Download 417_1.PDF

Spallation Neutron Source Beam Current
Monitor Electronics1
M. Kesselman, W. C. Dawson
Brookhaven National Laboratory
Abstract. This paper will discuss the present electronics design for the beam current monitor
system to be used throughout the Spallation Neutron Source (SNS) under construction at Oak Ridge
National Laboratory. The beam is composed of a micro-pulse structure due to the 402.5MHz RF,
and is chopped into mini-pulses of 645ns duration with a 300ns gap, providing a macro-pulse of
1060 mini-pulses repeating at a 60Hz rate. Ring beam current will vary from about 15ma peak
during studies, to about 50Amps peak (design to 100 amps). A digital approach to droop
compensation has been implemented and initial test results presented.
BACKGROUND
This is the fourth1'2'3 paper presented on this subject. The earlier papers describe the
decision process to attempt to use identical electronics throughout to provide a
compatible approach to system diagnostics. A PC based instrument design philosophy
was adopted for the diagnostics wherever it would apply, and a compromise in
transformer frequency response, droop, and standardization resulted in a decision to
use Bergoz® FCT (Fast Current) transformers. Individual areas within the SNS will
have current transformers that suit the dimensional requirements, while maintaining
electrical performance compatible with required performance and the system
electronics. All electronics would be identical and could be placed anywhere in the
system. Therefore, the electronics has been designed with flexibility in its
configuration.
CHALLENGES
There were a number of major challenges to be addressed:
•Measurement of chopper characteristics vs. Ring turn by turn current.
•Goal of a single design to minimize design cost, and provide interchangeability
•Large dynamic range in Ring and RTBT
•Baseline restoration and droop compensation
•Integration accuracy vs. sampling rate
•Noise, response characteristics, filtering and digitizing aliasing
•Testing
•Calibration
1
Work performed under the auspices of the U.S. Department of Energy
CP648, Beam Instrumentation Workshop 2002: Tenth Workshop, edited by G. A. Smith and T. Russo
2002 American Institute of Physics 0-7354-0103-9
417
FCT
14
14 Bit
Bit 60MSa/S
60MSa/S
INPUT
INPUTSSIGNAL
IGNAL
CONDITIONING
CONDI1
ADJ.
GAIN
ADJ. G/JN
LPF
TPF
fc(X^
'x
ADC
Anr
,,
'
BASELINE
BASELINE
RESTORATION
RESTORATION
DROOP
DROOP
COMPENSATION
COMPENSATION
CURRENT
CURRENT
........................................................................
BEAM
i
CIRCULAR
,AR
BUFFER
BUFFEIi
INTEGRATOR
INTEGRATOR
accumulator
accumulator
>
OUTPUT
OUTPUT
——». DATA
DATA
FORMATTER
FORMATTER
HISTORY
HISTORY
FF rev
iev
CURRENT
CURRENT
DATA
DATA
MINI-BUNCH CHARGEE
Frev /N
INTEGRATOR
Frev/N
INTEGRATOR
accumulator
————*
accumulator
HEBT
HEBT ONLY
ONLY
MACRO-BUNCH
MACRO-BUNCH CHARGE
CHARGE (HEBT)
(HEBT)
DIGITAL
DIGITAL PROCESSING
PROCESSING
88 Bit
Bit1GSa/S
IGSa/S
—1»
INPUT
INPUT SIGNAL
SIGNAL
CONDITIONING
CONDITIONING
ADJ.
ADJ. GAIN
GAIN
^/ ADC
Anr
x
L—
i>
MINI-BUNCH
MINI-BUNCH DATA
DATA ACQ.
ACQ. SYSTEM
S\ STEM
CIRCULAR
CIRCULAR
BUFFER
BUFFER
-J
I
FIGURE
FIGURE 1.
1. Block
Block diagram
diagram of
of BCM
BCM electronics.
electronics. This
This shows
shows two
two digitizing
digitizing systems.
systems. The
The faster
faster
digitizer
digitizer is
is aa separate
separate system
system used
used for
for analysis
analysis of
of chopper
chopper characteristics
characteristics and
and is
is intended
intended for
for short
short time
time
samples.
samples. The
The slower
slower speed
speed digitizer
digitizer is
is intended
intended for
for general
general current
current monitoring
monitoring and
and has
has aa reduced
reduced
bandwidth.
bandwidth. Baseline
Baseline restoration
restoration and
and droop
droop compensation
compensation are
are accomplished
accomplished digitally
digitally in
in the
the slower
slower
system.
system.
DETAILS
DETAILS
The
The selection
selection of
of the
the FCT
FCT transformer
transformer permits
permits high
high speed
speed response,
response, while
while allowing
allowing
droop
be compensated
compensated digitally.
droop to
to be
digitally. The
The electronics
electronics will
will provide
provide aa broad-band
broad-band output
output
suitable
the chopper
chopper characteristics.
jumper selected
suitable for
for observing
observing the
characteristics. Provisions
Provisions for
for jumper
selected
configurations
configurations for
for the
the lower
lower current
current MEBT,
MEET, Linac,
Linac, and
and HEBT
HEBT and
and the
the higher
higher current
current
Ring
and
RTBT
allow
the
same
electronics
to
be
used
throughout
the
SNS.
Ring and RTBT allow the same electronics to be used throughout the SNS. The
The large
large
dynamic
dynamic range
range in
in the
the Ring
Ring and
and RTBT
RTBT is
is handled
handled by
by employing
employing different
different gain
gain paths
paths
that
that are
are selected
selected by
by switchable
switchable amplifiers.
amplifiers. These
These amplifiers
amplifiers (OPA680
(OPA680 series)
series) switch
switch in
in
less
than
100ns,
allowing
them
to
be
switched
during
the
“gap”
time
in
the
Ring.
less than 100ns, allowing them to be switched during the "gap" time in the Ring. The
The
large
voltages expected
large voltages
expected (25
(25 to
to 50
50 volts)
volts) are
are sufficient
sufficient to
to cause
cause sensitive
sensitive amplifiers
amplifiers to
to
fail.
Therefore,
the
use
of
protected
amplifiers
is
necessary
in
the
Ring
and
fail. Therefore, the use of protected amplifiers is necessary in the Ring and RTBT.
RTBT. A
A
system
system block
block diagram
diagram is
is shown
shown in
in figure
figure 1.
1. The
The input
input signal
signal conditioning
conditioning circuitry
circuitry
will
paths. The
will separate
separate the
the signal
signal into
into two
two paths.
The broadband
broadband path
path will
will be
be sent
sent to
to aa separate
separate
high-speed
digitizer
capable
of
analyzing
the
chopper
characteristics.
The
high-speed digitizer capable of analyzing the chopper characteristics. The second
second path
path
will
will be
be channeled
channeled to
to aa signal
signal processing
processing system
system capable
capable of
of amplifying
amplifying the
the signal
signal
adequately
adequately for
for digitization
digitization by
by aa 14
14 bit
bit 68MSa/s
68MSa/s digitizer.
digitizer. This
This digitizer
digitizer is
is
).
synchronized
synchronized with
with the
the revolution
revolution frequency
frequency of
of the
the Ring
Ring (64xF
(64xFrev
).
rev
418
RESISTIVE
MATCHING
100 :50 OHMS
WIDE-BAND OUTPUT
AD600
0dB to +40dB
7.5mV
to
30 mV
+35dB
50 Ohm
Resistor
-6dB
Gaussian
FILTER
5 Pole
0.211 V for 15mA
7MHz LP
0.846 V for 60mA
29dB Gain
SINGLE PRE-SET GAIN
1.05MHz rf
AD8138
Diff. Out
+
+0dB
-
ADC
AD6644
14 Bit
+/- 1.1V
0.21 V for 15mA
0.85 V for 60mA
14 Bits
DATA
ANTI-ALIAS
FILTER
DIGITAL INTERFACE
DAC
GAIN
8 Bits
8 Bits
FRONT END
REG.
1:50 FCT
15mA to 60 mA
40MHz or 60MHz Clock (continuous)
DIGITIZER
FIGURE 2. Block diagram of BCM electronics used for the
the MEET,
MEBT, Linac
Linac and
and HEBT.
HEBT. AA lOOMhz
100Mhz
bandwidth output is provided
provided for
for monitoring
monitoring purposes.
purposes. Jumper
Jumper selection
selection configures
configures this
this processing
processing
configuration.
configuration.
MEET,
MEBT, Linac, and HEBT Analog System
A high gain configuration
configuration is jumper
jumper selected
selected to
to accommodate
accommodate the
the lower
lower currents
currents
expected in
in the Front End, MEET,
expected
MEBT, Linac and HEBT, and
and is
is shown
shown in
in figure
figure 2.
2. AA
variable gain
variable
gain controlled
controlled amplifier
amplifier (AD600)
(AD600) is
is used
used to
to amplify
amplify signals
signals in
in the
the 00 to
to 25mV
25mV
range (for
(for 0 to 50ma) to a voltage range
range of
of 0 to
to 500mV
500mV for
for the
the ADC.
ADC. AA factor
factor of
of two
two
in headroom
headroom has
has been
in
been reserved
reserved for
for current
current peaking.
peaking. The
The wideband
wideband transformer
transformer signal
signal
is buffered
buffered by
by an
an amplifier
amplifier with
with bandwidth
bandwidth >100MHz
>100MHz (OPA3680).
(OPA3680).
is
The Ring
The
Ring rise
rise time
time is
is expected
expected to
to deteriorate
deteriorate to
to about
about 50ns.
50ns. This
This requires
requires aa 7MHz
7MHz
bandwidth.
Increasing
the
bandwidth
increases
the
noise
and
the
resolution
bandwidth. Increasing the bandwidth increases the noise and the resolution degrades.
degrades.
Therefore, aa 55 pole
Therefore,
pole -- 7MHz
7MHz Gaussian
Gaussian filter
filter was
was chosen
chosen to
to minimize
minimize overshoot,
overshoot, and
and
provide
significant
filtering
for
aliasing
considerations.
This
filter
will
provide significant filtering for aliasing considerations. This filter will provide
provide about
about
42dB attenuation
attenuation at
at 34MHz.
34MHz. Additional
34dB by
42dB
Additional attenuation
attenuation of
of 34dB
by aa 55 pole
pole -– 17MHz,
17MHz,
O.OldB,
Chebyshev
filter
and
two
additional
amplifier
stages
limited
0.01dB, Chebyshev filter and two additional amplifier stages limited to
to 34MHz
34MHz
provides
attenuation at
at the
the Nyquist
Nyquist frequency.
provides >80dB
>80dB attenuation
frequency.
Ring
Ring and
and RTBT
RTBT Analog
Analog System
System
To accommodate
accommodate the
the large
To
large dynamic
dynamic range
range of
of the
the Ring
Ring and
and RTBT,
RTBT, the
the configuration
configuration
of figure
figure 33 is
of
is employed
employed by
by aa jumper
jumper selection.
selection. This
This configuration
configuration employs
employs protected
protected
419
high
stages
and
number
of
lower
gain
paths that
that are
are digitally
digitally selected
selected to
to
high gain
gain stages
stages and
and aaa number
number of
of lower
lower gain
gain paths
paths
that
are
digitally
selected
to
establish
a
variable
gain
capability.
The
paths
are
switched
by
fast
paths are
are switched
switched by
by fast
fast switching
switching
establish a variable
variable gain
gain capability.
capability. The
The paths
switching
amplifiers.
Switching
during
“gap”
time
will
allow
for
no loss
loss of
of turns.
turns.
amplifiers.Switching
Switchingduring
during“gap”
"gap"time
timewill
will allow
allow for
for no
no
loss
of
turns.
RESISTIVE
RESISTIVE
MATCHING
MATCHING
RR :50
:50 OHMS
OHMS
WIDE-BAND
WIDE-BAND
OUTPUT
WIDE-BANDOUTPUT
OUTPUT
GAIN
GAINSET
SET
88Bits
Bits
DAC
DAC
7.5mV
7.5mV
to
to
50
50 V
V
PROTECTED
PROTECTED
PROTECTED
PRE-AMPS
PRE-AMPS
PRE-AMPS
WITH
WITH
WITH
DIFFERENT
DIFFERENT
DIFFERENT
SELECTED
SELECTED
SELECTED
ATTENUATION
ATTENUATION
ATTENUATION
AND
AND
AND
TWO
TWOSTATE
STATE
TWO
STATE
OUTPUT
OUTPUT
OUTPUT
88Bits
Bits
REG.
REG.
Zi=50
Zi=50
GAIN
GAIN
1:50
1:50 FCT
FCT
15mA
15mA to
to 100
100 A
A
1.05MHz
1.05MHzrfrf
Amp
Ampselect
select
50
50Ohm
Ohm
Resistor
Resistor
-6dB
-6dB
FILTERs
FILTERs
55Pole
Pole
7MHz
7MHzLP
LP 0.211 V for 15mA
0.211 V for 15mA
55Pole
Pole
0.846
0.846VVfor
for60mA
60mA
17MHz
17MHz
TO
TO ADC
ADC
BUFFER
BUFFERAMP
AMP
BUFFERED
BUFFERED
OUTPUT
OUTPUT
FIGURE
Ring
FIGURE3.
3. This
Thisshows
showsthe
thegeneral
generalblock
the jumper
jumper
selected
configuration
for
the
FIGURE
3.
This
shows
the
general
block diagram
diagram for
for the
jumper selected
selectedconfiguration
configurationfor
forthe
theRing
Ring
and
andRTBT
RTBTBCM
BCMelectronics.
electronics. A
Aprotected
protectedamplifier
amplifier system
system handles
the high
high
voltage
levels
expected
inin
and
RTBT
BCM
electronics.
A
protected
amplifier
system
handles the
highvoltage
voltagelevels
levelsexpected
expectedin
the
theRing
Ringand
andRTBT.
RTBT. Different
Different amplifier
paths are
digitally to
to
permit
handling
the
1000:1
the
Ring
and
RTBT.
Different
amplifier gain
gain paths
are selected
selected digitally
topermit
permithandling
handlingthe
the1000:1
1000:1
dynamic
range.
dynamic range.
To monitor amp
To monitor amp
R10
R10
150
150
R1
R1
100
100
R3
R3
0
0
52.3
52.3
R6
R6
866
866
33
R5
R5
866
866
2
2 -
1.96K
1.96K
R7
D1
R7
2D1
161
2
161
0
0
0
0
0
0
1
1
BAS70-04
BAS70-04
OUT
OUT
V-V-
150:150 -3dB
150:150 -3dB
R4
R4
U1
U1
3
3 +
+
R2
R2
7
V+ 7
V+
3.92K
3.92K
0
0
150
150
To other attenuators
To other attenuators
-
44
R11
R11
6
6
OPA620/BB
OPA620/BB
R8
R8
100
100
R9
R9
100
100
0
0
0
0
For
Forthe
theattenuator:
attenuator:
For
the
attenuator:
/
Z
=2*(∆Z
ratio=Pi/P0o
∆Z
AZj/
Z
2*(AZ
ZLL)/[2*N+(N-1)*
)/[2*N+(N-1)* (∆Z
(AZLL// Z
ZL)]; N= power ratio=Pi/P
i=
i
i
Z
∆Zi/ Zi=2*(∆ZLLL/// Z
L)/[2*N+(N-1)* (∆ZL/ ZL)]; N= power ratio=Pi/Po
Z
=161||Zii ;;;Z
Zi{ ===V/I
V/I===V/(V-V
V/(V-Vdd)/1.96K
)/1.96K ~- 1.96K*(1+
1.96K*(1+ V
Vdd/V)
/V)
ZZLLL=161||Z
=161||Z
Z
V/I
V/(V-V
i
i
d)/1.96K ~ 1.96K*(1+ Vd/V)
For
ForV
Vdd/V=1/50=0.02;
/V=l/50=0.02; ∆Z
AZLL=149.000648-148.7788
=149.000648-148.7788 ==
= .2218
.2218 and
and
(AZLL///ZZ
)=0.001491
LL)=0.001491
/V=1/50=0.02;
∆Z
=149.000648-148.7788
.2218
and(∆Z
(∆Z
For
V
d
L
L ZL)=0.001491
For
the
3dB
pad
N=2
and
∆Z
/
Z
=
.074%
For
the
3dB
pad
N=2
and
AZ/
Z
.074%
i=
i
i
For the 3dB pad N=2 and ∆Z / Z = .074%
i
i
FIGURE4.
4. A
Aprotected
protectedamplifier
amplifier must
must both
bothprotect
protect the
the amplifier
amplifier against
against high
high
input
signal
voltage
FIGURE
FIGURE
4.
A
protected
amplifier
must
both
protect
the
amplifier
against
highinput
inputsignal
signalvoltage
voltage
aswell
wellas
asassure
assurethat
thatinput
input impedance
impedance isis
is maintained
maintained constant.
constant.
as
as
well
as
assure
that
input
impedance
maintained
constant.
420
Protected Amplifier
Protected Amplifier
Signal levels in the Ring can get as high as 25 Volts for 50 amps. If one considers
doubling
forinheadroom,
thegetsignals
clearly
large
for If
theone
amplifiers
Signalthis
levels
the Ring can
as highare
as 25
Voltstoo
for 50
amps.
considersto
handle.
the high gain
paths must
be protected.
showntoin
doublingTherefore,
this for headroom,
the signals
are clearly
too largeThe
for amplifier
the amplifiers
figure
4 provides
both
a near
constant
input
impedance.
handle.
Therefore,
theprotection
high gain and
pathsassures
must be
protected.
The
amplifier
shown The
in
input
vary by and
moreassures
than 0.1%
assure proper
accuracy. This
figureimpedance
4 providesmust
bothnot
protection
a neartoconstant
input impedance.
The is
achieved
by providing
of more
1K Ohm
before
the protection
input impedance
must an
notinput
vary resistance
by more than
0.1% than
to assure
proper
accuracy.
This is
achieved
inputdistortion
resistancefor
of high
more amplitude
than IK Ohm
before
the protection
diodes.
Inby
soproviding
doing theandiode
signals
is not
reflected to
diodes.
In sopaths,
doing however,
the diode noise
distortion
for high amplitude
signals
is notinput
reflected
to
the
other gain
is increased.
One of the
150 Ohm
resistors
the other gain
noise
is increased.
One of is
theselected
150 Ohm
input
represents
a 150paths,
Ohmhowever,
attenuator
chain.
The attenuation
such
thatresistors
after the
represents
150 Ohm attenuator
The attenuation
is selected
suchinput
that signal
after the
first
stage ofa attenuation
the signalchain.
is maintained
below +/2.0 V (max
for
first
stage of attenuation the signal is maintained below +/- 2.0 V (max input signal for
the
AD600).
theAD600).
Adjustable Gain Stage
Adjustable Gain Stage
The gain changing configuration is shown in figure 5. The attenuation provided is
gain The
changing
configuration
is shown
figure 5.
The attenuation
provided is
also The
shown.
amplifiers
are part of
a tripleinop-amp
(OPA3680).
This amplifier
is a
also
shown.
The
amplifiers
are
part
of
a
triple
op-amp
(OPA3680).
This
amplifier
is a
voltage feedback type amplifier with a digital control input permitting it to switch
voltage feedback type amplifier with a digital control input permitting it to switch
“off” within 100ns. The amplifiers of figure 5 are all of this type. The switched
"off within 100ns. The amplifiers of figure 5 are all of this type. The switched
amplifier is configured as a +2 gain amplifier, providing no signal punch through
amplifier is configured as a +2 gain amplifier, providing no signal punch through
when switched “off”. The summer, adds the outputs of each path (all paths “off”
when switched "off. The summer, adds the outputs of each path (all paths "off
except
has been
been established
established atat 0.5%
0.5% ofoffull
fullscale,
scale,
except one).
one). The
The system
system resolution
resolution has
allowing
4
gains.
A
table
of
gains
and
expected
resolution
is
shown
in
Table
1.
allowing 4 gains. A table of gains and expected resolution is shown in Table 1.
150:50
MONITOR
24.5 dB
8.1dB
G=67
G=7.2
0V-35V
150:150
-3dB
0V-50V
0-100A
12 dB
G=4
Protected
T1-T3
29 dB
G=28.2
150:50
-28.2dB
G=1.09
0-1.95V
T4-T26
G=0.039
100:100
-6.25dB
G=0.019
SWITCH
AMP
8.35 dB
G=2.615
G=0.102
T27-T276
0 dB
G=1
G=0.019
0-0.95V
T277-T1060
FIGURE 5.
5. Gain
Gain Changing
FIGURE
Changing configuration
configuration
421
SUM
TABLE 1. Gain Switching
Turn
Beam
Est.
Current
Input
Signal
1 test
15mA
7.5mV
1
4
5
26
27
276
277
1060
38mA
152mA
190mA
988mA
1.026A
10.488A
10.526A
40.28A
19mV
76mV
95mV
0.494V
0.513V
5.244V
5.263V
20.14V
Gain
67
7.2
7.2
1.09
1.09
0.102
0.102
0.019
0.019
Est.
Output
Noise
0.65mV
0.65mV
0.65mV
0.485mV
0.485mV
0.246mV
0.246mV
0.246mV
0.246mV
Output
Voltage
Resolution
0.5
0.136
0.547
0.103
0.538
0.052
0.535
0.1
0.382
0.13%
0.47%
0.12%
0.47%
0.09%
0.47%
0.046%
0.246%
0.064%
Digital control of gain is accomplished via separate DACs to set gain voltage levels
on the AD600 variable gain amplifiers, and a gain storage register that establishes
which of the gain paths is made active. The gain storage register is updated by a
sequence generator on the digital interface section. The gain register information is
stored in a FIFO, along with data to tag each data point with a gain path.
ADC Driver Stage
The driver for the ADC has been selected as per the manufacturer's
recommendation. An ADS 138 wide-band differential ADC drive amplifier is
employed to shift the reference to match the reference of the AD6645. The summer
feeds the 5 pole Gaussian filter as described earlier. This is buffered by an amplifier
and fed to the 17MHz Chebyshev filter.
Digital Processing
The key to the system is the digital processing. This permits us to use transformers
that have a very wide bandwidth, and compensate for the droop. The digitized data
are transferred to a FIFO for DMA transfer to the PC. A Lab VIEW® program
processes the data and interfaces the PC with the network.
The DC offset is calculated by averaging points prior to the arrival of the beam.
This is subtracted from the data to provide a data set that has been corrected to a zero
baseline. The data is then compensated for droop with a digital IIR filter algorithm
that cancels the transformer low frequency pole and establishes a new low frequency
pole of 1 rad/sec. To cancel the transformer pole, it is necessary to calculate the droop
time constant. This is accomplished by providing a calibration pulse and computing
the exponential time constant during the transformer recovery time. The sensitivity of
resulting compensated droop to errors in this calculation requires a good estimate. An
analysis of this indicates a sensitivity of -0.6%/% (error in droop/error in transformer
time constant). It is, therefore, necessary to average many data points or calculations.
422
The sensitivity of the droop to sampling time has a similar sensitivity, +0.6%/% (error
sensitivityinofsampling
the droop
to sampling time has a similar sensitivity, +0.6%/% (error
inThe
droop/error
time).
in droop/error in sampling time).
The droop compensation formula is:
The droop compensation formula is:
Y(n)={1/(2/T+1/τ2)}{y(n-1) (2/T-1/τ2) + x(n) (2/T +1/τ1) + x(n-1) (-2/T +1/τ1)}
Y(n)={ l/(2/T+l/T2)} {y(n-l) (2/T-l/T2) + x(n) (2/T +I/TI) + x(n-l) (-2/T +I/TI)}
Where: T is the sampling period, 1/τ1 is the transformer lower cut-off radian
Where: T is the sampling period, I/TI is the transformer lower cut-off radian
frequency,
2 is the desired new transformer lower cut-off radian frequency.
frequency, and
and1/τ
I/TZ
is the desired new transformer lower cut-off radian frequency.
The
and filtered
filtered for
for aacomfort
comfortdisplay.
display.
Thedata
dataare
areintegrated
integrated to
to determine
determine total
total charge
charge and
An
analysis
of
integration
errors
due
to
insufficient
samples
indicates
a
sampling
An analysis of integration errors due to insufficient samples indicates a sampling
frequency
0.1% error
error in
in the
the integral.
integral. This
This
frequency of
of more
more than
than 25MSa/s
25MSa/s is
is required
required for
for aa 0.1%
analysis
was
carried
out
using
both
a
simple
sum
and
a
Simpson’s
rule
algorithm.
analysis was carried out using both a simple sum and a Simpson's rule algorithm. ItItisis
interesting
at 64MSa/s,
64MSa/s, and
and differ
differ only
only slightly
slightly
interestingto
to note
note that
that the
the two
two methods
methods converge
converge at
atat lower
sampling
rates.
Therefore,
there
is
little
advantage
to
using
the
more
lower sampling rates. Therefore, there is little advantage to using the more
computer
intensive
Simpson’s
rule.
An
example
of
a
simulated
beam
processed
by
computer intensive Simpson's rule. An example of a simulated beam processed by
this
software
is
shown
in
figure
6.
this software is shown in figure 6.
FIGURE6.6. Screen
Screendump
dump of
of aa simulated
simulated beam
beam processed
processed by
FIGURE
by the
the BCM
BCM electronics.
electronics. The
Theupper
upperleft
left
graph
is
raw
data
showing
a
simulated
1ms,
645ns
pulse
with
a
945ns
period,
pulse
train.
Droop
graph is raw data showing a simulated 1ms, 645ns pulse with a 945ns period, pulse train. Droopisis
obvious,and
andshown
showncompensated
compensated in
in the
the graph
graph below.
obvious,
below. The
The filtered
filtered comfort
comfortdisplay
displayisisshown
shownwith
with256
256
points
in
the
lower
right,
and
the
integrated
charged
particle
count
calculation
shown
above
points in the lower right, and the integrated charged particle count calculation shown aboveit.it.
423
Testing
A test apparatus was constructed using an 8 inch 50 Ohm coax line. The outer
conductor was cut, insulated, and a shroud built to allow the transformer to measure
the current in the center conductor. This was found to have a single resonance near
500MHz and provided a good 50 Ohm match for excellent current transient response
measurements.
Calibration
An isolated dual, current output, DAC (DAC2902) provides the fundamental source
for the calibrator. This is a fast current output device that settles quickly, and can
deliver up to 20mA. The calibrator is isolated to avoid ground loops, and is AC
terminated to back terminate the calibration winding, while allowing a DC current
previously measured by an accurate DMM to flow into the winding. To simulate
larger currents for the Ring, an additional current amplifier will be used.
ACKNOWLEDGMENTS
The authors would like to acknowledge Chris Degen for his assistance with the
Lab VIEW software, and Richard Witkover and Julian Bergoz for their technical
assistance during system development.
REFERENCES
1. Kesselman, M. et-al., "SNS Project-Wide Beam Current Monitors", BIW 2000, May 8-11, 2000,
Cambridge MA.
2. Kesselman, M., et-al. "SNS Project-Wide Beam Current Monitors", EPAC 2000, Vienna, Austria,
June 26-30, 2000
3. Kesselman, M., "Spallation Neutron Source Beam Current Monitor Electronics", PAC 2001,
Chicago, II., June 18-22.
424