ANALOG FRONT-END ELECTRONICS FOR BEAM POSITION
MEASUREMENT ON THE BEAM HALO MEASUREMENT
*
R. B. Shurter, T. Cote, J. D. Gilpatrick, Los Alamos National Laboratory, Los Alamos, NM 87545,
USA
Abstract
Enhancements have been made to the log-ratio analog
front-end electronics based on the Analog Devices 8307
logarithmic amplifier as used on the LEDA accelerator.
The dynamic range of greater than 85 dB, has been
extended to nearly the full capability of the AD8307 from
the previous design of approximately 65 dB through the
addition of a 350 MHz band-pass filter, careful use of
ground and power plane placement, signal routing, and
power supply bypassing. Additionally, selection of high-
isolation RF switches (55dB) has been an integral part of
a new calibration technique, which is fully described in
another paper submitted to this conference [1]. Provision
has also been made for insertion of a first-stage low-noise
amplifier for using the circuit under low-signal
conditions.
1 BACKGROUND
A BPM signal processor based on the AD8307 was
previously used by us on the Low Energy Demonstration
Accelerator (LEDA), part of the Accelerator Production
of Tritium (APT) project at Los Alamos. Experience with
that system was reported at the 2000 Beam
Instrumentation Workshop [2]. Several of the problems
encountered with that system have been corrected, as well
as new capabilities added, with the resultant system being
presented here. Currently the LEDA accelerator is
configured to conduct a beam halo experiment.
Beam diagnostics include fifteen beam position
monitors in the 52-quadrupole magnet focusing-
defocusing (FODO) lattice, along with wire scanners,
beam-loss monitors and current monitors [7]. The nature
of the experiment places a requirement for additional
dynamic range on the BPM electronics. Beam de-
bunching, which occurs as the beam is transported
through the lattice, reduces the signal power by about 40
dB from the last BPM with respect to the first.
2 CIRCUIT DESCRIPTION
Referring to the block diagram in Figure 1, signals from
the four electrodes of the beam position monitors are
band-pass filtered and input to the AD8307 log amplifier
through a transformer balun, which is required because of
the differential inputs with an impedance of 1 kΩ shunted
by 1.4 pf to common. The circuit layout can accommodate
a Minicircuits Lab fixed attenuator (PAT-X) and/or a
Cougar Components AC566 pre-amplifier [8] installed
between the filter and the transformer. This attenuator-
amplifier combination allows us to adjust the signal high-
end for the log amp’s maximum input of approximately
+17 dBm, thereby obtaining the greatest signal to noise
ratio and dynamic range (approximately 92 dB is
specified for the AD8307.)
The log amp output is then amplified, scaled and offset-
adjusted by an AD8041 amplifier stage for a 0 to
approximately 4.6 V range. The signal is then low-pass
___________________________________________
* Supported by US DOE, Office of Defense Programs, and the Office of
Nuclear Energy, Science and Technology.
Digital Processing
Circuitry
AD9241 A/D
Converter
Xformer/Balun
AD
8307
Log
Amp
AD8041
Op Amp
2MHz Low Pass
Filter
350 MHz
Bandpass Filter
AC566
Preamp
Selectable
Attenuator
Solid-state
Switch
Σ
Summing
Junction
BPM
Signal Sources
T
B
R
L
Calibrator
'Top'
Channel
(of Four Identical
Channels)
From
Bottom, Right and
Left Channel Analog
Circuitry
D/A
AD8041
Op Amp
B
R
L
T
To Bottom, Right and Left Channels
Front Panel
Intensity Monitor
Output
Front Panel
Channel Monitor
Outputs
Four-Way
Splitter
Σ
Figure 1: BPM Signal Processing Block Diagram.
0-7803-7191-7/01/$10.00 ©2001 IEEE. 1378
Proceedings of the 2001 Particle Accelerator Conference, Chicago
filtered to a 200-KHz bandwidth and digitized by an
AD9241 14-bit analog to digital converter. Only 12 bits of
the A/D is used, giving a resolution of 5V/4096 = 1.2
mV/count. The result of one channel’s raw (calibrator
errors not corrected) calibration response, as seen on a
LabVIEW PC display, is given in figure 2.
4000
0
500
1000
1500
2000
2500
3000
3500
90
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
Counts
Relative input power (dB)
Figure 2. Calibration display
3 IMPROVMENTS
The improvements to the previous design include:
• Correction of overlapping digital and analog power
and ground planes. Signal line re-routing to eliminate
crossing gaps in the planes and/or keeping the plane-
signal types coplanar.
• Addition of an RF shield around the RF components.
• Input filter change from low-pass to band-pass. This
resulted in the largest dynamic range gain.
• Impedance matching all input RF traces and careful
attention to shortness of trace length between
components and their bypass capacitors.
• Elimination of one baseband gain stage.
3.1 Input Filter
Our previous design used a low-pass filter at the RF
input. We found that most of the debilitating noise we
were experiencing could be eliminated by using a band-
pass filter custom made by TTE, Inc [3].
3.2 Circuit Layout
Attention was given to the routes return currents
traveled in order to make sure that those currents did not
induce signals in low-signal parts of the circuitry. Because
we were not absolutely certain about this, we provided
jumper pads for tying the analog and digital ground planes
together at various points. Additionally, the recommended
printed circuit layout for the Analog Devices AD9241
A/D converter was followed. The AD9241 product
specifications define how the analog and digital ground
planes are separated under and around the part [4][5].
Large co-planar power and ground planes were used to
provide low-impedance power and return paths, as well as
to increase the capacitive coupling between them for
shunting unwanted signals. Care was taken to make sure
signal lines were not routed between these planes.
Signal traces were never allowed to cross gaps in
ground planes, as this would create current loops, thereby
inducing currents in nearby components [6].
3.3 Dynamic Range improvement
Figure 3. shows a graph of the dynamic range of the
AFE circuit as previously used on the LEDA accelerator.
Figure 4. shows the improved range of the circuit used on
the HALO measurement.
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
0204060
Relative Power (dB)
Figure 3. LEDA BPM AFE Linearity
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
-100-80-60-40-200
Power (dBm)
Figure 4. HALO BPM AFE Linearity
4 CONCLUSION
The Analog Devices AD8307 Logarithmic amplifier’s
superior 92dB dynamic range and 500 MHz bandwidth
makes an excellent beam signal detector as compared to
other techniques, e.g.. synchronous detection, as long as
careful attention is given to following good noise
reduction techniques.
Since the logarithmic non-conformance of this type of
signal processing can introduce deterministic errors of up
to +/-0.5 dB, calibration is used to further enhance
accuracy [9].
Our experience with the BPM VXI processing
electronics modules has shown them to be highly reliable.
The few failures we have experienced have been due to
mis-soldered components, mother-board problems, and in
two cases, defective VXI crates.
1379
Proceedings of the 2001 Particle Accelerator Conference, Chicago
Figure 5. HALO BPM AFE Circuit Board
Figure 6. HALO BPM VXI Module withAFE on
Motherboard
5 ACKNOWLEDGMENT
The authors wish to thank John Power for his initial
work on this measurement design, as well as his
continued help.
6 REFERENCES
[1] D. Barr, J. D. Gilpatrick, R. Shurter, “Upgrade to
Initial BPM Electronics Module and Beamline
Components for Calibration of the LEDA Beam
Position Measurements,” published at this conference.
[2] R.B. Shurter, J. D. Gilpatrick, J. Power, “BPM Analog
Front-End Electronics Based on the AD8307 Log
Amplifier,” 9
th
Beam Instrumentation Workshop,
Cambridge, MA, May 8-11, 2000.
[3] TTE, Inc., http://www.tte.com/
. Filter specifications-
Model number: K4034-350M-20M-50-484, 3-pole
chebyshev, F
0
: 350 MHz, Insertion loss @ F
0
: 5dB
max., Rejection (typical): -40 dB @ 270MHz and
360MHz, Impedance: 50 ohms.
[4] Analog Devices, Inc., www.analogdevices.com
AD9241 specification
[5] Analog Devices, Inc., AN-404 Application note.
[6] T. Van Doren, “Circuit Board Layout to Reduce
Electro-magnetic Emission and Susceptibility,
National Technological University live video course
#MC00032802, March 28, 2000.
[7] J. D. Gilpatrick, et al., "Experience with the Low
Energy Demonstration Accelerator (LEDA) Halo
Experiment Beam Instrumentation,” published at this
conference.
[8] Cougar Components, Inc.,
http://www.cougarcorp.com/
[9] D. Barr, J. D. Gilpatrick, R. Shurter, “Upgrade to
Initial BPM Electronics Module and Beamline
Components for Calibration of the LEDA Beam
Position Measurements,” published at this conference.
1380
Proceedings of the 2001 Particle Accelerator Conference, Chicago