The Nuclear and Ionic Charge Distribution Particle Experiments on the ISEE-1 and ISEE-C Spacecraft D. HOVESTADT, G. GLOECKLER, C. Y. FAN, L. A. FISK, F. M. IPAVICH, B. KLECKER, J. J. O'GALLAGHER, M. SCHOLER, H. ARBINGER, J. CAIN, H. HOFNER, E.KUNNETH, P. LAEVERENZ, AND E. TUMS Abstract-The ISEE-1 and ISEE-C instruments have been designed to measure the elemental abundances, charge state composition, energy spectra, and angular distributions of energetic ions in the energy range 2 keV/charge to 80 MeV/nucleon and of electrons between 75 and 1300 keV. By covering the energy range between solar wind and low-energy cosmic rays the instrument will fill a gap in the knowledge especially of the nuclear and ionic composition of solar, interplanetary, and magneto- spheric accelerated and trapped particles. The instrument consists of three different sensor systems: ULECA is an electrostatic deflection analyzer system with rectangular solid-state detectors as energy deter- mining devices, its energy range is ~3 to 560 keV/charge; the ULEWAT is a double dE/dX versus E thin-window flow-through proportional counter/solid-state detector telescope covering the energy range from 0.2 to 80 MeV/nucleon (Fe); the ULEZEQ sensor consists of a com- bination of an electrostatic deflection analyzer and a thin-window dE/dX versus E system with a thin-window proportional counter and a position- sensitive solid-state detector. The energy range is 0.4 MeV/nucleon to 6 MeV/nucleon. While the ULECA and the ULEWAT sensors are de- signed mainly for interplanetary and outer magnetospheric studies, the ULEZEQ sensor will also obtain composition data in the trapped radia- tion zone. 65 rates and pulse-height data can be obtained with sector- ing in up to 16 sectors. 1. INTRODUCTION OUR ISEE EXPERIMENTS have been designed to measure the elemental abundances, charge state composition, en- ergy spectra, and angular distributions of energetic ions in the energy range 2 keV/charge to 80 MeV/nucleon and of elec- trons between 75 and 1300 keV. At the lower end of the en- ergy range, the instruments will measure the ionization states of heavy ions in the high speed (>500 km/s) solar wind; while throughout most of the upper energy range (0.3 to 80 MeV/nu- cleon) we will determine the energy spectra, anisotropies, and composition of energetic solar particles as well as magneto- spheric and interplanetary ions. In a limited energy range be- tween 0.4 and 6 MeV/nucleon, a simultaneous determination of the ionic and nuclear charge is possible. II. SCIENTIFIC OBJECTIVES The low-energy portion of interplanetary and magneto- spheric particles (energies 1 keV/nucleon to several hundred keV/nucleon) has been virtually unexplored to date. Measurements in this region, however, are expected to provide sensitive probes of solar, interplanetary, and magnetospheric phenomena. Our investigation, which will provide the first detailed measurements of low-energy ions (and also measurements of low-energy electrons and higher energy ions) is thus expected to yield significant new results. Examples of areas of likely impact include the following. A. Solar and lnterplanetary Phenomena 1) Solar Flare Acceleration Mechanism: The mechanism by which ions are accelerated in flares is not presently known. Important information on this mechanism, however, is available from studies of the energy spectra of the particles which the mechanism produces. This is particularly true in the case of heavier ions at low energies. These particles are not expected to be fully stripped of their electrons. Thus in the same energy range different ions cover different rigidity intervals. By examining, then, the energy spectra of many ions at low energies, and also electrons, we will place stringent constraints on the rigidity dependence of the flare acceleration process. 2) The Location of the Flare Acceleration Region: Low-energy ions also provide in their charge states information on the region in the corona where they are accelerated. The charge of solar particles is "frozen-in" as the particles leave the corona; the plasma density in interplanetary space is too low to cause additional ionization or recombination. Thus the charge composition, particularly of very low energy ions, can reflect the temperature of coronal electrons in the region where the particles are accelerated, propagate, or are stored, For example, if the ions are accelerated in the flare site itself, we should expect a high degree of ionization, appropriate to the high temperature in the flare. In contrast, if the particles are accelerated in the surrounding corona by, e.g., plasma disturbances emitted from the flare, we should expect a charge composition more similar to that of the solar wind. 3) Coronal Propagation and Storage: Low-energy ions also provide some of the most detailed information on coronal propagation mechanisms, and coronal storage. Measurements in the broad rigidity range covered by partially stripped ions at low energies, and also by low-energy electrons, will provide stringent tests of the current idea that coronal propagation is rigidity-independent. Further, ionization loss which is a consequence of extended coronal storage is most evident at low energies, where it should produce in the differential energy spectra a systematic flattening which depends on the charge squared to mass ratio. From the flattening observed in the spectra of different ions, it should be possible to determine the time during which the particles are stored and/or the coronal density and thus the location of the region where the storage occurs. 4) Compositional Variations in Solar Flares: The composition of energetic flare particles is known to vary widely from flare to flare, particularly at low energies. The cause of this variability is not presently known. It may result from the fact that heavier ions, which may not be fully stripped of their electrons, can exhibit different rigidities in different flares, and thus will respond differently to the flare acceleration process. It may be also that the variations reflect compositional anomalies in the coronal material from which the particles are accelerated. Our investigation, which can determine the charge states of low-energy ions, will probe the former explanation in detail. From these charge-state measurements, which can indicate the coronal conditions where the particles are accelerated, and from our general ability to observe flares in considerable detail, we will also provide information on the latter possibility. Of particular interest in the study of compositional variations in solar events are the so-called 3He-rich event, in which the 3He/4He ratio can exceed unity. These flares are one of the more dramatic examples of compositional anomalies since the coronal abundance is 3He/4He <10^-3. These flares also have the peculiarity that there is no accompanying increase in deuterium and tritium, which are equal byproducts of 3He in spallation reactions, and there is a general enhancement in heavier elements, particularly in iron. Our investigation will extend the measurements of 3He and heavier ions in these flares to much lower energies than has been possible to date. 5) Interplanetary Propagation: The current theory for en- ergetic particle interactions with the magnetic field in the solar wind is inadequate in that it predicts more pitch-angle scatter- ing than is observed. These differences are most pronounced, and thus most easily studied at low energies (< 1 MeV/nucleon). We will make a detailed study of propagation at low energies by observing the anisotropies of protons and helium through- out our energy range, as well as the time profiles of different ion species and electrons. The former measurement is a sensitive indicator of the extent of the scattering; both mea- surements provide information on its rigidity dependence. 6) Interplanetary Acceleration: It appears from studies in recent years that the majority of co-rotating particle streams are the result of interplanetary acceleration, as opposed to of direct solar origin. The origin of the particles, however, is not presently known. It may be that the particles are accelerated out of the solar wind. It is possible also that the particles enter the solar wind as energetic solar particles and receive here an additional acceleration. Our measurements of the spectra, anisotropies, and charge states of ions, over an es- sentially continuous energy range from the solar wind to high- energy particles, should provide stringent tests of these pos- sible origins. If these particles are accelerated out of the solar wind, their charge states should be those of solar wind ions, and their spectra a continuous extension of the solar wind distribution. Particles which originate as more energetic solar particles, in contrast, may be more highly ionized and may exhibit a behavior at higher energies which is uncorrelated with that closer to solar wind energies. 7) The Anomalous Cosmic-Ray Component: From 1972 on, a component with the anomalous composition of only helium, nitrogen, oxygen, and neon has been observed in the cosmic-ray flux at energies 1-30 MeV/nucleon. The origin of this component is presently being debated, although there is mounting evidence that it results from interstellar neutral particles which are swept into the heliosphere and ionized and accelerated in the solar wind. This component appears to be a feature of solar minimum conditions. Our observations over the next few years will record the behavior of this component as solar activity increases with the onset of the new solar cycle. 8) Correlated Studies with Deep-Space Missions: Our in- vestigation will also provide measurements at earth for cor- related studies of galactic cosmic-ray modulation and solar flare propagation with deep-space missions such as Pioneer and Voyager. B. Magnetospheric Acceleration and Trapped Particles 1) Origin of the Trapped Radiation: The origin of the en- ergetic ions comparing the radiation belts has not been un- ambiguously determined. Both the earth's ionosphere and the solar wind are possible candidates for the trapped radia- tion. A very sensible test that can be employed to decide whether the source of the trapped radiation is the ionosphere or the solar wind is the measurement of the C/O abundance ratio. This ratio is about 0.5 in the solar wind and less than 10^-5 in the ionosphere. Other sensitive tests are the deter- mination of the charge states of He and O. In the solar wind He+/He++ is approximately equal to 10^-3, in the ionosphere He+/He++ is approximately equal to 10^2; also the O+/O+6 ratio is very small in the solar wind and very large in the ionosphere. However, charge exchange with the residual atmosphere is an important process for low-energy radiation belt ions and, depending on energy, can obscure the charge state of the source particles. 2) Transport Processes in the Magnetosphere: Although charge state measurements of radiation belt particles are only of limited use as an indicator of their source, they are im- portant in order to study diffusion processes in the mag- netosphere. Radial and pitch angle diffusion rates depend on mass and charge of the diffusing particles and this de- pendence is different for different diffusion mechanisms, so that knowledge of the charge state is essential for any quantitative work concerning loss and source processes. A very important measurement in this respect will be the ob- servation of the charge state evolution of specific ions after impulsive injection events. 3) Origin of the Ring Current: There is strong evidence that ions heavier than protons play a fundamental role in the ex- traterrestrial ring current and are, therefore, important in the dynamics of the magnetosphere. The origin of the ring current particles is unclear; observations of the Lockheed/Palo Alto group indicate that O+ is the dominant species in the storm- time ring current, suggesting an ionospheric source. On the other hand, the same group observed during a magnetic storm precipitating helium ions which were primarily doubly charged. Charge state measurements over a wide energy range for a number of different ions are, therefore, essential in determin- ing the origin of the ring current. 4) Energetic Magnetospheric Bursts: Bursts with mildly rela- tivistic electrons and energetic ions up to ~1 MeV have been re- peatedly observed in the geomagnetic tail. The nature of these bursts is still unclear, although they are believed to be gen- erated by rapid field line merging and correspondingly large- amplitude cross-tail electric fields. By examining the energy spectra of bursts of ions with different charge states we will place stringent constraints on the rigidity dependence of the acceleration mechanism. 5) Energetic Particles Upstream of the Bow Shock: Par- ticles in the energy range 30 to 100 keV have been observed continuously in that part of the upstream interplanetary medium that is connected to the earth's bow shock by the interplanetary magnetic field. It is not clear at present whether these particles are protons or heavier ions. Identification of these upstream particles will shed light on the acceleration process and will answer the question whether these particles are originally solar wind particles or very-low-energy cosmic- ray particles which are further accelerated. III. DESCRIPTION OF THE DETECTOR SYSTEMS The detector system for each of the experiments on ISEE-1 and ISEE-C consists of three sensors, each of which is optimized for a given energy range and sensitivity. A brief description of these sensors is given below. A. ULECA The ULECA (Utra Low Energy Charge Analyzer) sensor incorporates techniques of electrostatic deflection and a total energy measurement to provide the charge composi- tion of ions in the energy range 2 to 560 keV/charge. The simplified top and side cross-sectional views of ULECA are shown in ufigure1. ULECA represents an adaption of the de- sign of the University of Maryland ECA (Energy Charge Analyzer) instrument flown on IMP 7 and 8 satellites [1]. Low-energy ions pass through a multislit focusing collimator and enter one of three deflection regions designed by L, M, and MP. Seven rectangular surface-barrier (Au-Si) solid-state detectors are placed at fixed positions at the exit of the de- flection regions, each defining a given energy per charge window. The output of each of these detectors is pulse- height analyzed provided the solid-state anticoincidence detectors are not triggered. A number of counting rates are also recorded as shown in Table I. A residual background (due to cosmic-ray produced secondaries) and the geometrical factor of the collimator determine the minimum flux which can be measured by each detector (see Tables I-IV). The majority of penetrating particles is eliminated from analysis by the solid-state anticoincidence detectors. Ufigure2 shows a computer simulation of the response of detector M2. The assumed abundances are H+:He+1:O+5:O+7::100:20:0.5:0.5. The following effects were included: 1) dispersion caused by different entrance angles, 2) fringe field effects, 3) collimator dispersion, 4) electronic and detector noise, 5) nuclear defect in the solid-state detector, and 6) the differential intensity spectral index (assumed to be -3 in the figure). TABLE I RATE CHANNEL CHARACTERISTICS ULECA SENSOR ____________________________________________________________________________ Rate Readout Designation Particle Energy Range Period Conversion Factor Min. Flux (N/S)(1) Type(2) (keV/charge) (S)(3) (cm^2*sr* (cm^2*sr* (4) keV/charge)^-l s*keV/ charge)^-1 ____________________________________________________________________________ Ll (S) Q >= 3 2.88-3.08 16 581 290 8.68-9.29 191 95 L2 (S) Q >= 1 8.14-9.38 16 94 47 26.4-30.5 28 14 M11 (N) Q = 1 27.4-32.7 8 18.5 1520 (S) 32 5 M12 (N) Q = 2 27.4-32.7 8 18.5 8 (S) 32 3 M13 (N) Q > 2 27.4-32.7 32 18.5 61 (S) 64 15 M21 (N) Q = 1 53-69 8 5.35 17 (S) 64 1.3 M22 (N) Q = 2 53-69 8 5.35 32 0.7 M23 (N) Q > 2 53-69 64 5.35 2.2 M31 (N) Q = 1 103-140 8 2.52 298 32 0.7 M32 (N) Q = 2 103-140 8 2.52 0.3 M33 (N) Q > 2 103-140 32 2.52 0.2 MP12(N) Q = 1,2 105-560 64 0.33 3.5 MP3L Q >= 3 105-560 64 0.33 0.2 MP1L Q = 1 105-560 32 0.33 3.3 MP2L Q = 2 105-560 32 0.33 0.1 (1) N = nonsectored; S sectored into 8-45 degree sectors in ecliptic plane. (2) Q = charge state of ion. (3) At high bit rate divide period by 4. (4) L1, L2 energy range assumes medium voltage mode. The major effect not included is the secondary electron background in the solid-state detector. This effect is caused by the interaction of high-energy particles in the spacecraft. Our experience with similar detector systems on IMP's 7 and 8 indicate that this will not be an important effect for mag- netospheric events. In the L deflection region, a variable power supply steps the deflection voltage in 32 logarithmic increments from 600 to 1550 V, stepping once every 5 spin periods (~16 s). The voltage range may be changed up or down by 50 percent via ground command. Two low-noise (15-keV energy threshold) rectangular solid-state detectors L1 and L2 measure the energy and record the counting rate of deflected ions for each voltage step. The relative energy per charge windows delta(E)/E are 0.07 and 0.15 (FWHM), and the energy ranges over which measurements are made are 1.8 to 9.5 and 5.1 to 30.5 keV/charge for L1 and L2, respectively. In the M and MP deflection region, two high-voltage supplies are used to provide deflection fields of 1.5 and 6.7 kV/cm, respectively. At the exit of the M section 3 rectangular solid- state detectors, M1, M2, and M3 measure the fluxes and anisotropies of protons, alpha's and Q >= 4 ions in the energy range 25 to 140 keV/charge. These detectors are also pulse- height-analyzed for more detailed charge spectra. In the MP section, a single position-sensitive rectangular Si detector is used to determine the charge spectrum of ions (H to Fe) from 105 to 560 keV/charge. B. ULEWAT The ULEWAT (Ultra Low Energy Wide Angle Telescope) is a dE/dx versus E type particle composition telescope con- sisting of two thin-window flow-through proportional counter delta(E) (energy loss) elements P1 and P2 and three large-area rectangular solid-state detector elements D1, D2, and A of various thicknesses as shown in ufigure3. Depending on the range of the incoming particles, the two proportional counters P1 and P2 and the first detector D1, or, for higher energies, P2, D1, and D2 form a double dE/dx versus E (residual) tele - scope. A large area Li-drifted detector is used as an anti- coincidence to veto analysis of penetrating particles. This detector system is similar to the ULET (ULtra Low Energy Telescope) flown by our group on IMP 7 and 8 [2]. A pro- portional counter hodoscope provides information on par- ticle arrival directions off the satellite equatorial plane and allows for path length corrections in the delta(E) counters which, along with the E detectors, are pulse-height analyzed. Ufigure4 shows a preliminary two-dimensional dE/dx versus E (right panel) and a (dE/dx) P1 versus (dE/dx) P2 printout (left panel) from ULEWAT obtained on ISEE-1 during a 2-h period of a small solar particle flux increase on November 1, 1977. Basic rate information determined by the various logic and energy threshold conditions on the delta(E) and E detectors, and detailed pulse-height information is telemetered allowing TABLE II RATE CHANNEL CHARACTERISTICS ULEWAT SENSOR _____________________________________________________________________________ Rate Designation Particle Detector Energy Range Readout (N, S, U/D) Type Combination (MeV/nuc) Period(s) (1) (2) (3) (4) (5) _____________________________________________________________________________ LP(N, S) protons POS-P1-P2 0.17-0.4 8, 64 LA(N, S) alphas POS-P1-P2 0.12-0.2 8, 64 LH1(N, S) Z >= 1 P1 0.1-0.17(p) 8, 64 LH2(N, S) Z > 2 POS-P1-P2 e.g., N:0.1-0.14 8, 64 Fe:0.025-0.05 MPA(N, S)(6) protons POS-P1-P2-D1 16, 32 alphas MP1(N, S, U/D) 0.48-0.66 8, 64, 64 MP2(N, S, U/D) protons POS-P1-P2-Dl 0.66-1.20 8, 64, 64 MP3(N, S, U/D) (8) 1.2-2.1 8, 64, 64 MA1(N, S, U/D) 0.31-0.36 8, 64, 64 MA2(N, S, U/D) alphas POS-P1-P2-P1 0.36-0.47 8, 64, 64 MA3(N, S, U/D) 0.47-0.68 8, 64, 64 MA4(N) 0.68-1.37 16 MA5(N) 1.37-2.55 16 MA6(N) 2.55-5.2 16 MH(N, S, U/D) Z > 2 POS-P1-P2-D1 32, 64, 32 MX(N)(7) Z > 2 POS-P1-P2-D1 64 HA(N) alphas POS-D1-D2 5-20 16 HN(N) Z > 2 POS-D1-D2 1-80(Fe) 32 HP1(N) protons D1-D2 5-10 16 HP2(N) 10-20 16 HA1(N) 5-7.5 32 HA2(N) alphas POS-D1-D2 7.5-11 32 HA3(N) 11-20 32 El(N, S) 0.075-0.115 8, 32 E2(N) electrons D1 0.115-0.30 16 E3(N) 0.30-0.44 16 E4(N) electrons D1*D2 0.44-1.30 64 (1) Geometrical factor of ULEWAT is approximately 2 cm^2 * sr. N = nonsectored; S = sectored into 8-45 degree sectors in the ecliptic plane. U/D = full spin accumulation of rates of particles arriving up or down with respect to ecliptic. (Average elevation angle 47 degree.) (2) Z = atomic number of particles. (3) POS: hodoscope; P1, P2: first and second proportional counter delta(E) detectors; D1: solid-state E detector; D2: solid-state detector. (4) Preliminary. (5) First, second, and third entry refer to nonsectored, sectored, and up-down rate readout periods, respectively. (6) Basic rate used with pulse-height data; energy range variable, depending on particle type. (7) See Table III. (8) Threshold on POS raised to exclude protons if POS single rate exceeds 6 x 10^4 counts/s. precise determination of the three-dimensional anisotropies. energy spectra, and elemental composition (isotopic up to O) of particles in the 0.1- to 20-MeV/nucleon energy range and of electrons from 75 to 1300 keV. In addition, we employed a MATRIX-rate system, which uses two-dimensional pulse- height informations from P2 and D1 to determine on board 16 selected ranges in nuclear charges and energy per nucleon. This-system allows for a high temporal resolution rate mea- surement for a well matched set of the most abundant ele- ments individually (Li + Be + B, C, O, Ne + Mg + Si + S, Fe). The ULEWAT sensor is also capable to measure electrons. Electrons with energies between ~75 keV and ~1300 keV are measured in four rate channels, contiguous in energy. The four rate channels are designated E1(~75-115 keV),E2(~115- 300 keV), E3(~300-440 keV), and E4(~440-1300 keV). The geometrical factor is ~2.0cm^2 * sr. Anticoincidence requirements placed on the proportional counters and solid- state detectors in ULEWAT ensure that all four electron rates are free of contamination by protons or heavy ions. For example, P21 rejects all protons with energies less than ~2 MeV, D11 simultaneously rejects protons between ~450 keV and ~150 MeV, and A simultaneously rejects all protons >18 MeV. The E1 and E2 rate channels are singles rates requiring a signal in the D1 detector, while the E3 and E4 channels require a simultaneous signal in both the D1 and D2 detectors. C ULEZEQ The third sensor in the experiments is the ULEZEQ (Ultra Low Enerergy Z, E, and Q) detector system whose top and side, cross-sectional views are shown in ufigure5. ULEZEQ is literally a combination of a deflection analyzer system similar to TABLE III ENERGY RANGES OF THE ULEWAT MX (MATRIX) RATE CHANNELS ____________________________________________________________________________ Rate Particle Energy Range Channel Type (MeV/nuc) ____________________________________________________________________________ 1 Li, Be, B 0.75-2.0 (Be) 2 C, O 0.3-0.7 (O) 4 0.46-0.98 6 carbon 0.98-1.45 8 1.45-2.50 5 0.7-1.0 7 oxygen 1.0-1.4 9 1.4-2.5 10 0.3-0.7 11 Ne-S 0.7-1.5 (S) 12 1.5-2.5 13 0.2-0.64 14 Fe group 0.64-1.0 (Fe) 15 1.0-1.5 3 1.5-2.56 TABLE IV RATE CHANNEL CHARACTERISTICS ULEZEQ SENSOR _____________________________________________________________________________ Energy Geometrical Minimum Rate Particle Range Readout Factor Flow Designation Type (MeV/nuc) Period(s) (cm^2* sr) (cm^2*sr*s (1) (2) *MeV/nuc) _____________________________________________________________________________ UH1(3) Z >= 1 >O.1 32 ~0.02 UH2(3) Z >= 1 0.45 (p) 32 ~0.02 UHP1 protons 0.45-1.2 64 ~0.02 0.006 UHP2 protons 1.2-3.0 64 ~0.02 0.002 UHA helium 0.26-3.0 64 ~0.02 0.002 UHH(3) Z > 2 >0.3 (O) 64 ~0.02 >0.2 (Fe) (1) All rates are nonsectored in the interplanetary mode. (2) Z = atomic number of particles. (3) Integral rate channel. ULECA and of a simplified version of the ULEWAT dE/dx versus E system. A single deflection region consisting of a multislit focusing collimator, 20-kV power supply and de- flection plates, and a position sensitive solid-state detector comprise the deflection analyzer portion of the sensor. A thin-window flow-through proportional counter placed in front of the position-sensitive detector is used as the delta(E) element and together with the position-sensitive detector represents the dE/dx versus E telescope of ULEZEQ. Pulse- height analysis of the delta(E), energy, and position signals pro- vides the information from which the energy, nuclear charge, and charge states of each individual particle in the energy range ~0.4 to 6 MeV/nucleon may be determined. Ufigure6 shows a two-dimensional display of an accelerator run with sulphur ions scattered off a thick target. The lower panel shows the dE/dx versus E matrix of PC*D, the upper panel the position signal x*E versus the energy signal E in D. For the lower track, the 20-kV deflection voltage is off, and for the upper track on. IV. DATA OUTPUT The scientific data from all three sensors consist of two kinds: a) counting rate data and b) pulse-height analyzer events. A. Counting Rate Data Each detector is monitored by one or more threshold dis- criminators whose outputs are passed through a system of coincidence-anticoincidence logic which generates 65 dif- ferent counting rates corresponding to a variety of detector combinations, particle energy windows, etc. The number of accumulated counts for any particular rate is stored in one 24-bit register in a 256 register memory. (There are two such memories, identical in all respects, for redundancy. For the most part only one memory is used with the second remaining as a backup, with data storage able to be switched between the memories by command.) Individual rates are read out at predetermined intervals and logarithmically com- pressed to either 10 or 12 bits and inserted into the telemetry stream. Some rates are accumulated continuously over all directions during the spacecraft spin (omnidirectional rates), others are sorted into 8 directional angular sectors so as to measure flux anisotropies (sectored rates). Of these rates, the BASIC rates are of special importance. Their logic require- ments are identical to the ones required to trigger individual pulse-height analyzed (PHA) events. Absolute fluxes can, therefore, be computed from PHA events. A list of rate desig- TABLE V SUMMARY OF PULSE-HEIGHT-ANALYSIS DATA _____________________________________________________________________________ Required Directional Detector Analyzed No. of Gain Information Sensor Combination Signal Channels Range (No. of Sectors) _____________________________________________________________________________ ULECA none L1 256 1 16 L2 256 1 16 M1 1024 1 1 M2 1024 1 16 M3 or MB 1024 1 16 MP energy 256 2 MP position 256 2 16 ULEZEQ(1) PC*D PC 256 2 16 D energy 256 2 16 D position 256 2 ULEWAT(2) P1(3) 256 2 16 POS*P1*D1 P2(4) 256 2 16 or D1 256 3 16 POS*P2*D1*D2 D2 256 2 16 (1)Proportional counter. (2)Hodoscope. (3)First delta(E) proportional counter. (4)Second delta(E) proportional counter. nations and associated information grouped according to sensor subsystem is given in Tables I-IV. Due to the large number of rate channels generated, the telemetry of in- dividual rates involves a rather complicated submultiplexing scheme, which cannot be described here. B. Pulse Height-Analyzed Events Each pulse-height analysis (PHA) consists of a comprehensive measurement of the energy deposited by a single incident particle in one or more detectors as it passes through a par- ticular sensor, together with information on which detectors were triggered and directional information about the particle. The energy signals from the various detectors are pulse-height analyzed with either a 1024 channel analyzer or 256 channel analyzers with multiple (2 or 3) gain levels to provide the re- quired dynamic range. The complete description of each such analyzed event requires a relatively large number of bits so that, in general, the number of events which can be telemetered is limited to one event each time the associated storage register is read out. For most detector channels, then, at high rates in a flare event or within the magnetosphere, only a sample of the incident particle distribution can be analyzed. A list of pulse-height-analyzed detectors and relevant information grouped according to sensor subsystem is given in Table V. C Special Features 1) Rotating Priority Scheme: The counting rates provide an absolute measure of the flux of particles whose detailed characteristics are sampled by the PHA. Since the objectives Of this instrument concern the relatively rare heavy ion events, a system of priorities is imposed on the so-called "event selec- tion logic" ESL by the rotating priority system. This system prevents the telemetry from being dominated at high rates by any one particular type of particle. 2) On-Board Pulse-Height Accumulation: A unique feature Of the pulse-height-analyzed data for the L1 and L2 detectors in the ULECA sensor is that in addition to the sampling of pulse heights by a 256-channel analyzer in the analysis mode described above, a 32-channel (nonlinear) analysis is per- formed on those events occuring after the telemetry registers are full awaiting readout and the resulting distribution is ac- cumulated in 32 registers of the memory. Only one detector (L1 or L2) can be accommodated at a time, however, so each distribution is accumulated for a full voltage step on one detector and then switched to the other detector for the next step. V. APPENDIX(1) THE ISEE-A AND ISEE-C GAMMA-RAY BURST DETECTORS T. L. CLINE(2), G. GLOECKLER, D. HOVESTADT, AND B. J. TEEGARDEN(2) Gamma-ray bursts were discovered in 1973, using the Vela satellite monitors of nuclear test explosions, by Klebesadel, Strong, and Olson of the Los Alamos Scientific Laboratory. The phenomenon they found consisted of brief ( 1/10 to 30 s ) but intense (10^-5 to 10^-4 erg cm^-2) bursts of hard X-ray or soft gamma-ray photons (energy > 100 keV.). These events triggered their instruments only several times per year and came from directions in the sky that had no correlation with other unusual astronomical observations and, therefore, no obvious source identification. The directions of arrival of the photon wavefronts were determined by triangulation, i.e., by the timing delays at the four satellites used, assuming speed of light propagation and knowing the detector positions in their 120,000-km radius circular orbits. Given the timing delays of up to 800 light*ms, and given typical timing accuracies of wavefront arrival of about 50 to 100 ms (limited by counting rate statistics, rather than clock stability), the resulting source locational accuracies were only a few degrees at best. Since that discovery over four years ago, no definitive progress in the identification of the burst source objects or in the under- standing of the emission mechanism has yet been possible, due to the facts that all other observations possible at the time of discovery were also circumstantial and that it has taken several years to plan and execute new space-flight investigations spe- cificary designed to investigate this phenomenon. The first experimental advance possible resulted from the in- stitution of a NASA policy endorsing appropriate modifica- tions of space-flight instruments under construction, but de- signed for other purposes, in order to make gamma-ray burst observations. Simultaneously, experiments were proposed for various missions not yet having selected payloads. The ISEE-A and -C missions were two of the obvious candidates for modification, as well as Helios-2, HEAO-A, and -C, and several others in the European space programs. The Solrad- 11 and Pioneer-Venus missions, in addition, included new experiments for burst detection. The goal of this overall policy was to instrument as many spacecraft as possible, so as to maximize the probability of having several in orbit simultaneously in order that further triangulation measure- ments could be made. Improved directional accuracies would result if the temporal accuracy were maximized in each instru- ment, and if at least some of these detectors were placed in interplanetary space, the advantage accrued of having much longer baselines, thereby making greater relative timing ac- curacy possible. Both the ISEE-A and -C modifications satisfy the require- ment of maximum possible temporal accuracy. Also, since the ISEE-C spacecraft will be placed in an orbit at 0.01 AU, or about 10 times that of the typical distant elliptical earth orbiter, the near-interplanetary baseline objective is also achieved. In addition, the ISEE-C instrument will perform high-resolution gamma-ray burst spectroscopic measure- ments, an exciting possibility that could go far in describ- ing some of the physical parameters involved in the source mechanism. The ISEE-A instrument is more or less a modification of the Helios-2 gamma-ray burst detector, changed to fit the telemetry format of the ISEE host apparatus. In this ar- rangement, an intensity increase in either of the two sensors used can cause a trigger to take place, freezing the circulating memory of the immediate past counting rate history and filling another memory with the counting rates for 1 min follow- ing the trigger. The time of trigger and its location in the temporal history are also placed in memory. All recorded information are then read out at a very low bit rate during the succeeding several hours. A number of commands are available to adjust the various gains and sensitivity levels of the sensors and trigger counts, making possible a flight adjustment for optimum signal-to-noise and minimal dead time on background triggers. The sensors incorporated are a 4-cm diameter photomulti- plier Csl-scintillator system supplied by the Max-Planck- lnstitut, and a 6-cm^2 solid-state (cadmium telluride) array supplied by Goddard Space Flight Center, the first NASA satellite use of CdTe gamma-ray detectors. Each of these have command-adjustable gains, and each can be turned on or off in flight. Three triggers are used, in order to take into account the variability of burst temporal histories and rise times: one is based on the total counts in 4 ms, the next on a 32-ms basis, and the slowest on a 256-ms basis. Each of these triggers is independently command-adjustable to fire on some number of counts between 1 and 1027. Typical settings for Hehos-2 have been 51 counts in a quarter of a second, 24 in 1/32 of a second and 9 in 1/256 of a second. These are not multiples of one-eighth, due to the greater fluctuations in the background for smaller numbers, i.e., non-Poisson, due to the variety of secondary particle and gamma-ray interactions in the spacecraft. Six memories are used, three before and three after trigger, on 4-ms, 32- ms, and 256-ms time bases, for 1/64th, 1/8th, and 1-min total storage, respectively, making possible the accurate study of the rise-time structure without having a contin- uously high storage requirement. This scheme was shown, on Helios-2, to make possible the detection and accurate study of every gwnma-ray burst that triggers the Vela and/or Solrad systems. ISEE-A has, since launch, triggered on at least one solar flare X-ray event, and is expected to provide timing data for most known burst events that will take place when it is outside the magnetosphere. The ISEE-C gamma-ray burst detection system is a second- generation improvement over the ISEE-A instrumentation. The principal difference is the addition of the capability of performing high-resolution spectroscopy on the spectrum of gamma-ray burst photons between 0.05 and 6.5 MeV. To ac- complish this we have incorporated on ISEE-C a radiatively cooled high-purity germanium detector. In addition, we also analyze signals from the central Csl crystal in the MEH elec- tron experiment. Both temporal and spectral information are obtained from this detector. It is a large crystal and repre- sents the most sensitive gamma-ray burst detector on the spacecraft. Finally, we obtain temporal information from a smaller CsI crystal in the HOH experiment. A cross-sectional view of the cooler/sensor is given in ufigure7. Its basic structure is divided into two parts, an outer stage and an inner stage. The outer stage is designed to operate at an intermediate temperature (~160 K) and thereby provide a thermal buffer between the sensor (residing in the inner stage) and the spacecraft. The outer stage has a basic conical shape to define a field of view (126 degrees full angle) for the inner stage. This field of view must not contain the sun, earth, or any sig- nificant portion of the spacecraft. The mechanical interface between the outer stage and the spacecraft is via a st:.pport ring. This support ring will be in good thermal contact with the spacecraft and be close to room temperature. As the outer stage radiatively cools (via its radiating surface shown in ufigure7) it will contract slightly. The support points are designed such that this contraction causes a break in the thermally conductive path between the support ring and the outer stage, thereby then-nally isolating it from the space- craft. The inner conical surface of the outer stage is highly polished specular aluminum so that no sunlight can be scat- tered into the inner stage. The germanium crystal is housed in the inner stage of the cooler in a hermetically sealed magnesium enclosure. The ex- posed surface is coated with a highly emissive white paint to allow efficient radiation of heat. The inner stage is mechani- cally supported in a similar fashion to the outer stage so that a second level of thermal isolation is attained. The predicted equilibrium operating temperature of the inner stage is 101 K (-172 C). The germanium crystal itself is 4-cm diameter by 3-cm depth and will have an energy resolution of < 3.5 keV at 1 MeV. It is made from hyperpure or intrinsic germanium. The use of this material permits the crystal to be stored indefinitely at room temperature without sustaining any loss in performance. The weight of the cooler/sensor/electronics package is ~2.6 kg. A charge-sensitive preamplifier and bias supply are located in a small package immediately adjacent to the cooler/sensor so that lead length and capacitance will be maintained. A 4096-channel ADC is also contained in this package which digitizes the germanium signals. This is a lightweight low- power ADC which has already been developed for the Caltech- Goddard cosmic ray experiment on MJS. The digitized sig- nals will be fed to the gamma-burst digital instrumentation in the low-energy cosmic-ray (HOH) experiment. The digital electronics in the HOH experiment contains a 10^5 -bit memory that is partitioned into three sections. Two of these are devoted to storage of temporal histories of gamma- ray bursts. Any of the three (2 CsI and 1 germanium) can be programmed into either of these two portions of the memory. Each of these time history memories (THM's) has trigger cir- cuitry associated with it that samples the incoming count rate and determines if it exceeds a ground-commandable threshold. Instead of accumulating the number of counts in a fixed time interval the electronics measures the time interval over which a fixed number of counts are accumulated. This number is selectable by command in binary increments between 1 and 128. In addition, the clock used for time interval measure- ment has ground-selectable frequencies of 1, 2, 4, and 8 kHz. Each of the THM's has 2000 12-bit words so that each can store 2000 X N events when N is the fixed number of counts accumulated. This will permit the complete storage of the time histories of nearly all of the gamma-ray bursts that one expects to occur. The remainder of the memory is devoted to the storage of spectral information from either the MEH CsI crystal or the germanium detector. This part of the memory is organized into 3072 16-bit words. Of these 16 bits, 12 are used to store pulse height and 4 are for time tag. The full 12 bits are used for the germanium spectrum storage giving 4096 channel resolutions. For the MEH CsI crystal only 512-channel (9-bit) analysis is performed. The clock frequency used for the time tag is independently ground programmable in binary increments between 64 and 2048 Hz. This is to allow the time tagging to be optimally adjusted according to the detector background counting rate. When a trigger occurs, all three memories are allowed to fill. When full they can be dumped either automatically or by ground command. Because of the low bit rate (1.5 bits/s) allocated to the gamma-ray burst instrumentation it takes 20 h to complete a full memory dump. It is, however, possible to have only partial dumps by ground command. ACKNOWLEDGMENT The authors are grateful to the many individuals and institu- tions who contributed to the successful design and fabrica- tion of the instrumentation. They are especially grateful to O. Vollmer who contributed substantially to the basic design. They greatly acknowledge the work of F. Eberl, W. Goebel, G. Hartmann, N. Heinecke, W. Lang, E. Seidenschwang, and H. Waldleben of the Max-Planck-Institut fur extraterrestrische Physik and that of R. Cates, G. Umberger, R. Lundgren, L. Reed, A. Galvin, R. Sciambi, and B. Lambird of the University of Maryland. For the integration into the ISEE-1 and ISEE-C spacecraft, and the successful post-launch performance, they thank the ISEE project office of the Goddard Space Flight Center, especially M. A. Davis, J. Madden, K. Ogilvie, T. von Rosenvinge, J. Hrastar, G. Kowalski, and R. Beatty. They acknowledge the cooperation of the DFVLR, Bereich Pro- jekttrigerschaften, especially R. D. Giinther and B. Klinkmann. REFERENCES [1] C. Y. Fan, G. Gloeckler, and E. Turns, "An electrostatic deflection versus energy instrument for measuring interplanetary particles in the range 0.1 to ~3 MeV/charge," in Cosmic Ray Conf Papers (Hobart, Tasmania), vol. 4, p. 1602, 1971. [2] D. Hovestadt and 0. Vollmer, "Satellite experiment for detecting low energy heavy cosmic rays," in Cosmic Ray Conf. Papers (Hobart, Tasmania), vol. 4, p. 1608, 1971.