900 1000
Output pulse shape Rise time as in table above; exponential
fal l with 50-psec time constant
Integral nonlinearity sO.1% for 0-5V with internal series
termination
Temperature coefficient ±0.01% per °C
Detector bias isolation 1000V dc with Microdot connector
in place; 2000V dc without Microdot
connector
1 I

2-2
I
Input open loop gain >20,000
Power required +24V dc at 18 mA, -24V dc at 18 mA;
supplied from ORTEC main amplifier or
from on ORTEC 115 Preamplifier
Power Supply
Cable required 10-foot compatible cable supplied with
preamplifier, type 108-C2
Power connector Ampheno! 17-20090
Saturated output amplitude 7V at end of several hundred feet of
unterminated 93-ohm cable
Output source impedance Adjustable from 50 to 150 ohms
Charge gain 300 mV/MeV in XI0 gain position
(silicon), 365 mV/MeV (germanium);
30 mV/MeV in XI gain position
(silicon), 36.5 mV/MeV (germanium)
Detector, output, and test pulse connectors BNC
Detector bias connector MHV, Gremar 6317
Size 1.8 X 4 X 6 inches (4.5 x 10.2 x 15.3 cm)
Weight Net, 1.5 pounds (0.7 kg); gross,4.0
pounds (1.8 kg)
S
3.
AI V
CM-C
r
0-
O
kf
C
V« ,
•1/
c
A
^7 ,
% ^
.0

3-1
3. INSTALLATION INSTRUCTIONS
3.1 Connection to Detector
A direct connection with shielded coaxial cable should be made between the
detector and the Microdot connector labeled DET. INPUT on the front
panel. DO NOT CONNECT THE DETECTOR TO THE INPUT CONNECTOR
UNTIL THE DETECTOR BIAS VOLTAGE HAS BEEN REDUCED TO ZERO.
FAILURE TO DO THIS WILL CAUSE CATASTROPHIC DAMAGE TO THE INPUT
FIELD EFFECT TRANSISTOR.
The performance of the 118A Preamplifier, like that of all other such low-
noise nuclear amplifiers, is degraded as the capacity at the input of the
amplifier increases; for this reason, it is important that the length of coaxial
cable used between the amplifier and the detector be kept at the minimum
necessary. Also, it is preferable to use 93- or 100-ohm impedance cable rather
than 75- or 50-ohm cable, since the capacity per foot is less for the higher-
impedance cable, Microdot cables and fittings of the series stocked and
supplied by ORTEC are suitable. The cable is 100-ohm impedance,
13 pF per foot. An adapter, ORTEC No, C-16, may be used on the input
Microdot connector to permit use of the BNC cables, (See pages 6-6
and 6-7,)
Once the input cable installation has been made, the electronic noise perform
ance of the preamp can be predicted by calculating the cable capacity from the
above information, adding the capacity expected from the detector, and referring
to the table of typical performance versus input capacity (Section 2).
3.2 Operating Above 1000V Detector Bias
If it is desired to operate the detector at more than 1000 volts, the Microdot
detector input connector must be removed, since it wil l not take more than
1000 volts. The components in the bias circuitry are rated for up to 2000 volts,
and the unit may be operated at up to 2000 volts if some arrangement is made to
get the detector connection inside the preamp with good shielding. This is some
times done by removing the input connector, mounting the preamp rigidly against
the outside of the vacuum wall, and letting the vacuum feedthrough protrude
directly into the inside of the preamp, where connection is made with the wire
that was connected to the input connector.
3.3 Connection to a Shaping Main Amplifier
The preamp can be used to drive long 93-ohm line to a shaping main amplifier,
and is designed to be directly compatible with the ORTEC transistor main ampli
fiers. It can be used with any shaping main amplifier if a power supply is used
to power the preamp ,

3-2
3.4 Input Power
Power for the preamp is supplied through the Amphenol connector (17-20090)
on the rear of the chassis. Power may be supplied by a single 45-volt battery
with a tap at 22.5 volts (the tap is used as ground, providing +22.5V and
- 22.5V; current drain is 18 mA), or any well-filtered ±24V pov/er supply such
as the ORTEC 115 Preamplifier Power Supply.
If the preamp is used with ORTEC transistor main amplifiers, power for the
preamplifier will automatically be supplied to the preamplifier from the tru in
amplifier via the interconnecting cable supplied with the preamplifier.
3.5 Test Pulse
A voltage test pulse can be inserted at the TEST PULSE connector on the rear
of the preamp without the use of an external charge terminator, since the preamp
has a built-in charge terminator. The shape of this voltage pulse must have a
fast rise (less than 10~8 seconds) followed by a slow exponential decay back to
the baseline (2 to 4 x 10"4 seconds). The input amplitude can be set to any
desired level with the knowledge that 92 mV amplitude at the TEST PULSE
connector is equal to approximately 1 MeV energy loss in a silicon detector.
Also, the test pulse can be inserted into the DET. INPUT connector simultaneously
with an operating detector by using an external charge terminator, provided the
charge terminator will withstand the detector bias voltage (s200 volts with
standard ORTEC charge terminator).
NOTE: In most experimental situations the optimum signal-to-noise ratio occurs
with the preamp gain switch in the XI0 position. When this switch is in the XI
position, the signal-to-noise ratio at the preamplifier output is the same as that
of the XI0 position, but the signal level is only l/lO. At this low signal level,
main amplifier noise contribution becomes more significant. The signal-to-noise
ratio at the output of the main amplifier can be degraded by the amount of the
main amp noise contribution and resolution sometimes suffers.

4-1
OPERATING INSTRUCTIONS
4.1 Detector Bias
Detector bias is applied to the preamplifier through the high-voltage MHV
connector (UG-931/U). The detector bios components are rated at 2000V
and connectors are rated at 1000 volts. Inside the preomp, the detector bias
is applied to the DET. INPUT connector through a decoupling network con
sisting of R1 (10-megohm), R2 (2000-megohm), and C1 (0.0022pF) so that
the total resistance in series with the detector bias is 2010 megohms. This
resistance must be taken into account when considering the bias voltage actu
ally applied to a leaky detector. The voltage dropped in this network will
be 2010 volts for each microampere of detector leakage current. It will be
necessary to increase the bias supply output voltage to compensate for this
drop.
EXAMPLE: It is desired to operate a detector at 100 volts bias. Detector
leakage at 100 volts = 0.05 microampere. The drop in the bias
network =(0.05 x 10"®)= (2000 x 10®) = 100.5 volts
Voltage required at bias supply = 100 + 100.5 = 200.5 volts.
4.2 Linear Output
The output of the preomp is a "step" of voltage of 300 mV/MeV with the
gain switch in the X10 position, or 30 mV/MeV in the XI position (silicon
detector equivalents). The dynamic range of the output is 7 volts, either
polarity, when the output is "sending-end" terminated. "Receiving-end"
termination will result in a 3.5-volt dynamic range, but this may be preferable
to sending-end termination. The integral nonlinearity in the preomp is
specified at not greater than 0.1% over the 0 to 10 MeV range.

5- 1
5. CIRCUIT DESCRIPTION (See Schematic Diagram 118-0000-A-S)
5.1 Charge-Sensitive Loop
The charge-sensitive loop consists of five transistors acting as an operatiorKil
amplifier with capacitive feedback. Transistors Q1 and Q2 operate in cascode
and drive Q3, Q4, and Q5 in a low impedance driver configuration for low
output impedance and fast rise time. The rise time of the charge-sensitive
loop output increases as the external input (detector) capacitance is increased.
5.2 Voltage Amplifier
The voltage amplifier is designed for fast rise time (20 nanoseconds in the XI
gain position) so as to faithfully reproduce the pulse from the charge-sensitive
loop. In the XIO gain position the voltage amplifier has a rise time of approx
imately 40 nanoseconds even at low capacitance values. In the XIO gain
position the preamplifier overall rise time does not change until the charge-
sensitive loop rise time approaches 40 nanoseconds, as external capacitance is
increased.
5.2.1 Pole-Zero Cancellation Network
The decay time constant of the output signal from the preamplifier is
deteimined by C12 and the parallel combination of R17, R18, and R36.
It is accurately set at 50 microseconds. R36 C12 is a 1 millisecond time
constant in the proper configuration to provide a "zero" type frequency
response which cancels out the 1 mil lisecond "pole" generated by the
charge sensitive loop feedback time constant. The purpose of this "Pole-
Zero Cancellation" is to obtain a pulse response which has a step rise
with a single 50 microsecond decay time constant back to the baseline
without appreciable undershoot. This will allow accurate pole-zero
cancel lation in the shaping amplifier.
5.3 Cable Driver
The cable driver consists of Q8, Q9, QIO, and Q11 operating in a complementary
Darlington connection. This circuit gives extremely good linearity and an output
impedance of a few ohms. However, 51 ohms is inserted in series with each cir
cuit, so that the minimum output impedance (R35 at 0 ohms) is 51 ohms. The
maximum output impedance is 150 ohms (R31 at ICQ ohms), so that cables in the
ra nge of 50 to 150 ohms can be series (sending-end) terminated.
5.4 Performance Data
Figure 1 shows typical performance characteristics of the preamp on relative
amplitude and resolution (fwhm) as a function of external capacitance on the
input of the preamp for 1 psec, 2psec, 5Hsec, and lOpsec single RC integration
and differentiation time constants. Figure 2 shows rise time as a function of in-

5-2
put capacitance. All data were taken with an ORTEC 410 Linear Amplifier.
Figure 3 shows the detrimental effect of decreasing the bias resistor, R3, to
values less than 2000 megohms. It is conceivable that it may be desired to de
crease the value of R3, e.g., when using the preamp with room temperature
detectors which may have appreciable leakage current. The effect of decreasing
R3 is shown at both zero external capacitance and 55 pF external capacitance.
The detrimental effect of decreasing R3 is less when the external (detector) capaci
tance is appreciable.

100
i±H44=t
RELATIVE AMPLITUDE
RESOLUTION (FWHM) AS A FUNCTION OF
INPUT CAPACITANCE AT VARIOUS SHAPING TIME
CONSTANTS. MODELS 109A & 118A.
H 10
5 M Sec
2 uSec
10 m Sec.
118A
•■m 2
200168
20 40 60 100
INPUT CAPACITANCE (pf)
RESOLUTION (FWHM) VERSUS INPUT CAPACITANCE
200
400 600 1000
Oi
I
CO
Figure 1. Resolution (fwhm) as a Function of Input Capacitance at Various Shaping Time Constants

8 10
Qtirn
200160
20 40 60 80 100 200
INPUT CAPACITANCE (pF)
400 600 800 1000
Oi
Figure 2. Rise Time as a Function of Input Capacitance

700
<
s
ce
ui
(S
>
«
o
I/)
Ui
fl C
600„
400 O
200200
50 100 200 500 1000 2000
BIAS RESISTOR VALUE (MEGOHMS)
5000
Oi
Figure 3. Resolution (fwhm) os a Function of Bias Resistor

6- 1
6. MAINTENANCE INSTRUCTIONS
6.1 Testing Performance
As ordinarily used in a counting or spectroscopy system, the preamp is one
part of a series system involving the source of particles to be analyzed, the
detector, the preamp, the main amplifier, and the pulse height analyzer.
In situations where proper results are not being obtained and tests for proper
performance of the preamp and the other components are indicated, it is
important to realize that rapid and logical testing is possible only when the
individual components are separated from the series system. In proving the
performance of the preamp, this consists of removing it from the system and
dealing with it alone, by providing a known electrical input signal and testing
for proper output signal with an oscilloscope.
6.1.1 Use a voltage pulse in the TEST PULSE jack, as outlined in Section
3.5, or use a pulser with a charge terminator at the DET. INPUT
jack. The polarity of the test pulse signal should be in agreement
with the expected signal input polarity from a detector.
6.1.2 If a suitable input signal has been obtained for the preamp as outlined
in the preceding section, the performance of the instrument may be
checked by observing the pulse waveform at the OUTPUT jack. If an
input signal of 920 mV, corresponding to about 10 MeV, has been
obtained as described above, one can expect an output pulse amplitude
of about 3.0 volts with the gain switch in the XIO position, and 0.3
volt with the gain switch in the XI position.
6.1.3 The noise contribution of the preamplifier may be verified by two basic
methods. In either case, the normal capacity of the detector and
associated cables should be replaced by a capacitor of equal value
connected to the DET. INPUT jack. This is necessary because the
noise contribution of the preamplifier is dependent upon input capacity,
as can be seen from the noise specifications given in Section 2. The
only meaningful statement of the noise level of the preamplifier is one
that relates to the spread caused by the noise in actual spectra. This
can be measured and expressed in terms of the ful l width at half maxi
mum of a monoenergetic signal after passing through the preamplifier
and main amplifier system. The noise performance referenced in Section
2 is stated in these terms, and verification methods wil l be described.
If desired, the preamplifier can be tested with no external capacity on
the DET. INPUT jack, inwhich case the noise width should be approxi
mately that shown for zero external capacity. In any case, the input
jack and capacitors, when used, should be completely shielded electri
cal ly. A wrapping of aluminum foil around the input jack will suffice
for testing at zero capacity, taking care not to short the center conductor
to ground.

6-2
The preamplifier must be tested in conjunction with an associated main
amplifier that provides the required pulse shaping. The typical noise per
formance given in Section 2 is based on main amplifier pulse shaping
consisting of equal RC differentiation and integration of 2-microsecond
time constant. For comparison to these tabulated values, it is preferable
to test the preamplifier under identical pulse shaping conditions. It is
also important to ensure that the noise level of the input stage of the
associated main amplifier does not contribute materially to the total
noise. This is usually no problem provided input attenuators, if any,
on the main amplifier are set for minimum attenuation.
If a multichannel pulse height analyzer is used following the main ampli
fier, testing of the noise performance can be accomplished merely by
using a calibrated test pulse generator with charge terminator, as outlined
in Section 6.1.1 . With only the charge terminator connected to the
DET. INPUT jack, the spread of the pulser peak thus analyzed will be
due only to the electronic noise contribution of the preamplifier and
main amplifier. The analyzer can be calibrated in terms of keV per
channel by observing two different pulser peaks of known energy, and
the fwhm of a peak can be taken directly from the analyzer readout.
It is also possible to determine the noise performance of the preamplifier
by the use of a wide-bandwidth rms ac voltmeter such as the Hewlett-
Packard 400D, reading the main amplifier output noise level and correlating
with the expected pulse amplitude per keV of input signal under the
same conditions. Again, a calibrated test pulse generator is required
for an accurate measurement.
In this method, the preamplifier and main amplifier are set up as they
would be used normal ly, but with a dummy capacitor (or no capacity)
on the DET. INPUT jack, and with the ac voltmeter connected to the
amplifier output. The noise voltage indicated by the meter, designated
Erms, 'S read and noted.
Then, a test pulse of known energy, Ejp (in keV), is applied to the input
jack, and the amplitude of the resulting output pulse, Eq^j. is measured
in volts with an oscilloscope. The noise spread can then be calculated
from the formula
fwhm (keV, Si det.) = 2.66 (Epms) (Ein)
^out
where Epms 's output noise in volts on the 400D meter, E|n is the input
signal in keV particle energy, Eout output signal in volts corresponding
to the above input. The factor 2.66 is the product of two relations:
correction from rms to fwhm (2.35), and correction of the 4000 meter
from sine-wave to white noise (1.13).

6-3
The noise performance of the preamplifier, as measured by the above
methods, should not differ significantly from that given in the specifi
cations in Section 2.
6.1.4 When testing the preamplifier and detector, if the noise performance of
the preamplifier has been verified as outlined in the preceding section,
or is otherwise not suspect, a detector may be tested to some extent by
duplicating the noise performance tests with the detector connected in
place, and with normal operating bias applied. The resulting combined
noise measurement, made either with an analyzer or by the voltmeter
method, indicates the sum in quadrature of the separate noise sources
of the amplifier and the detector. In other words, the total noise is
given by (N^ot)^ = (Njet)^ (Namp)® • Each quantity is expressed
in keV fwhm. The quantity Njet 'S known as the "noise width" of the
detector, and is included as one of the specified parameters of each
ORTEC semiconductor detector. By use of the above equation, with
a knowledge of the noise of the preamplifier, the noise width of the
detector can be determined. The significance of this noise width in
evaluating the detector is subject to interpretation, but general ly the
actual resolution of the detector for protons or electrons will be approxi
mately the same as the noise width; the resolution of the detector for
alpha particles will be poorer than the noise width. The most useful
application of determining the noise width of a detector is in the occasional
monitoring of this quantity to verify that the detector characteristics
have not undergone any significant change during use.
6.2 Suggestions for Troubleshooting
In situations where the preamp is suspected of malfunction, it is important to
isolate the preamp and test it alone, not in a system involving other units such
a sourceof particles to be analyzed, the detector, the preamp, a main amplifier
and subsequent sealers and/or analyzers. Such logical isolation and individual
testing of components will be the most productive approach.
6.2.1 Charge-Sensitive Loop
The function of the preamp is simple and lends itself to relatively easy
scrutiny. The charge-sensitive loop performs a charge-to-voltage
conversion on the input signal. It has an output signal that manifests
itself as a fast rise (~40 nanoseconds at 0 pF external capacitance) step
of voltage whose height is determined by the input charge, followed
by a 400-microsecond decay back to the baseline. This signal can be
observed at the emitter of Q4 while impressing a signal, as described
in Section 3.5. The amplitude of this signal should be 90 mV per MeV
equivalent input signal .