All figures and tables have been removed from this text-only version
of this document. Scanned GIF89A images of all figures and tables are
provided separately. These are named according to the convention:
EQ_#.GIF for equations, FIG##.GIF for figures and TAB##.GIF for 
tables. Captions for figures and tables are included at the end of 
this document. In addition the following conventions have been used 
in this document to represent non-ASCII characters appearing in the 
original document:

1.  '~' before another symbol represents approximately (e.g. ~= is
    'approximately equal', etc.).
2.  >= and <= represent greater than or equal and less than or equal
    respectively.
3.  characters preceded by the carat character '^' are superscripted
4.  characters enclosed in square brackets '[]' are subscripted
5.  non-ASCII characters (Greek or mathematical symbols, etc.) are
    replaced by a description also contained in square brackets '[]'
    (e.g. [pi] = lower case Greek letter pi, [DELTA] = upper case
    Greek delta, [integral] = integral sign, etc.)
6.  the '*' character represents either multiplication or a dot product 
    depending upon whether the operands are scalar or vector quantities

Space Science Reviews 21 (1977) 289-308. All Rights Reserved.
Copyright 1977 Kluwer Academic Publishers, Dordrecht, Boston, London.
Reprinted with permission of Kluwer Academic Publishers.

This material is posted here with permission of Kluwer Academic 
Publishers (Kluwer). Such permission of Kluwer does not in any way 
imply Kluwer endorsement of any PDS product or service. Internal or 
personal use of this material is permitted. However, permission to 
reprint/republish this material for advertising or promotional 
purposes or for creating new collective works for resale or 
redistribution must be obtained from Kluwer.

By choosing to view this document, you agree to all provisions of the 
copyright laws protecting it.

A PLASMA WAVE INVESTIGATION FOR THE VOYAGER MISSION

FREDERICK L. SCARF
TRW Defense & Space Systems Group, One Space Park, Redondo Beach, 
Calif. 90278, U.S.A

and

DONALD A. GURNETT
Department of Physics & Astronomy, University of Iowa, Iowa City, Iowa 
52242, U.S.A.

(Received 24 May, 1977)

Abstract. The Voyager Plasma Wave System (PWS) will provide the first 
direct information on wave-particle interactions and their effects at 
the outer planets. The data will give answers to fundamental questions 
on the dynamics of the Jupiter and Saturn magnetospheres and the 
properties of the distant interplanetary medium. Basic planetary 
dynamical processes are known to be associated with wave-particle 
interactions (for instance, solar wind particle heating at the bow 
shock, diffusion effects that allow magnetosheath plasma to populate 
the magnetospheres, various energization phenomena that convert 
thermal plasma of solar wind origin into trapped radiation, and 
precipitation mechanisms that limit the trapped particle populations). 
At Jupiter, plasma wave measurements will also lead to understanding 
of the key processes known to be involved in the decameter bursts such 
as the cooperative mechanisms that yield the intense radiation, the 
observed millisecond fine-structure, and the Io modulation effect. 
Similar phenomena should be associated with other planetary satellites 
or with Saturn's rings. Local diagnostic information (such as plasma 
densities) will be obtained from wave observations, and the PWS may 
detect lightning whistler evidence of atmospheric electrical 
discharges. The Voyager Plasma Wave System shares the 10-meter PRA 
antenna elements, and the signals are processed with a 16-channel 
spectrum analyzer, covering the range 10 Hz to 56 kHz. At selected 
times during the planetary encounters, the PWS broadband channel will 
operate with the Voyager video telemetry link to give complete 
electric field waveforms over the frequency range 50 Hz to 10 kHz.

1. Introduction

The Voyager plasma wave system (PWS) will measure electric field 
components of local plasma waves over the frequency range extending 
from 10 Hz to 56 Hz. The PWS shares the two extendible 10-meter 
electric antenna elements provided by the planetary radio astronomy 
(PRA) investigators, but the two groups use these sensors in different 
ways. For the radio measurements the antennas are connected as a pair 
of orthogonal monopoles, while the PWS investigators use these same 
elements to form a balanced electric dipole. In the normal format, the 
plasma wave signals are processed with a simple 16-channel spectrum 
analyzer, and at the planetary encounters this system will provide a 
full spectral scan every four seconds. The PWS also has a broadband 
amplifier that will utilize the Voyager video telemetry link to give 
electric field waveforms (frequency range 50 Hz to 10 kHz) at selected 
times during planetary encounters.

The Voyager plasma wave investigation is designed to provide key 
information on the wave-particle interaction phenomena that control 
important aspects of the plasma dynamics in the magnetospheres of 
Jupiter and Saturn. It is well known that wave-particle interactions 
play extremely important roles at Earth, and it is also known that the 
dynamics of at least the inner magnetosphere of Jupiter is 
conceptually similar to that of Earth, despite the vast difference in 
size and in energy of the trapped particles. In addition, the 
satellites of Jupiter and Saturn provide important localized sources 
of plasma and field-aligned currents, and they must significantly 
affect the trapped particle populations. Pioneer 10, 11 measurements 
indicate that in the inner region of Jupiter the trapped electron 
fluxes are near the stable limit set by whistler mode and half- 
gyrofrequency harmonic mode wave-particle interactions; however, the 
ion fluxes are probably controlled by other instabilities (ion 
cyclotron, drift wave, loss cone, etc.). The plasma waves cause pitch- 
angle diffusion, and the precipitating electrons and ions must 
significantly affect the ionospheric properties and influence 
ionosphere-magnetosphere coupling. It is also expected that some kinds 
of enhanced precipitation and wave-induced anomalous conductivity 
effects develop along field-lines threading the inner satellites, and 
strong plasma wave radiation field coupling must account for the 
intense levels of decametric emissions.

Studies of wave-particle interaction phenomena in the outer 
magnetospheres of Jupiter and Saturn are also of great importance. 
These outer regions should be dominated by the high-[beta] 
[=4*[pi]*N*[kappa]*T/B^2] plasma spun out by centrifugal forces, and 
here wave-particle interactions can provide local acceleration; they 
can affect particle diffusion; and they can lead to field-line 
merging. If the magnetospheric plasma at Jupiter or Saturn flows 
radially outward to form a planetary wind, wave-particle interactions 
should lead to a second collisionless shock within the magnetopause. 
Other relevant wave-particle interactions involve whistlers generated 
by atmospheric lightning, and electron plasma oscillations associated 
with suprathermal particles in the upstream solar wind.

The fundamental roles of wave-particle interactions and plasma 
instabilities at Jupiter have been discussed in recent reviews by 
Coroniti (1975), Goertz (1976), Scarf (1976), and Kennel and Coroniti 
(1977). Scarf (1975) recently speculated on the structure of Saturn's 
magnetosphere, and in the context of a Uranus magnetosphere 
configuration of the novel type discussed by Kennel (1973) and by 
Siscoe (1975), it is anticipated that the Voyager plasma wave 
investigation can also provide unique and valuable information on 
wave-particle interaction phenomena near Uranus.

In this report we first present some background information on plasma 
waves and magnetospheric dynamics, with examples drawn from near-Earth 
observations. This section is followed by a discussion of specific 
outer planet science objectives, and a summary of the science 
rationale for the Voyager PWS design. The instrument details are 
summarized in the following section.

2. Background

Extensive laboratory studies and many theoretical analyses show that 
waves play dominant roles in all non-equilibrium plasma systems (see 
for instance, Stix, 1962; Montgomery and Tidman, 1964; Sagdeev and 
Galeev, 1969). In particular, when the plasma is dilute and cool, 
ordinary coulomb collisions are unimportant and wave-particle 
interactions provide the scattering and acceleration mechanisms that 
govern the dynamics. As noted above, turbulent wave-particle 
scattering leads to an effective conductivity and resistive-type 
heating; resonant wave-particle interactions give rise to acceleration 
similar to that in a cyclotron; and waves also allow diffusion that 
can be directly associated with particle energization (i.e., cross L 
diffusion with conservation of magnetic moment, [mu] = m*v[perp]^2/B, 
gives v[perp] ~ B^(1/2)). Waves also cause diffusion in pitch angle 
and particle precipitation (see Fredricks, 1975, for a recent 
summary). In all cases, the most significant interactions involve 
waves that are strongly emitted and absorbed by the plasma ions and 
electrons. The local interactions of importance are associated with 
waves having phase speeds comparable to plasma particle speeds, the 
'radiation' mechanism is essentially a generalized Cerenkov process, 
and the spectral region where the strong wave absorptions and 
emissions occur is called the plasma wave region.

Plasma wave modes of importance in magnetospheric physics are related 
to the electron and ion gyrofrequencies, f[c]^(+/-) = 
e*B/(2*[pi]*m[+/-]*c), and to the electron and ion plasma frequencies, 
f[p]^(+/-) = (4*[pi]*N*e^2/m[+/-])^(1/2)/2*[pi], where N is the plasma 
density and B is the local magnetic field strength. As an example of 
the spectral range for the characteristic resonant frequencies, when N 
= 5 cm^-3 and B = 5 gamma (typical upstream solar wind conditions near 
1 AU), these expressions give f[p]^- = 20 kHz, f[p]^+ = 470 Hz, f[c]^- 
= 140 Hz, and f[c]^+ = 0.076 Hz. These plasma wave modes can also be 
categorized as generalized electromagnetic waves (having both E and B 
wave components), or as electrostatic oscillations (similar to 
compressional sound waves, with space charge variations that produce 
only electric field wave components). For instance, the whistler mode 
(f <= f[c]^-) is electromagnetic, but ion sound waves (f ~<= f[p]^+) 
and electron plasma oscillations (f ~= f[p]^-) are electrostatic.

Figure 1 shows some examples of natural plasma waves observed near 
Earth using the broadband or waveform channels of the plasma wave 
instruments on the OGO-5 and IMP-6 spacecraft. As shown in the top 
panels of the figure, the electron plasma frequency is significant in 
two distinct ways. Suprathermal electrons in the region upstream from 
the bow shock lose energy by radiating narrowband electrostatic waves 
(upper left) at f=f[p]^- (Scarf et al., 1971). In addition, f[p]^- is 
a well-known critical frequency for electromagnetic waves in the sense 
that the usual free space electromagnetic modes can be transmitted 
through a plasma if f>f[p]^-, whereas for f<f[p]^- they cannot 
propagate. The cutoff of outer magnetosphere electromagnetic noise at 
the plasma frequency is shown in the upper right panel of Figure 1. 
Measurement of this cutoff frequency provides an excellent technique 
for determination of the plasma density; in fact, since this technique 
is based on analysis of electromagnetic wave modes having wavelengths 
in the kilometer range, the presence of the spacecraft and its plasma 
sheath is unimportant, and observation of the f[p]^--cutoff provides 
an absolute and sheath-independent evaluation of the total plasma 
density (Gurnett and Frank, 1974).

Wave investigations in Earth orbit have also demonstrated that two 
general classes of wave-particle interactions are of great importance 
for magnetospheric dynamics, and these measurements are illustrated in 
the other panels of Figure 1. Electromagnetic and electrostatic plasma 
instabilities give rise to relatively narrow-banded spontaneous 
emissions (e.g., ELF hiss, chorus, three-halves noise, ion cyclotron 
and ion-plasma-frequency turbulence) that can scatter trapped 
particles into the loss cone, leading to modified pitch-angle 
distributions, stable trapping limits, diffuse aurora, proton 
precipitation events, etc. The current-driven plasma instabilities 
generate intense impulsive ion acoustic or Buneman mode turbulence 
that provides very effective energy transfer (via the anomalous 
conductivity mechanism) at the bow shock and in regions where strong 
field-aligned currents are observed.

In all cases, the plasma wave observations, in conjunction with 
measurements from the energetic particle detector, the plasma probe 
and the magnetometer, provide quantitative information needed to 
evaluate effective transport coefficients associated with the wave- 
particle interactions, as indicated by the following examples:

Anomalous Conductivity. At the collisionless bow shock, the radial 
outflow shock, at plasma boundaries, and on field lines threading the 
satellites, strong currents flow and the two-stream instability should 
produce intense electrostatic turbulence in the ion acoustic mode, 
with f ~<= f[p]^+, and V(phase)= [omega]/k ~= [sqrt]([kappa]*T[-
]/m[+]). For this type of interaction, the wave-particle scattering 
can be described in terms of an anomalous electrical conductivity 
[sigma] = j/E(DC) = N*e^2/m*[nu](eff), (Papadopoulos, 1977) where

     [nu](eff) ~= (2*[pi]^3)^(1/2)*
                  f[p]^-[epsilon][0]*E^2(wave)/2*N*[kappa]*T].     (1)

The resistive heating associated with this interaction is very 
effective in producing the shock dissipation and particle 
acceleration, and we plan to use the observations to evaluate 
[nu](eff) and to study the effect of this important interaction.

Whistler Mode Scattering. The simple Kennel-Petschek theory (1966) 
predicts that electrons with resonant energy

     E[R] = (B^2/(8*[pi]*N))*(f[c]^-/f)*[1-(f/f[c]^-)]^3     (2)

amplify waves with frequency f<f[c]^-, and the effective pitch-angle 
diffusion coefficient, D[alpha], is then roughly given by

     D[alpha] = (2*[pi]*f[c]^-)*[n*E(wave)/c*B]^2,     (3)

where n is the index of refraction, and E is the amplitude of the 
electric component of the wave. We intend to use the observations to 
derive energy-dependent and L-dependent pitch- angle distribution 
lifetimes and equilibrium distribution functions, and similar 
calculations will be carried out for other wave-particle interactions 
(e.g., ion cyclotron waves and protons; 3*f[c]^-/2 waves and 
electrons, etc.).

3. Specific Scientific Results Anticipated at Jupiter, Saturn and 
Uranus

For the Voyager mission, the most important outstanding questions 
involve (a) the dynamics of the outer planet magnetospheres; (b) the 
satellite-magnetosphere interactions; (c) the mechanisms affecting the 
radio emissions from the plasma environment of Jupiter, Saturn and 
Uranus; and (d) the ways in which dynamical magnetospheric processes 
affect the ionospheres and atmospheres of these outer planets. The 
remote sensing planetary science investigations on Pioneer 10, 11 and 
on Voyager utilize ultraviolet, radio, infrared or visible wave 
emissions, and these investigations do provide non-local information 
of great scientific value; however, corresponding information on 
magnetospheric dynamics and plasma processes can be obtained only by 
flying the relevant diagnostics on spacecraft which move through the 
interaction regions. Electrostatic plasma waves do not propagate any 
significant distance, and VLF electromagnetic waves, with f<
f[c]^(+/-), have very restricted propagation characteristics. 
Important corresponding perturbations in the magnetic field(e.g., 
field-aligned currents) and in the plasma distribution functions must 
also be studied with local diagnostics. Since Pioneer 10, 11 did not 
carry any wave instruments, the Voyager mission will provide the first 
measurements of wave-particle interaction phenomena at the outer 
planets.

The results anticipated in the primary scientific areas to be 
addressed by the Voyager plasma wave investigation can best be 
presented with reference to a drawing of the planned flyby 
trajectories (Figure 2) as follows:

(1) From present knowledge of Earth's magnetosphere (see Figures 1, 2) 
it is known that wave-particle interactions will play important roles 
in accelerating electrons and protons at the bow shock, in neutral 
sheet merging regions ('fireballs'), in the turbulent equatorial 
plasma sheet region of the outer magnetosphere, in the region of the 
internal (radial outflow) shock at the co-rotation boundary or Alfven 
point, wherever beam-plasma interactions provide field- aligned 
streaming, and on the field lines threading the satellites, where 
strong plasma currents must flow. The relevant measurements concerning 
acceleration will involve detection of intense and varying wave levels 
in association with significant local changes in the particle 
distribution functions.

(2) It is anticipated that the Voyager plasma wave experiment will 
provide extremely important information about the mechanisms that 
control the interactions of the satellites with the rapidly rotating 
magnetospheres. Some of the ways in which wave-particle interactions 
can dominate the interactions are indicated in Figure 3, which is 
presented in the context of the family of Io sheath-field line 
coupling models derived from the works of Goldreich and Lynden-Bell 
(1969), Gurnett (1972), and Shawhan et al.(1975)(1976) and co- 
workers. It is anticipated that similar classes of wave-particle 
interaction phenomena will prove to be significant at all of the inner 
satellites of Jupiter and Saturn, and the Voyager mission will provide 
unique opportunities to collect the vital experimental data on plasma 
instabilities associated with satellite-magnetosphere interactions.

(3) Measurements of plasma wave E component amplitudes and overall 
frequency characteristics will provide quantitative information on 
pitch-angle diffusion phenomena that lead to particle precipitation 
and stable trapping limits (see Equation 3). Detailed wave 
measurements are needed to sort out the mechanisms because 
electromagnetic whistlers and ion cyclotron waves, electrostatic 
gyrofrequency harmonics and ion acoustic modes, and drift-type waves 
can all scatter trapped particles into the loss cones. All of these 
wave-particle interactions are expected to operate on subgroups of 
particles having different energies, streaming characteristics, pitch-
angles and charge-to-mass ratios. The effects of the waves on the 
population of particles will vary with plasma density and, [beta]-
value, corotation speed, etc., and in this context it is clear that 
detailed measurements throughout the magnetospheres are needed to 
achieve the necessary understanding of the interaction mechanisms so 
that the associated pitch angle diffusion coefficients can be 
evaluated numerically. When this study is completed, the Voyager 
investigators will be able to understand the balance between particle 
sources and sinks in the Jovian magnetosphere.

(4) Sheath independent evaluations of the plasma density profiles will 
be available from analysis of the local plasma wave frequency 
spectrum. Within the Jupiter and Saturn magnetospheres, the value of 
the electron plasma frequency (f[p]^-) can be obtained by examining 
the lower cutoff of the synchrotron emission or the local lower hybrid 
resonance emission. In the upstream region, suprathermal electrons 
generate electrostatic emissions with f~f[p]^-, and identification of 
this mode leads to a determination of the plasma density. In both 
cases, the signals have wavelengths very large in comparison with the 
Voyager spacecraft, and the spacecraft and its sheath produce 
negligible perturbations, so that the density-value obtained in this 
way is sheath-independent.

(5) Earth-orbiting spacecraft carrying plasma wave instruments also 
commonly observe atmospheric discharges by detecting lightning 
whistler signals that escape into the magnetosphere. We intend to 
search for lightning whistlers at Jupiter and Saturn, and if we are 
successful this result would have great significance in terms of the 
far-ranging atmospheric implications. Detection of lightning whistlers 
from the Voyager is potentially of great value in other respects; the 
individual frequency components travel at different speeds in a 
magnetized plasma, and by analyzing the arrival times of different 
components, it will be possible to obtain non-local information on the 
density and magnetic field distributions along the field line from the 
ionosphere to the spacecraft. In addition, lightning whistlers 
commonly trigger local plasma emissions, and it can be anticipated 
that study of such emissions will give unique information on the 
stability of the Jupiter and Saturn magnetospheres.

(6) The Voyager plasma wave investigation can also provide basic 
information on wave-particle interactions at Uranus. The results 
anticipated vary considerably, depending on the large scale 
classification of the actual interaction, which may be dominated by 
the planetary magnetic field (as at Earth, Jupiter, Mercury, and 
perhaps Mars and Saturn), or by the planetary ionosphere (as at Venus, 
and perhaps Mars). There are also very significant distinctions and 
uncertainties because the overall dynamical interaction may involve 
the conventional supersonic solar wind, a solar wind modified by 
interstellar medium effects, or the interstellar medium itself. Even 
if the interaction is magnetic, the results will depend critically on 
the inclination of the magnetic dipole axis to the Uranus spin axis.

In the context of an interaction of the type discussed by Kennel 
(1973) and by Siscoe (1975), it is most interesting to consider the 
case where Uranus has a magnetic moment essentially parallel to its 
spin axis, with a field magnitude sufficiently large to stand off a 
conventional streaming solar wind. If the physics of the Uranus plasma 
environment is even roughly given by these models, then the 
magnetosphere of Uranus will be as shown in Figure 4 (adapted from 
Siscoe, 1975), and the anticipated results of the plasma wave 
investigation are summarized on the figure.

4. The PWS Design Approach

The plasma wave investigation was added to the Voyager mission in July 
1974, when the mission plans and spacecraft design were already fairly 
well defined. This late change in the payload was made with the 
understanding that the PWS-investigators would share the two 
orthogonal 10-meter Planetary Radio Astronomy electric antennas, so no 
measurements of magnetic wave components were contemplated. In 
addition, at this point in the mission planning the available mass, 
power and telemetry allocations were limited. However, in July of 1974 
we did have some very important advantages not previously available. 
For Jupiter, the December 1973 encounter of Pioneer 10 provided direct 
information on the magnetic field profile throughout the magnetosphere 
(Smith et al., 1974a, b), and this allowed us to determine the range 
of electron and ion gyrofrequencies and associated frequency ranges 
for whistler mode turbulence and 3*f[c]^-/2 signals (see Figure 1). 
The Pioneer 10 measurements of the extended magnetodisc (Smith et al., 
1974a, 1974b) and the direct observations of low energy plasma in the 
inner magnetosphere (Frank et al., 1976) also gave information on the 
plasma density distributions, and this could be used to determine the 
range of electron and ion plasma frequencies. This information was 
also used to construct models of the Saturn magnetosphere (Scarf, 
1973, 1975), and subsequent measurements of Saturnian radio emissions 
(Brown, 1975) show that these models have some validity.

In terms of characteristic frequencies, the spectral range of interest 
for local plasma wave phenomena extends from slightly below the proton 
gyrofrequency, f[c]^+, to slightly above either the electron 
gyrofrequency, f[c]^-, or the electron plasma frequency, f[p]^-, 
whichever is larger. Considering the radial variations involved and 
the relatively unknown nature of the Saturn and Uranus magnetospheres, 
the range of frequencies which must be covered is enormous. The 
central panel of Figure 5 shows the expected variation of the electron 
and proton gyrofrequencies and plasma frequencies in the Jovian 
magnetosphere, based on the Pioneer measurements and the plasma 
density model of Axford and Mendis (1974). The earlier models of Brice 
and co-workers give similar density profiles, and the measurements of 
Frank et al. (1976) give somewhat higher densities only within about 
L=10. At the closest approach of the Voyager to Jupiter (~5 R[J]) the 
electron gyrofrequency is approximately 100 kHz. Since PRA will 
provide its best coverage above 100 kHz, we have set the highest PWS 
spectrum analyzer channel at 56 kHz, as shown in the left-side of 
Figure 5.

In the outer regions of the magnetosphere the lowest frequency of 
interest, f[c]^+, is extremely low, ~10^-2 Hz. Practically it is not 
possible, or even desirable, to extend the range of this experiment to 
such low frequencies. At low frequencies, <10 Hz, conventional (dc) 
magnetometers have good sensitivity for electromagnetic waves, and 
since the magnetometer on Voyager will generally have an upper cutoff 
frequency of a few Hz (the basic sampling rate gives 16 2/3 vectors 
per second) , we have chosen 10 Hz for the lowest PWS spectrum 
analyzer channel. The filter response curves for the sixteen-channel 
spectrum analyzer are drawn on the extreme left-hand side of Figure 5.

In addition to the spectrum analyzer, the PWS has a broadband channel 
designed to utilize the 128 kb/sec video telemetry and data storage 
link for selected periods during the magnetosphere encounters. The 
lower frequency limit has been set near 50 Hz, in order to provide an 
opportunity to search for proton cyclotron waves during the JSI 
closest approach (r ~= 5 R[J]). At the upper end, the filter has been 
designed to roll off strongly between 10 and 12 kHz in order to avoid 
signal aliasing (the Nyquist frequency is 14.4 kHz) and the 10 kHz 
bound is indicated on the figure.

The right panel of Figure 5 contains a corresponding display of the 
characteristic wave frequency variations anticipated for the Saturn 
encounters. The Saturn magnetosphere measurements are, of course, 
exploratory, and this illustration must not be taken to represent more 
than a reasonable expectation, based on some general models of outer 
planet magnetospheres (Kennel, 1973; Scarf, 1973, 1975) and on the 
interpretations of low level radio signals from Saturn (Brown, 1975). 
To the extent that these models have validity, the PWS spectrum 
analyzer measurements at Saturn should be comparable with those made 
at Jupiter, in terms of characteristic frequency coverage. In 
particular, for the planned close flyby of Titan, the plasma wave 
measurements, in conjunction with observations from the plasma probe, 
the magnetometer, and the energetic particle detector, should give 
full information on the interaction between the Titan atmosphere-
ionosphere system and the rapidly rotating magnetosphere of Saturn.

The outer planet scaling concepts discussed by Kennel (1973) and by 
Scarf (1973, 1975) can also be used to estimate plausible ranges of 
characteristic plasma wave frequencies for Uranus, and with somewhat 
more certainty it is possible to extrapolate the range of 
interplanetary wave modes out past the orbit of Neptune. For the 
distant solar wind, out to the heliosphere boundary or the limit of 
tracking, the range of the PWS should include the local ion and 
electron plasma frequencies, but beyond about 15 AU the electron 
cyclotron frequency should fall below the 10 Hz lower limit of the 
spectrum analyzer, so that this characteristic electromagnetic mode 
will then be detectable only in the magnetometer data. This ability to 
measure wave phenomena in the distant solar wind means that the PWS 
should be able to yield necessary information on the dynamical 
phenomena that develop in the region of the Uranus (or Neptune) bow 
shock, and at a possible heliosphere boundary shock. In fact, for the 
Siscoe (1975) model of the Uranus magnetosphere, the anticipated 
ranges of f[p]^(+/-) and f[c]^(+/-) during encounter are similar to 
the Saturnian frequency profiles of Figure 5 (to within a factor of 
two) and the Voyager PWS, together with the other Voyager instruments, 
may provide essentially all of the information indicated on Figure 4.

5. Instrumentation

The Voyager plasma wave instrumentation has two basic elements: the 
electric antenna system to be shared with the Planetary Radio 
Astronomy Group and the main electronics amplifier and all associated 
electronics.

A. SENSORS

The basic antenna sensors for the plasma wave experiment consist of 
two extendible elements each with a nominal length of 10 m, mounted as 
shown in Figure 6. These antennas are beryllium-copper overlapped 
tubes, approximately 0.5 inches in diameter, which are rolled flat 
onto a spool prior to extension. The elements are extended by a motor-
driven mechanism which is turned on by command; this stops 
automatically when the antennas are fully extended. In the deployed 
configuration, the balanced Vee has an effective length of about 7 m.

B. ELECTRONICS

All of the signal processing for the plasma wave experiment is 
performed in a single main electronics box which is mounted near the 
electric antennas, above the PRA, as shown in Figure 6. The block 
diagram of the instrumentation in the main electronics box is shown in 
Figure 7. Signals from the electric antennas go to two pairs of 
preamplifiers; one pair is connected directly to the antennas (x1) and 
the second pair has a 40 db attenuator between the antennas and the 
preamplifiers (x0.01). The outputs of either pair of preamplifiers can 
be selected by command. This command permits us to shift the dynamic 
range of the experiment by 40 db if unusually large signals are 
detected during a planetary encounter. The selected preamplifier 
outputs then go to a differential amplifier which provides an output 
signal proportional to the voltage difference, hence electric field, 
between the antenna elements. A notch filter is used after the 
differential amplifier to attenuate interference signals at the 
spacecraft power supply frequency 2.4 kHz and its 3rd harmonic, 7.2 
kHz.

Signals from the differential amplifier are processed in two ways: (1) 
by a 16-channel spectrum analyzer and (2) by a waveform amplifier. The 
16-channel spectrum analyzer consists of 16 discrete filters with four 
frequencies per decade from 10 Hz to 56.2 kHz (i.e., 10 Hz, 17.8 Hz, 
31.1 Hz, etc.). These filters have bandwidths of +/- 15.0% for the 
lowest 8 frequencies (active filters) and +/- 7.5% for the highest 8 
frequencies (passive filters). The outputs from these filters are 
switched into two logarithmic detectors which provide output voltages 
(0 to 3 V) to the spacecraft data system. A reset pulse from the data 
system is used to synchronize the counter which controls the channel 
selection. The frequency selection and sampling rate of the spectrum 
analyzer outputs are completely controlled by programs in the flight 
data system. The normal operating mode assigned for planetary 
encounters involves one scan through all 16 frequencies in 4 sec 
(i.e., 32 bps).

The spectrum analyzer system gives high sensitivity but the threshold 
does vary somewhat with frequency. In terms of sine wave signals at 
the center frequencies of the bandpass channels, channel 1 (10 Hz) has 
a threshold of about 1.7 [micro]V m^-1, while channel 16 (56 kHz) has 
a threshold of 0.3 [micro]V m^-1.

The waveform amplifier and associated A/D converter shown in Figure 7 
are intended to provide very high time resolution measurements of 
electric field waveforms at selected times during planetary 
encounters. A bandpass filter ahead of the waveform amplifier limits 
the bandwidth of these measurements to the range from 50 Hz to 10 kHz. 
The waveform amplifier consists of an automatic gain control circuit 
which maintains an essentially constant output amplitude independent 
of the amplitude of the input signal, with a time constant of 0.5 sec. 
The output of the waveform amplifier is sampled at a rate of 28 800 
samples/second by a 4-bit A/D converter. The 115.2 kbps output from 
the A/D converter is recorded by the spacecraft data storage system at 
selected periods during planetary encounters.

The waveform channel provides an enormous increase in our ability to 
identify and analyze wave phenomena because the broadband measurement 
gives virtually continuous coverage in frequency and time, subject 
only to the constraints of the [DELTA]f*[DELTA]t ~= 1 limitation. The 
data in the broadband panels of Figure 1 were all acquired using the 
waveform capabilities of the IMP-6 and OGO-5 plasma wave instruments, 
together with the high rate analog spacecraft telemetry links. For 
Voyager, the 128 kb/sec transmissions are digital, but the JPL Imaging 
Processing Laboratory has existing data processing routines that can 
readily be used to make up broadband frequency-time spectrograms 
similar to those shown in Figure 1. For instance, we show in Figure 8 
how an S^3-A recording of a lightning whistler would appear when the 
signal is played through the Voyager PWS waveform link, and processed 
through the Imaging Processing Laboratory facilities. Each flyby of 
Jupiter should provide a total of more than 10 hours of such waveform 
data, and for JST we anticipate that frequency-time spectrograms 
similar to Figure 8 will be available from the region of the Io flux 
tube.

C. PHYSICAL CHARACTERISTICS

The packing of electronics in the main electronics housing consists of 
a combination of cordwood modules and custom hybrid circuits attached 
to a 0.062-inch printed circuit motherboard. The mechanical housing is 
machined from magnesium and is gold plated to provide a protective 
finish. Feed through filters on bulkheads and other filtering 
techniques are used to eliminate conducted RFI from the spacecraft. 
Physical and electrical characteristics of the instrument are 
summarized in Table 1.

D. PRIOR PERFORMANCE AND DESIGN HISTORY

The basic University of Iowa hybrid circuits and electronics designs 
used in this investigation have had extensive use in spacecraft plasma 
wave instrumentation over the past seven years, including flights on 
the IMP-6, S^3-A, IMP-8, Hawkeye-1 and Helios-1,2 spacecrafts. Related 
plasma wave measurement techniques were used to fabricate instruments 
for Pioneer 8, 9, OGO-5, IMP-7, and similar hardware is now being 
integrated on the ISEE A, B, C and Pioneer-Venus-Orbiter spacecraft. 
The specific design of the Voyager Plasma Wave System bears a very 
close similarity to that of the S^3-A plasma wave investigation, and 
the relatively minor changes that were imposed were primarily needed 
to: (a) accommodate the Voyager power supply and interface 
requirements, and (b) reduce susceptibility to radiation interference 
effects. The design, fabrication and testing of the Voyager PWS was 
performed by the staff of the University of Iowa.

E. RADIATION ENVIRONMENT EFFECTS

As part of the Voyager experiment development, extensive analysis and 
testing have been performed to assure that this subsystem will operate 
at Jupiter and survive the passage through the Jovian radiation belts. 
The proper operation and survival of the flight equipment in the 
Jovian radiation environment has been full confirmed by these analyses 
and tests. Specific steps that were taken to assure satisfactory 
operation include: (a) the exposure of samples of transistors from 
each wafer used to construct the flight unit hybrid circuits to 
suitable radiation levels, to verify that these specific transistor 
wafers are not defective or unusually sensitive to radiation; (b) the 
use of devices screened and tested by JPL for radiation resistance; 
(c) redesign of some circuits to make the operation more immune to 
radiation-induced parameter shifts.

G. SPACECRAFT CHARGING AND INPUT CIRCUIT PROTECTION

In a space plasma, photoelectron emission and secondary electron 
emission from the spacecraft surfaces are both generally significant, 
and plasma probe measurements always involve ambient distributions 
plus populations of charged particles that are emitted from or trapped 
near the spacecraft. At Jupiter, these complexities may also involve 
sputtering from spacecraft surfaces. Moreover, a plasma probe on a 
spacecraft always measures the spectrum of ions or electrons arriving 
at the detector, but these charged particles are locally accelerated 
or retarded by sheath fields. The energy shifts associated with the 
plasma sheath can be extremely important; DeForest (1972) showed that 
ATS-5 charged up to 10 kV (negative) when the electron temperature at 
synchronous Earth orbit (6.6 R[E]) reached 10-20 kV during substorms, 
and the possible scientific relevance of this charging problem at 
Jupiter has been discussed by Scarf (1973, 1975) and by Mendis and 
Axford (1974). Higher spacecraft charging levels can generally be 
expected at Jupiter, and Pioneer 10, 11 measurements do indicate that 
large sheath field fluctuations did develop within about 13 R[J]. For 
completely unconnected metal surfaces, potential differences of many 
kilovolts are likely at Jupiter, although the associated problems 
depend on the details of the spacecraft electrical and mechanical 
configuration. For Voyager wave experimenters, these considerations 
lead to special concerns about arcs and input circuit protection since 
the 10-m antennas must be floating with respect to spacecraft ground 
at all frequencies above 10 Hz.

Fortunately, Voyager is implementing an equipotential spacecraft 
design requirement so that there should be no possibility of arcs from 
thermal blankets, for instance, to the antennas. In this case, any 
change in antenna potential with respect to ground will be current 
limited to the maximum collection capability of the antenna, and we 
simply have to evaluate the maximum current that can be carried.

For E>10 keV, Pioneer 11 found J(max) ~= 10^9 cm^-2 sec^-1 or i(max) = 
1.6x10^-10 A cm^-2 (Fillius et al., 1975). We increase this maximum by 
a factor of five to account for space, time variations, E<10 keV 
electrons, Io emissions, etc. In this case, the maximum charging 
current to the antenna (assuming no arcing) is 4 [micro]A, and a 20 
M[Ohms] input resistor to ground will keep the dc antenna-to-ground 
potential difference at or below 80 V, which is well below the 500-V 
rating of this resistor and the 500-V to 1 kV ratings of the PWS-PRA 
coupling capacitors. This input resistor has been inserted to provide 
protection against the development of extreme sheath field transients.

6. Concluding Remarks

The inherent noise detected in the plasma wave subsystem when mounted 
on the Voyager spacecraft and interconnected to the PRA front end and 
the antenna mechanisms appears to be far below the levels anticipated 
for ambient plasma waves at Jupiter or Saturn. In fact, the equivalent 
sine wave sensitivities of 1.7 [micro]V/M (10 Hz) to 0.3 [micro]V/M 
(56 kHz) are far below the ranges for signal levels customarily 
encountered in Earth orbit when discrete emissions or current-driven 
plasma waves are being studied. With reference to Figure 1, typical 
sine wave levels for 3*f[c]^-/2 emissions or ion acoustic waves at the 
bow shock range between 100 [micro]V/M and 10 mV/M [see, for instance, 
Kennel et al. (1970); Fredericks et al. (1970); Gurnett and Shaw 
(1973); Rodriguez and Gurnett (1975)], and even the high frequency 
f[p]^- emissions (the weakest signals studied in Earth orbit) 
customarily have amplitude levels that are an order of magnitude above 
the PWS high frequency threshold.

Of course, these PWS sensitivities have not yet been measured in a 
plasma, and spacecraft noise coupling to the electric antenna through 
the plasma sheath can frequently produce serious problems that are 
only evident after launch. For Voyager, with its 2.4 kHz square wave 
power distribution, this kind of prospective spacecraft interference 
concerned us greatly during the design phase and led to the inclusion 
of the 2.4 kHz and 7.2 kHz notch filters. The use of a balanced dipole 
should also strongly reduce any in-orbit interference effects of this 
type since the balanced dipole gives a 40 db common-mode rejection 
that is very effective against spacecraft noises. Finally, the 
imposition of the Voyager electrostatic cleanliness specification 
should provide considerable additional protection since the thermal 
blankets and the mission Module structure form a relatively continuous 
Faraday shield that will keep much of the spacecraft noise away from 
the plasma. Since Voyager does not have solar arrays, which frequently 
provide strong coupling between spacecraft noise signals and the 
plasma, the prospects for a quiet environment around the Voyager 
spacecraft appear to be very good.

Acknowledgments

The plasma wave investigation was added to the Voyager payload at a 
late date, and the success of the PWS project is directly attributable 
to the exceptional support that has been provided by our colleagues at 
the University of Iowa, by so many members of the Voyager Project 
Staff at JPL and at NASA Headquarters, and by the Planetary Radio 
Astronomy team, led by James Warwick.

We are especially greatful to M. Agabra, the PWS Cognizant Engineer at 
JPL, and to R. Shaw, W. Kurth, S. Remington and R. Randall of the 
University of Iowa for their excellent and dedicated support and their 
many invaluable contributions throughout the design, fabrication, 
integration and launch preparation phases of this project.

References

Axford, W. I. and Mendis, D. A.: 1974, Ann. Rev. Earth Planet. Sci. 2, 
419.
Brown, L. W.: 1975, Astrophys. J. 198, L89.
Coroniti, F. V.: 1975, in V. Formisano (ed.), The Magnetospheres of 
the
Earth and Jupiter, D. Reidel Publishing
     Company, Dordrecht, Holland, p. 391
De Forest, S. E.: 1972, J. Geophys. Res. 77, 651.
Fillius, R. W., McIlwain, C. E., and Mogro-Comparo, A.: 1975, Science 
188,
465.
Frank, L. A., Ackerson, K., Wolfe, J. H., and Mihalov, J. D.: 1976, J.
Geophys. Res. 81, 457.
Fredricks, R. W., Crook, G. M., Kennel, C. F., Green, I. M., Scarf, F. 
L.,
Coleman, P. J., Jr., and Russell, C. T.: 1970,
     J. Geophys. Res. 75, 3751.
Fredricks, R. W.: 1975, Space Sci. Rev. 17, 449.
Goertz, C. K.: 1976, in T. Gehrels (ed.), Jupiter, University of 
Arizona
Press, p. 32.
Goldreich, P. and Lynden-Bell, D.: 1969, Astrophys. J. 156, 59.
Gurnett, D. A.: 1972, Astrophys. J. 175, 525.
Gurnett, D. A. and Shaw, R. R.: 1973, J. Geophys. Res. 78, 8136.
Gurnett, D. A. and Frank, L. A.: 1974, J. Geophys. Res. 79, 2355.
Kennel, C. F. and Petschek, H. E.: 1966, J. Geophys. Res. 71, 1.
Kennel, C. F., Scarf, F. L., Fredricks, R. W., McGehee, J. H., and
Coroniti, F. V.: 1970, J. Geophys. Res. 75, 6136.
Kennel, C. F.: 1973, Space Sci. Rev. 14, 511.
Kennel, C. F. and Coroniti, F. V.: 1977, Ann. Rev. Astron. Astrophys. 
(in
press).
Montgomery, D. C. and Tidman, D. A.: 1964, Plasma Kinetic Theory,
McGraw-Hill Book Company, New York.
Papadopoulos, K.: 1977. Rev. Geophys. Space Phys. (in press).
Rodriguez, P. and Gurnett, D. A.: 1975, J. Geophys. Res. 80, 19.
Sagdeev, R. Z. and Galeev, A. A.: 1969, Non-Linear Plasma Theory, W. 
A.
Benjamin, Inc. Book Company, New
     York City, N. Y.
Scarf, F. L., Fredricks, R. W., Frank, L. A., and Neugebauer, M.: 
1971, J.
Geophys. Res. 76, 5162.
Scarf, F. L.: 1973, Cosmic Electrodyn. 3, 437.
Scarf, F. L.: 1975, in F. Formisano (ed.), The Magnetospheres of the 
Earth
and Jupiter, D. Reidel Publishing
     Company, Dordrecht, Holland, p. 433.
Scarf, F. L.: 1976, in T. Gehrels (ed.), Jupiter, University of 
Arizona
Press, p. 870.
Shawhan, S. D., Goertz, C. K., Hubbard, R. F., Gurnett, D. A., and 
Joyce,
G.: 1975, in V. Formisano (ed.), The
     Magnetospheres of Earth and Jupiter, D. Reidel Publishing 
Company,
Dordrecht, Holland, p. 375. Shawhan,
     S. D.: 1976, J. Geophys. Res. 81, 3373.
Siscoe, G. L.: 1975, Icarus 24, 311.
Smith, E. J., Davis, L., Jones, D. E., Colburn, D. S., Coleman, P. J., 
Jr.,
Dyal, P., and Sonett, C. P.: 1974a, Science
     183, 305.
Smith, E. J., Davis, L., Jones, D. E., Coleman, P. J., Jr., Colburn, 
D. S.,
Dyal, P., Sonett, C. P., and Frandsen, A. M. A.:
     1974b, J. Geophys. Res. 79, 3501.
Stix, T. H.: 1962, The Theory of Plasma Waves, McGraw-Hill Book 
Company,
New York.

TABLE I
Voyager plasma wave instrument

Fig. 1. Examples of plasma waves and wave-particle interaction 
phenomena detected near earth. Broadband telemetry observations from 
OGO-5 and IMP-6 were used to make up these frequency-time diagrams. 
The arrows show typical locations in the upstream region, at the bow 
shock, and in the near-equatorial region of the magnetosphere. 
Examples of the important high-latitude interactions and wave-particle 
interactions in the proton mode are not shown, but these effects are 
summarized in the lower part of the figure.

Fig. 2. The upper drawings show the JST and JSX (Uranus option) flyby 
trajectories in relation to the anticipated extent of the nominal 
Jupiter and Saturn magnetospheres. The lower panel contains a partial 
listing of important outer planet wave-particle interaction phenomena, 
with rough indications of the regions of interest for the various 
observations.

Fig. 3. Many of the recent models constructed to account for the Io 
modulation of Jupiter's decametric emissions invoke plasma sheath 
effects and field-line coupling via current systems, as sketched here. 
The labels indicate the many ways in which wave- particle interactions 
and plasma instabilities are expected to affect the microscopic 
processes along the field-lines threading the Io sheath and 
ionosphere. Related processes should be important at other satellites 
of Jupiter and Saturn.

Fig. 4. If the physics of the Uranus plasma environment is given by a 
model similar to that discussed by Siscoe (1975), then the 
magnetosphere of Uranus and the important wave- particle interaction 
regions may be as shown (the sketch is adapted from Siscoe's paper).

Fig. 5. The central panel shows how the characteristic gyrofrequencies 
(f[c]^(+/-)) and plasma frequencies (f[p]^(+/-)) should vary with L-
value at Jupiter. These gyrofrequency profiles are derived from the 
Pioneer 10 data (Smith et al., 1974a, b), and the plasma frequency 
profiles are constructed from the theoretical model of Axford and 
Mendis (1974). The Pioneer 10 plasma probe measurements (Frank et al., 
1976) give somewhat higher plasma frequencies within L ~= 6-7 (f[p]^- 
up to about 70 kHz), but this does not require any significant change 
in range for the Voyager mission. The panel on the right contains a 
plausible scaled version for Saturn, and the Voyager PWS frequency 
coverage (16-channel spectrum analyzer plus broad wave-form channel) 
is roughly indicated on the left (the lower eight channels actually 
have 30% bandpass filters, with filter response curves that are 
broader than indicated here).

Fig. 6. The PWS, PRA electronic boxes and the extendible antenna 
elements are mounted as shown on the left. The PWS uses the orthogonal 
10-m elements as a balanced Vee-type dipole, and the effective length 
is about 7 m, as indicated.

Fig. 7. PWS block diagram.

Fig. 8. A frequency-time spectrogram made up using an S^3-A lightning 
whistler recording as input for the Voyager PWS broadband channel. The 
horizontal line is present because of the action of the PWS notch 
filter.