Transmission and Attenuation of Electromagnetic Pulses

The purpose of this experiment is to acquaint you with some of the principles
involved in the manipulation of electrical pulses. You will study reflections,
measure the effects of termination resistances, and measure pulse propagation
velocities for coaxial cables.

1. PRE-LAB

 Review Melissinos (2003) chapters 3.2 [Basic electronic equipment], 3.3 [Oscilloscopes
and digitizers], and 3.4 [Simple measurements]. Pulses are ubiquitous in timing
experiments, in all digital systems (computers, cell phones), and we must use
transmission lines (ribbon cable, coaxial cable) to move these pulses from one place to
the other.

 Pulses are also sometimes used to excite oscillators. Consider the tuning fork analogy,
where you strike a tuning fork. The oscillations decay over time. The longer that time
constant the higher the Q. The Q of an RLC circuit is 2π times the ratio between the
energy stored in the circuit (in C and in L) and the energy dissipated (by R), and equals
the resonant frequency divided by the bandwidth at half power. Extra credit: For a
damped harmonic oscillator derive a relation between the number of cycles to reach 1/e
of some initial amplitude and the Q.

 Read the files Coax impedance and Characteristic impedance on the 123 Electronics Lab
web page. Derive the 1-dimensional wave equation for a transmission line.

 Draw the predicted shape and amplitude of an ideal rectangular pulse of amplitude 1 volt
and duration 100 nsec after it has traversed a coaxial cable 100 m long and returned
following reflection from an open end.

2. INTRODUCTION

Many experiments involve the production and measurement of electrical pulses. Depending
on the timescale of the signals Δt, the approaches are very different:

Δt > 1ms do with your wires whatever you want. If the signals are small, you may want to use
shielded cables. (You did this in the analog LRC exercise)

1ms > Δt > 0.1μs typical for computers. Use ribbon cable, “twisted pair” or cables without
termination.

1
100ns > Δt > 0.1ns region of interest for “fast” signals to be studied here. Cables must be
terminated, or undesirable reflections will occur. In 122 Lab these types of signals are often
present with photomultiplier signals but the subsequent signal processing and logic slow the
signals into the middle category.

One may wish to know their rate, distribution of amplitudes, the relation of their occurrence
times relative to other pulses, etc. Such measurements are done with oscilloscopes, multichannel
analyzers, amplifiers, discriminators, coincidence circuits, etc.1, which may or may not be
working properly. It is essential, therefore, to gain facility in the use of test equipment such as
pulse generators and oscilloscopes so that the performance of a pulse measuring apparatus can be
checked, point by point. Electrical pulses are piped around a laboratory via transmission lines of
one sort or another, with consequent delays, attenuations, and reflections. It is important to
understand these effects and to know how to measure them. So this experiment is a study of
pulses in transmission lines. You learn that you must terminate all transmission lines properly so
as not to be deceived.

3. EXPERIMENTS

3.1. Reflection of electrical pulses from discontinuities in a transmission line

In this experiment you will be using an oscilloscope with better time resolution (250MHz
bandwidth). Connect the Tektronix PG502 pulse generator (be sure to terminate its output with a
50 ohm termination via a T connector) to the input of a Tektronix 2202B oscilloscope by means
of a BNC T connector. Before attaching your long cable, you need to set up the pulse generator
and scope. Put the scope on AC coupling, about 500mV per division, 25ns time base. On the
PG502 pulse generator, switch to a 5ns pulse with and a convenient 10 KHz repetition rate. Set
the variable pulse width control fully CCW to x1. There are grey knobs for output DC level in
the middle of the panel. Adjust the “Lower” level to zero volts, and then adjust the “Upper” level

1
Oscilloscopes display the signals V (t) vs. t. Multichannel Analyzers (MCA’s) sort signal events of
different heights into bins in histograms. Amplifiers enlarge the signal but may also alter it’s shape.
Charge sensitive preamplifiers produce an output voltage proportional the amount of charge generated in
a detector. Discriminators emit a logic (square V(t)) signal if the input exceeds a certain threshold.
Coincidence units produce a logic pulse if two (or more) inputs overlap in time.

2
for ~1 volt peak pulse height. Get your scope triggering on the leading edge of this pulse at
approx 500mV level. Now attach a long RG58 cable to the third end of the T, as shown in Figure
1. Observe the primary pulse from the pulser and describe the pulse reflected from the end of the
cable when the end is 1) open, 2) shorted, and 3) terminated with a resistor “termination” with
values of 50, 75, and 95 ohms (color coded), in order to determine the characteristic impedance
of RG58 cable. Remember to graph the observed waveforms in your notebook and label both
time and amplitude axes! Save the waveforms on a memory stick for your lab book if you like.
You will typically see one reflection. Why not n+1? Hint: You terminated the pulse generator in
50Ω at its output. Take this generator termination off and see what happens.

The dark green RG 58 cables in the lab are supposed to be about 19.7m long (but you should
check if the length is not marked on the cable!); one light green cable is labeled 50 feet. There
are enough of these long cables for 11 simultaneous experiments.

3.2. Speed and attenuation of pulses in transmission lines.

Determine the velocity of pulses in the cable by measuring the difference in the arrival time of
the direct and reflected pulses at the oscilloscope. Expand the scope sweep timescale so you just
see two pulses (you emitted pulse and the return reflected pulse having traveled twice the length
of the cable). Use the dual cursor on your scope to accurately measure the time delay of the
leading edge of your pulses at the same relative amplitude. Record sufficient data and other
information to permit an accurate assessment of the random and the systematic errors in your
determinations. Compare the velocity in the cable with the velocity of light in vacuum, and
explain the cause of the difference. Extra credit: Measure the attenuation of RG58 cable by
comparing the amplitude of the pulse reflected from the open end of the cable with and without
an additional length joined by a BNC connector. (Note that this strategy isolates for
measurement the effect of the delay in the cable from possible complicating effects of the
discontinuities in the circuit at the connections to the oscilloscope and pulse generator.)

3
4. ANALYSIS

1. Estimate the characteristic impedance Z for RG58 coaxial cable using only your
experimental data.

2. Determine the velocity of propagation Vprop for RG58 coaxial cable. Express this as a
fraction of the velocity of light in vacuum (this ratio is called the “velocity factor” of the cable)
and compare your value with the published values. Assess the random and systematic errors.

3. Determine the attenuation in dB/m for your RG58 coaxial cable. Compare your
coefficients versus published values. Assess the random and systematic errors. A dB (decibel) is
a measure of log power ratio. dB = 10 log(P1/P2), and since power is proportional to the square
of voltage dB = 20 log(V1/V2). 20 dB is a factor of 100. 3dB is a factor of 2.

References:
[1] Melissinos (2003) chapters 3.2, 3.3, 3.4.

[2] W. Leo, Techniques of Nuclear and Particle Experiments (Springer, 1992).

[3] G. Bekefi and A. Barrett, Electromagnetic Vibrations, Waves and Radiation (MIT Press,
1977).

[4] http://www.youtube.com/watch?v=LIvmfbMaTAU