Pulse Shaper for a Particle Detector
October 27 2011
Varun Gandhi
Supervised Learning Project report, Fall 2014
Guide: Prof. Pradeep Sarin
1
Certificate
Department of Physics
Indian Institute of Technology, Bombay
This is to certify that this Supervised Learning Project titled Pulse Shaper for a Particle Detector
by Varun Gandhi (120260003) was prepared under my guidance.
Prof. Pradeep Sarin
October 27, 2014
2
Declaration
I, Varun Gandhi, Roll No. 120260003 understand that plagiarism is defined as any one or the
combination of the following:
1. Uncredited verbatim copying of individual sentences, paragraphs or illustrations (such as graphs,
diagrams, etc.) from any source, published or unpublished, including the internet.
2. Uncredited improper paraphrasing of pages or paragraphs (changing a few words or phrases, or
rearranging the original sentence order)
3. Credited verbatim copying of a major portion of a paper (or thesis chapter) without clear
delineation of who did or wrote what. (Source: IEEE, The Institute, Dec. 2004)
I have made sure that all the ideas, expressions, graphs, diagrams, etc., that are not a result of
my work, are properly credited. Long phrases or sentences that had to be used verbatim from
published literature have been clearly identified using quotation marks.
I affirm that no portion of my work can be considered as plagiarism and I take full responsibility
if such a complaint occurs. I understand fully well that the guide of the seminar report may not
be in a position to check for the possibility of such incidences of plagiarism in this body of work.
Signature:
Name: Varun Gandhi
Roll No.: 120260003
Date: October 27, 2014
3
Table of contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Importance of Particle detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
What is a Pulse Shaper? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Necessity for a separate Pulse Shaping circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Circuit design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Implementation .. .. .. .. .. .. ... .. .. .. .. .. .. .. ... .. .. .. .. .. .. ... . 9
Results and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4
Introduction
Importance of Particle detectors
Measurement of various properties of high-energy particles produced in accelerators, nuclear decay
processes and cosmic radiation is of primary importance in several fields such as experimental
particle physics and nuclear physics. The variety of existing particle detector designs represents
the experimentalist’s varied requirements; energy, momentum, charge etc. are quantities that are
used to identify the reaction products amongst several possible different particle species.
Integrating a detector with an electronic circuit is absolutely essential; the data, in any application
of reasonable size, can then be processed, used and stored in a methodical fashion given the vast
literature on instrumentation and electronic circuit design. This circuit (or multitudes of them)
may then be interfaced with a computer which would provide sophisticated tools for data analysis.
5
What is a Pulse Shaper?
An ideal pulse shaping circuit would convert a narrow pulse (from a sensor) to a broader output
with a rounded maximum, whose maximum height is directly related to the peak height of the
input funciton. A realistic pulse shaper has to balance the following factors:
• Ensure that the shape of the output is not affected by changing the characteristics of the
input within a certain pre-specification.
• Ensure that the output returns to zero before another pulse is received. Otherwise pulses
would pile-up in amplitude, leading to incorrect readings.
• Ensure that the output pulse is wide enough to match the measurement time
• Ensure that noise is kept to a minimum by (ideally) not increasing the bandwith too much.
(Image source: Spieler H., Analog and Digital Electronics for Detectors)
Thus, the problem statement can be framed as: to design a semi-gaussian pulse shaper which
satisfies the criteria of low noise, no amplitude pile-up, with an overall time constant of ∼10µs (for
the given DAQ).
6
Necessity for a separate Pulse Shaping circuit
A typical detector consists of the the following setup
Trigger
Out
Pre-amp DAQ
Shaper
The task of the pre-amplifier circuit is to simply increase the order of magnitude of the signal while
retaining its shape without distortion; reducing noise in the pre-amplifier design is important.
The shaper makes the shape of the input pulse to be nearly Gaussian (in case of semi-Gaussian
pulse-shaping). The trigger is supposed to send a signal to the Data Acquisition system (DAQ) so
that it “knows” when to sample the shaper’s output. The DAQ’s task is to correctly sample the
shaper’s output at the appropriate time (determined by the trigger output and pulse shaper’s time
constant(s)) and then process the signals digitally.
While it is possible to simply use a pre-amp connected to a Data Acquisition system (DAQ)
directly, there are many advantages to using an additional step with a pulse shaper:
• The shaper’s time constant now determines the rate at which the DAQ should sample its
input. In the case that the time constant of the signal received by the pre-amplifier is very
short, one would require a very fast DAQ to directly read the signal. This can be avoided
when a shaper is used.
• The pulse shaper can also add in additional stages of amplification in the process of shaping.
This reduces the gain requirement on the pre-amplifier design, permitting larger bandwidth.
7
Circuit design
The basic idea of the circuit is to have three components
1. A passive differentiator : Creates an exponential curve from a rising edge
2. An inverting amplifier : Allows resistors to be changed for adjustable gain factor. The gain
ratio should be set according to the expected input size and required output size. In the
above example it is just set to 11.
3. An inverting active integrator : Integrates the exponential, which later decays (with a much
larger time constant) and slowly returns to zero.
It is well known that the time constants of the differentiator and integrator should be matched so
that the output is semi-gaussian.
(Image source: ORTEC® - Introduction to Amplifiers)
Additional components include decoupling capacitors for both the op-amps. There are multiple
test points marked TP for easily checking voltages at important points.
8
Implementation
The above PCB (and previous circuit schematic) was designed using CaDSoft Eagle software. The
PCB measures roughly 2.35 cm x 1.84 cm. By mistake, the SMA connectors have not been placed
on the periphery. 0603 resistors and capacitors have been used. Also there was a slight error in the
actual printing of the board; the pads have been printed on the back instead of the front, rendering
them nearly useless.
Since the board made by chemical etching, the vias are just holes with no copper walls so they
were connected with thin silver wires.
9
Results and Conclusions
The circuit was tested by giving 10kHz square pulses of varying amplitudes. At this frequency
the expected final output would be a sequence of semi-gaussians which decay well before the next
rising or falling edge is encountered. At the differentiator stage, an output of decaying exponential
curves is expected.
The current circuit I’ve made is not functioning as expected, even at somewhat low frequencies of
10kHz with low Pk-Pk input voltages (∼0.1V). Except for the output of the first stage (passive
differentiator), the rest of the circuit’s voltages are not correct; when tested with different supply
voltages (±8V,±5V) although the op-amp was not in saturation (output voltage < Vcc), the wave-
forms observed at the two op-amp terminals were significantly different.
I have double-checked the connections but I have not yet been able to debug the circuit completely
to pinpoint the exact source of the problem but I suspect that either (or both) of the op-amps have
stopped functioning correctly, perhaps due to improper handling on my part and/or poor soldering
technique.
A simpler and better approach would have been to first make the circuit on a breadboard as a proof-
of-concept using easily available components, which would function for relatively low frequencies.
This can also be done with more complicated shapers including two or more integration steps
(with appropriate time constants); a breadboard implementation would have enabled much faster
iteration and debugging. After demonstrating its functioning, making an actual circuit on a PCB
would have been more sensible.
10
