The Solar Probe ANalyzer—Ions on the Parker Solar Probe

SPAN-I's electrostatic analyzer (ESA) geometry draws its heritage from the STATIC instrument on the MAVEN spacecraft (McFadden et al. 2015). The top-hat toroidal approach to ESAs (Carlson et al. 2001) used for SPAN-I was originally designed for the Cluster mission (Reme et al. 1997) and successfully flown on the FAST satellite (Klumpar et al. 2001). The advantages of the top-hat design are its large geometric factor, optimal field of view, adequate energy resolution (dE/E 7%), and optics that allow exiting ions to be properly imaged by subsequent sectors. For SPAN-I, the electrostatic focal point is shifted from the exit grid of the ESA to the entrance of the −15 kV acceleration region to optimize the particle throughput within the TOF optics. The ESA's outer hemisphere is held at ground while the inner section is biased up to −4 kV, which provides an energy range between 125 eV and 20 keV for the first encounters. UV sunlight contamination and particle scattering off of the outer surface is reduced with the addition of Ebanol-C coating and scalloping features. As the exit of the ESA is close to the −15 kV HV acceleration sector, it is necessary to add a pair of grids at the exit of the ESA in order to reduce fringe fields. In front of the ESA aperture are two deflectors that allow the elevation angle of the instrument to be increased by up to +/−60° at energies as high as 4 keV per charge.

In front of the ESA aperture are two deflectors that allow the elevation angle of the instrument to be increased by up to +/−60° at energies as high as 4 keV.

SPAN-I is capable of measuring a large dynamic range of particle fluxes in the solar wind by using two modes of attenuation: a mechanical attenuator and an electrostatic spoiler. The mechanical attenuator is mounted between the deflectors and the ESA aperture. Before and during launch, the attenuator remains in a closed position, together with the one-shot TiNi cover, to prevent detector contamination and acoustic damage to the carbon foils. After launch, the cover is opened and the mechanical attenuator is allowed to move the multislit metal shield in and out of the ESA FOV by using a series of nanomuscle shape-memory alloy (SMA) actuators. The slits allow a reduction in ion fluxes by a factor of 10. In addition to the mechanical attenuator, SPAN-I includes an electrostatic spoiler, serving as an additional electrode that forms the lower half of the outer hemisphere (the upper half is maintained at ground). When the spoiler is held at a particular voltage (maximum of 80 V), the distribution of ions traveling through the analyzer is reduced by electro-optically narrowing the energy per charge passband. Initial calibration testing has shown the spoiler to be capable of reducing the ion flux to background levels, assuring an additional safety mechanism for saturation of the detector. Final calibration of the energy-dependent geometric factor is yet to be determined. Both attenuator mechanisms are under software control that monitors specific instrument parameters, such as counting rates and system attenuator state. Once the count rate exceeds a preset threshold, a set of logical and sequential combinations of the attenuation mechanisms are activated to maintain an ideal count rate. When transitioning the mechanical attenuator, the actuating nanomuscles require a 5 minute relaxation time to allow for thermal settling.

The mechanical TOF design is a direct copy of the TOF used on STATIC/MAVEN. The design uses two sets of carbon foils for both the START and the STOP signal generation, which simplifies the mechanical design by allowing a separation of the TOF HV region (−15 kV) from the MCP detector voltage (3 kV). In order for ions with a mass per charge heavier than H⁺ and He⁺⁺ to penetrate two carbon foils with high enough efficiencies, we selected ultrathin foils (<1 μg cm⁻²) combined with a post-acceleration voltage of −15 kV. The −15 kV TOF HV supply also produces a secondary voltage (11/12 of the full voltage), enabling the deflection of secondary electrons generated by the first set of carbon foils toward the START anodes below the TOF section. The carbon foils at the entrance and exit of the TOF analyzer are additionally shielded by grids to suppress field emissions generated by impurities and tears within the carbon foils.

ACF-Metals was the primary provider of the carbon foils. The production begins with the foils mounted on standard mica slides, which were placed on top of a cantilever base and lowered into a hot-water bath containing a surfactant solution. The floating carbon foil is then retrieved by replacing the now empty glass with the stainless steel frame containing a 333 line/inch grid mesh and raising the cantilever base above the water surface. Once removed, the foils were vacuum baked and then scanned for impurities using software developed for the MAVEN mission. The selection process of the foils included a thorough review of the high-resolution scans, the associated software results, and a calibration test using ion species of different mass per charge values.

The SPAN-I anode board is located below a Z-Stack MCP configuration and detects the START and STOP electrons. The flight MCPs have a resistance of 45 MΩ and a nominal gain of 2–3 × 10⁷. The gain is adjustable by controlling the MCP bias voltage through software commands and is continuously monitored in flight by performing multiple MCP-Gain tests at every encounter. The electron cloud generated at the bottom of the MCP is collected by a series of 11 inner discrete anodes (STARTs) and 16 outer discrete anodes (STOPs). Each of the 27 discrete anodes is connected to a dedicated CFD located close to the anode to reduce the signal travel time. The CFD enables high-resolution timing-of-arrival signals independent of the input pulse amplitude. The 11 inner (START) anodes have an angular size of 225 spanning a total azimuthal FOV of 2475. The 16 outer anodes (STOPS) are separated into 10 high-resolution anodes (1125) and 6 larger anodes (225). The STOP anodes permit a finer azimuthal resolution and are aligned such that the higher-resolution area is pointed toward the solar-wind direction.

The SPAN-I digital board contains the main instrument field-programmable gate array (FPGA) and four individual TOF chips, each having four input signals from both a START and STOP CFD for a combined 16 TOF measurements. The TOF chip acquires the input signals from the CFDs and passes on the processed results to the FPGA for further analysis. The FPGA is the main processing unit of the instrument and includes functions such as command execution and science data production. It communicates directly with the SWeap Electronics Module (SWEM), using the provided storage for data archiving and potential delivery to the spacecraft. The digital board also houses a set of MRAM and SRAM memory, where the digital-to-analog converter (DAC) control values for the HV components, the associated sweeping tables, and data acquisition schemes are stored. More about the instrument data acquisition scheme is detailed in Section 3.

There are two SPAN-I high-voltage power supply boards (HVPS) that operate the HV electrodes of the instrument. The first HVPS supplies high voltage to the hemisphere, spoiler, and both deflectors with voltage values controlled and set by a digital-to-analog converter (DAC) chip with a 4 V reference on the digital board. The stepping, or sweeping, from one voltage value to another, occurs every 0.2 ms for the full sweep (0.8 ms for the targeted sweep) with a voltage settling time of <1 ns. The second HVPS supplies high voltage to the MCPs and the TOF section, once again set via DACs, but held at nominal values pending further calibration. For the deflector supplies, however, the DAC is referenced relative to the hemisphere supply control voltage. Using this coupled DAC control voltage technique results in deflector biases scaled to the correct value for each hemisphere voltage step.

The LVPS generates 1.5V, 3.3V, +/−5V, +/−8V secondary voltages from the 22V supplied by the SWEM. As a fail-safe mechanism, the 22V source is routed through the backplane to a socket with a high-voltage enable plug. RIO ASIC monitors are also mounted to monitor the voltage and current draws.

The SPAN-I backplane board has several functions: 1. Provide high-voltage signals to the spoiler and deflector; 2. Transmit actuation commands to the cover mechanism and mechanical attenuator; 3. Provide access to the enable plug for ground testing and instrument safety.

SPAN-I is designed to perform a sweeping sampling of ions at a constant rate. A sweep is composed of either 1024 steps (Full) or 256 steps (Targeted), changing the instrument optics with each step by altering associated voltages and therefore sampling specific regions of phase space. There are a total of four sweeps that occur every "New York" second, which is derived from a 19.2 MHz clock and subdivided into bins to form an integration time of 2²⁴/19.2 × 10⁶ (0.874 s). A sweep therefore happens every 218.45 ms, alternating between a Full sweep and a Targeted sweep. A Full sweep uses high-voltage steps that allow for a coarse mapping of the entire range of the energy per charge space with the drawback of having regions where the spectrum is not sampled. This drawback is addressed with the Targeted sweep. Once the Full sweep is completed, the FPGA determines which bin contained the maximum number of counts and selects the appropriate Targeted table for a high-resolution scan around this region. Full sweeps contain a total of 1024 steps, which are reduced to a 256 bin product by summing to every fourth step (microstepping). The Targeted sweeps do not have microstepping and simply step through 256 bins in the same amount of time. Figure 5 shows example sweeps of the Full and Targeted data acquisition modes.

Zoom In Zoom Out Reset image size

Each of the four high-voltage electrodes (hemisphere, deflectors, spoiler) sweeps through voltages to sample the ambient plasma. The sweeping mechanism is controlled by the FPGA, which reads DAC values from lookup tables residing on the instrument SRAM memory and sets the voltage accordingly. There are a total of one HV lookup table (LUT) and two index LUTs: the Sweep HVLUT, Full IndexLUT, and Targeted IndexLUT. The Sweep LUT contains the 4 DAC values for controlling the hemisphere, the spoiler, deflector 1, and deflector 2. These values are all interspersed so that a reference to the start address of the first DAC setting locates the remaining three DACs. There are a total of 4096 DAC values, 16 bits long, for each of the four electrodes. The Sweep LUT is referenced with two tables, the Full–LUT and the Targeted LUT, which contain the correct indexes to perform a "coarse" and "targeted" measurement, respectively. The Full–LUT contains 1024 index values for the 1024 steps it sweeps through during a single cycle, with a microstepping feature that reduces the product to 256 bins by summing every fourth step. The Targeted LUT, on the other hand, contains 256 index values for the 256 steps it sweeps through during a targeted sweep cycle. Targeted sweeps focus around the previous high-voltage step where the peak counts occurred. Therefore, there are 256 separate tables of 256 indexes in the Targeted LUT based on the 256 steps where the peak can occur.

The TOF measures the time between START and STOP pulses from a single anode for a particular angle and energy per charge. The value measured is originally a value between 0 and 2047 representing the delays of 0 to 208.33 ns. Each count of the output represents 101.725 ps in delay, such that a value of 512 from the TOF converts to a value of 52.1 ns in delay. Before the TOF value is passed onto the data processing unit the FPGA either accepts the 9 most significant bits (MSBs) of the TOF, discarding the 2 least significant bits (LSBs), or it compresses the 10 MSB into 9 bits (discarding the LSB). The compression scheme is N for counts less than 256, N/2+128 for counts between 256 and 511, and N/4+256 for counts greater than 511. This compression emphasizes TOF resolution at low TOF values. The TOF value is further categorized into a mass per charge value by using a mass lookup table (MLUT) derived from ground calibration. For each energy per charge setting the 9 bits of the compressed TOF (cTOF) values are indexed into a 512 element table. Instead of using the full range of high-voltage settings for each energy per charge step (65536), the MLUT uses 128 tables based on the 7 most significant bits of the hemisphere DAC. The table converts the TOF values into 64 distinct mass per charge bins.

The first step in generating science products is converting the time-of-flight measurement to a mass per charge. Ions entering SPAN-I are first filtered by their energy per charge, meaning that the particle travel time will change for each energy step across the 2 cm TOF gap. For each energy per charge setting, a separate lookup table is used to convert the particles time of flight into one of 64 distinct mass per charge values. The next table, the Mass–Range–LUT, categorizes the 64 mass per charge bins into four separate mass products defined as: 0-Protons, 1-Alphas, 2-Higher M/Q, 3-Background. Finally, after the particle mass is categorized into a mass range, the FPGA will use a Mass–Bins–LUT to determine how much mass resolution to keep for each ion product. The result is an address space in memory defining a specific ion mass that will be filled with the appropriate science values.

The second step converts the high-voltage steps (energy and elevation), the anode number (azimuth), and the product number (mass address) into an address to increment and be filled with count measurements. The resulting products are then summed over a programmed number of sweeps (either all Full sweeps or all Targeted sweeps) defined in a Sum–LUT table that holds the exponent (n) of a 2ⁿ value.

The SPAN-I analyzer performs a single high-voltage sweep in 0.248 s, during which a set of voltages are applied to the hemisphere, deflectors, and spoiler based on a sweep lookup table. The first step within the sweep sets the hemisphere voltage to its highest value and subsequently steps through the remaining 31 values logarithmically toward the lowest voltage. For each hemisphere step, the deflectors are stepped through a series of voltages in order to scan in elevation for a specific energy per charge. Lastly, a spoiler voltage is set for each hemisphere step in order to reduce the total flux of incoming particles.

A slow rotation scan is performed across all 16 anodes with a 1 keV residual gas beam and −15 kV TOF acceleration. This is to test the azimuthal analyzer response and the associated START and STOP carbon foil efficiencies, which are a measure of the carbon foil secondary electron production efficiency as a function of the ion mass. Multiplying the START and STOP efficiency yields a measure of the total efficiency of the TOF section. The analyzer response function for each individual anode is shown in Figure 6 on the left. The anodes are drawn proportional to one another, but not to scale, and the normalized calibration measurements overlay the corresponding azimuthal angles. For larger anodes 225 (10–15) the relative efficiency is close to 90% with 5% cross talk interference between adjacent anodes. For the smaller STOP anodes of 1125 (0–9) a start anode of 225 is shared, and the signal is divided into its separate components. In this case, the cross talk between anodes is larger and closer to 30% while the relative efficiency stays high. The right side of Figure 6 shows the same anode configuration with carbon foil efficiencies for the STOP and START signals overlaid. The carbon foil efficiencies are derived from the following equation:

$$\begin{eqnarray*}&&{\mathrm{Start}}_{\mathrm{eff}}=(\mathrm{ValidRate})/(\mathrm{StopRate}),\end{eqnarray*}$$

$$\begin{eqnarray*}&&{\mathrm{Stop}}_{\mathrm{eff}}=(\mathrm{ValidRate})/(\mathrm{StartRate}),\end{eqnarray*}$$

where START_eff (STOP_eff) is the START (STOP) efficiency, ValidRate is the valid events rate, and the StartRate (StopRate) is the valid START (STOP) rate. The results show a fairly consistent carbon foil efficiency for both START (50%) and STOP (23%) across all anodes. At the edge of each anode pair there is a slight drop in both efficiencies due in part to the aforementioned cross talk and grids within the TOF section that are in the sensor's FOV. Cosmic rays and radioactive decay background is found to be minimal due to coincidence measurements with ion fluxes as high as 20 kHz.

Zoom In Zoom Out Reset image size

In order to avoid ion feedback and to improve the signal, a 500 nm thick Kapton foil was included after the second set of thin carbon foils. Ions that have passed through both thin carbon foils with an energy of >15 keV are stopped by the thick foil, whereas secondary electrons pass through almost unhindered. The improved signal comes from the scattering of the secondary electrons themselves as the pass through the thick foil, where the narrow beam is now spread over a greater area on the MCP and therefore reduce MCP droop.

The secondary electron yields of protons and helium are estimated to be two electrons from the thin carbon foils (Goruganthu & Wilson 1984; Ritzau & Baragiola 1998). Higher masses, with the same acceleration potential, will typically yield higher numbers of secondary electrons. A more detailed discussion can be found in McFadden et al. (2015). More calibration of efficiencies for different mass species and acceleration will follow in order to improve Venus flybys.

The analyzer concentricity is verified using a series of energy scans for each anode separately. Figure 7 shows test results for anodes 0–3, with the analyzer sweeping logarithmically from 5 eV–20 keV in order to verify the peak tracking mechanism for large energy steps in the full sweep. The top panel represents the targeted scan with peak tracking enabled, while the bottom panel represents the full sweep. Both panels show a clear tracking of the energy beam, even during instances when the beam energy was in between two full sweep energy bins.

Zoom In Zoom Out Reset image size

The analyzer k-factor is derived for each anode separately and is shown in Figure 8. Conversion of energy bins from instrument values to physical units in eV is achieved by assuming a linear relationship between the hemisphere voltage and the energy of the particle (k-factor)

$$\begin{eqnarray}&&E/Q=k\ast V.\end{eqnarray} \tag{ 1 }$$

Zoom In Zoom Out Reset image size

The results show a consistent k-factor for the first six anodes that closely matches the expected simulated value of 16.7. Anodes 7–15 appear to be slightly lower in value, highlighting a slight nonuniformity between the two concentric plates.

The TOF system is verified by testing the mass per charge resolution of individual ion species for a series of different energies. Figure 9 shows the ion travel time in nanoseconds for 1 keV H⁺, ${{\rm{H}}}_{2}^{+}$, He⁺, N⁺, O⁺, Ne⁺, ${{\rm{O}}}_{2}^{+}$, and Ar⁺. A clear separation is visible between H⁺ and ${{\rm{H}}}_{2}^{+}$ by up to 4 orders of magnitude, where ${{\rm{H}}}_{2}^{+}$ is a proxy of He⁺⁺ in the solar wind. The $\tfrac{{\rm{\Delta }}M}{M}$ are measured to be 15% and 20% H⁺ and ${{\rm{H}}}_{2}^{+}$, respectively. This value slightly changes for different particle energies and is taken into account using the mass–energy lookup table. SPAN-I is also capable of measuring higher-mass species such as O⁺ and CO₂ ⁺, which is ideal for measuring escape ions during the Venus gravity assists.

Zoom In Zoom Out Reset image size

In order to determine the correct ion travel time, and therefore the true mass per charge of the particle, it is important to include the energy loss component as the ions traverse the carbon foil. The resulting effect is a slower travel time and a broadening of the mass per charge distribution of individual species, commonly known as straggling. The peak of each distribution from Figure 9 is plotted in Figure 10 together with the corresponding ion mass per charge, taken from the calibration measurements of the ion gun. In addition, four simulations are plotted: The black curve represents the ion travel time with no carbon foil present, while the light blue, blue, and purple lines represent simulations using the "Stopping and Range of Ions in Matter" SRIM/TRIM software (Ziegler et al. 2010) using carbon foil thicknesses of 1.5, 1.0, and 0.5 μg cm⁻², respectively. The results show an agreement between the SPAN-I TOF results and the results for the 1.0 μg cm⁻² foil thickness, indicating a carbon foil thickness twice as thick as the nominal value.

Zoom In Zoom Out Reset image size

Operation of SPAN-I can be divided into two main modes based on spacecraft operations: the primary "encounter" phase of the orbits, which is approximately 10 days centered around PSP perihelion (variable by orbit profile), and the rest of the orbit, hereafter called the "cruise" phase. The instrument measurement rate is higher and uninterrupted during nominal encounter phases as compared to cruise phases, during which the data rate is considerably lower and the measurement periods are interrupted by spacecraft communications, power limitations, and other spacecraft critical operations. During periods of interest (e.g., Venus encounters), the sensors can be configured to collect more data than typically when outside of the encounter.

Level 0 (L0) files are unprocessed files downlinked directly from PSP through the Deep Space Network (DSN) in their original packetized format created by the spacecraft. Files contain a fixed volume of data and are named based on their date of acquisition. On their own, Level 0 files are not useful for scientific analysis but are archived for troubleshooting purposes.

Level 1 (L1) files are converted from the binary L0 format into a format readable by a standard data processing environment, such as IDL or Python packages. SPAN-I uses IDL routines in the data production pipeline to convert L0 files into L1 CDF files. To produce L1 files, minimal processing is performed as the intention of the L1 data is to serve as an archive of the instrument performance in its most raw state. All quantities in the CDF files are in engineering units, e.g., particle counts per accumulation period per energy bin number, deflection bin number, and anode number. Because of the units, L1 files are not useful for scientific analysis. The intention behind archiving L1 files is to keep a record independent from scientific conversions for pipeline debugging purposes and instrument calibration consistency checks over the course of the mission. Housekeeping values are converted into temperatures, currents, and voltages. L1 files are available by request.

Level 2 (L2) data files are generated from L1 files. Instrument units are converted into physical units. For example, the counts per accumulation period are converted into differential energy flux as a function of the energy in electron volts (eV), and deflection and anode bin numbers into degrees in ϕ or θ. The L2 data coordinates remain in the instrument frame of reference. Level 2 data are released to the public for scientific analysis.

Level 3 (L3) products are based on functions performed on L2 files or other processing that expands or reduces the number of dimensions of the L2 data set. Ion moments and fits, which produce values of the density, temperature, and velocity, are classified as L3 products as they are a combination of modified L2 data.

Figure 11 shows proton measurements for all of Encounter 2. The top panel shows the differential energy flux spectrogram of H⁺ ions summed over all FOV angles. The second panel shows proton measurements along the elevation angle of the instrument (deflectors) summed over the energy per charge and azimuth. The proton flux is clearly centered around 0° corresponding to a plane that aligns with the Sun line. The third panel shows the azimuthal FOV of the instrument, where each bin represents a small (1125) or large anode (225). Anode 0 is currently not included due to its partial obstruction by the TPS. The angle range from 16875 to 180°) requires additional flight calibration in order to determine the correct flux. The last panel shows distance from the Sun in million kilometers.

Zoom In Zoom Out Reset image size

The 3D VDFs produced by SPAN-I are akin to the ones reported in Marsch et al. (1982), who showcased field-aligned proton beams measured by Helios. Figure 2 of Verniero et al. (2020) similarly demonstrated the evolution of a proton beam during an ion-scale wave storm from Encounter 2. A single time slice, 2019-04-05/19:54:20, from that event is shown in Figure 12. The left panel shows individual energy sweeps for each anode and deflection combination plotted separately with the Alfvén velocity overplotted with a black dashed line. The middle and right panels of Figure 12 represent 2D contour elevations sliced through the ϕ and θ plane, respectively. The black arrow shows the orientation of the magnetic field, where the length of the arrow is the Alfvén speed and the head is placed at the SPC measured solar-wind velocity. Here, we refer to the Alfvén speed of the total proton distribution. Two separate proton distributions are clearly visible: the core and the beam. The "core" is defined as the population centered around the peak in the phase-space density, while the "beam" component comprises the tail.

Zoom In Zoom Out Reset image size

The middle panel of Figure 12 illustrates how much of the proton distribution lies in SPAN-I's FOV; in this particular time slice, we see that the core is partially visible. The 2D cut through the θ plane, in the right panel of Figure 12, reveals a dramatic field-aligned beam featuring a separate peak (green) that is markedly distinct from the core. The left panel of Figure 12 shows that the beam well surpasses the Alfvén speed. Previous observations of VDFs featuring differential flows between different ion populations (Feldman et al. 1973, 1974; Marsch et al. 1982; Marsch & Livi 1987; Neugebauer et al. 1996; Steinberg et al. 1996; Kasper et al. 2006; Podesta & Gary 2011) are known to drive the VDFs unstable, subsequently leading to wave generation (Daughton & Gary 1998; Hu & Habbal 1999; Gary et al. 2000; Marsch 2006; Maneva et al. 2013; Verscharen & Chandran 2013; Verscharen et al. 2013). The preliminary instability analysis conducted by Verniero et al. (2020) underscores the ability for SPAN-I data products to study fundamental processes that govern energy transfer mechanisms in the solar wind, such as wave–particle interactions.

Based on ground calibration, the beam measurement is completely unaffected by the presence of alpha particles at that same energy and are thus resolved without interference. Throughout the encounter, the appearance and disappearance of the proton beam can be attributed to the limited FOV of the instrument, hence the additional presence of SPC.

He⁺⁺ measurements can be see in Figure 13. For comparison, the first panel shows the proton spectrogram for Encounter 2. The second panel shows the He⁺⁺ differential energy flux spectrogram, once again summed over all FOV angles. A clear proton contamination is visible within the data that appear below the He⁺⁺ distribution. In addition, the highest energy bin contains counts from the energy sweep retrace as the instrument does not engage a dead time during the cycling of the voltage table. Roughly 1% of protons leak into the He⁺⁺ mass channel and present a large enough contamination that matches the density of He⁺⁺. This issue is addressed by taking the proton channel and subtracting a time-varying percentage of the flux from the He⁺⁺ channel. The bottom panel of Figure 13 shows the results of the subtraction algorithm performed on Encounter 2. The proton contamination is greatly reduced, and the He⁺⁺ distribution becomes apparent.

Zoom In Zoom Out Reset image size

SPAN-I is located behind PSP's TPS and can only observe the partial distribution function of the solar wind. The first 8° closest to the spacecraft z-axis are completely obstructed, meaning that anode 0, which has a 1125 azimuthal size, is 70% covered. For Encounters 1 and 2 the the instrument mostly measured the wings of the solar-wind distribution function, with occasional intervals when the full distribution was visible.

During first turn on, it was discovered that one of the instrument sweep tables was corrupted. This was evidenced by the fast housekeeping, which monitors the HV supply, and the resulting failed checksum. This issue was immediately addressed by selecting a backup table to avoid a total loss of data for the first encounter. The backup table has an energy range from 1 to 4 keV, resulting in the solar-wind beam flowing in and out of the instrument energy range.

Commissioning of SPAN-I, including its configuration and HV ramping, was limited in time due to spacecraft maneuvers that placed the Sun behind the heat shield closely after launch and therefore out of the FOV. The spacecraft did perform a transient slew in order to obtain the solar wind into the FOV of SPAN-I, which lasted 20 minutes. The test was used to confirm full functionality of the instrument.

When ions travel through the TOF they initially collide with the first set of carbon foils that generate the START signal. The interaction with the carbon foil causes a slight loss in kinetic energy; therefore, ions are measured to travel slower than their expected velocity. This straggling effect is especially noticeable in high flux beams such as protons, to the point that enough energy is lost to appear within the alpha product. This issue is addressed for the alpha channel by subtracting a percentage of the proton channel, as straggling protons appear at the same energy per charge in both channels.

Part of the background that is being measured by the sensor originates within the MCPs, such as radioactive decay of the glass and cosmic-ray penetration. This background was at most 10 Hz and does not significantly contribute to the overall valid events. Another source of constant background comes from coincidence measurements, where the START and STOP pulses are triggered by two different ions. This has been found to be on the order of 1% for fluxes 100 kHz and scales up for larger fluxes (McFadden et al. 2015). Ghost peaks are peaked signals in TOF spectrometers that are found in addition to the expected m/q values but in higher or lower mass bins. Their source is generally attributed to the charge state of ions that exit the carbon foil and to the production of secondary electrons. The SPAN-Ion sensor includes a thick foil placed after the ultrathin STOP carbon foil that prevents several of these spurious ghost peaks to form. For instance, an ion that emerges from the STOP foil as a neutral without generating a secondary electron (30% chance) can generate a valid TOF signal by striking the MCP directly and causing a delayed TOF. This effect is eliminated as the ion will impact the thick foil and generate a nonvalid event that gets filtered by the electronics. Similarly, an ion emerging with a negative charge is also prevented from striking the MCP by the thick foil. While neutral and negative ions are filtered out, positively charged ions still have a small probability of being reflected by the thick foil back onto the carbon foil, thus generating a secondary electron and therefore a delayed TOF. In addition, a molecular ion such as ${{\rm{H}}}_{2}^{+}$ can breakup at the START carbon foil resulting in particle components that emerge with a smaller kinetic energy and once again a delayed TOF signal. Both of these ghost peaks are seen in the H+ and ${{\rm{H}}}_{2}^{+}$ mass peaks in Figure 9.

The high-voltage sweeps are arranged so that the hemisphere (energy per charge) is held at a constant voltage value (starting with the highest within the series) while the deflectors are swept in one direction (see Figure 5). When plotting individual energy spectra, a clear hysteresis is observed, where the sweeping of the deflector voltages lags and differs depending on the direction in which it sweeps. This results in a slight offset in the deflection angle from the predetermined table values for alternating energy sweeps. This effect will be addressed in future in-flight calibrations.