00135-3650 This is a nearby pair of M0.5V dwarfs. With only four measures, the orbit is obviously preliminary, although this system has made two revolution since its discovery by Hipparcos.
00164-7024 This first preliminary orbit lacks coverage.
00258+1025 The pair of G-type dwarfs HDS 57 approaches the periastron, so fast motion is expected in the coming years.
01334-4354 This is the first, but well-defined, orbit with P=7.77 years (Figure 1). The SOAR data sample one full period. The Hipparcos measure is off because of the short period. The primary mass of 1M_{Sun}_ is too large for the spectral type K0.5V listed in SIMBAD. Both components are located on the main sequence in the CMD.
02166-5026 The pair TOK 185, discovered at SOAR in 2011, has covered more than half of its 14 year orbit, including the periastron; the orbit is updated.
02390-5811 This very fast (P=6 years) pair of M0V dwarfs was near the periastron in 2016.9 and has made four revolutions since its discovery by Hipparcos.
04007+2023 This star is a member of the Hyades cluster and is a single-lined spectroscopic binary with a period of 16.7 years (Griffin 2012JApA...33...29G). The SOAR measures cover only a part of the ellipse, but the combined orbit using the Griffin's RVs is well defined (Figure 1). Griffin remarked on the potential resolution of this pair: "The difference in luminosity in the V band may be expected to be around five magnitudes-an impossible prospect visually..." (Griffin 2012JApA...33...29G, p. 48). In fact, the binary was resolved at SOAR a year after the publication of the Griffin's paper. Indeed, {Delta}y=5.2mag, but the pair is far from being "impossible". The secondary component is located well above the main sequence in the CMD in Figure 2, and has V-I_C_=3.45mag. However, the five measurements of {Delta}I have a large scatter of 0.73mag, so the unusually red color of the secondary component needs confirmation. The RV amplitude corresponds to the secondary mass of 0.58M_{Sun}_ if the primary mass is 1.14M_{Sun}_, which is in agreement with the masses in Table 3.
04049-3527 This is a visual triple system where the outer pair A,BC is the 1'' binary I 152 with the estimated period of ~300 years. Although the 52 year orbit of the subsystem B,C is computed for the first time, it is well constrained (Figure 1); this pair, discovered in 1990.9 at 0.48'', passed through the periastron in 2009.7 and now opens up again. Its large eccentricity of 0.92 could be caused by the outer system.
04074-6413 The spectroscopic elements from Jenkins et al. (2015MNRAS.453.1439J) are fixed and only the period and the "visual" elements are fitted. Unfortunately, no RVs are given in that paper for computing the combined orbit. The spectroscopic mass of the secondary component, 0.53M_{Sun}_, matches the mass in Table 3.
04107-0452 This is HIP 19508, HD 26441, or ADS 3041. The previous 20 year visual orbit with high eccentricity is that of Tokovinin et al. 2014 (Cat. J/AJ/147/123). A concordant double-lined spectroscopic orbit based on 36 years of RV coverage has been published by Griffin (2015Obs...135..321G) in a paper that is, unfortunately, ignored by SIMBAD. The orbit here is a combined one based on the Griffin's RVs and the speckle data, with the visual micrometer measures given a very low weight. The combined orbit is highly accurate and deserves the grade 1; its spectroscopic elements (K_1_=11.61, K_2_=12.55, V_0_=26.59km/s) are close to those given by Griffin. The combined orbit leads to the masses of 1.120+/-0.017 and 1.036+/-0.016M_{Sun}_ for the primary and secondary components, respectively. The orbital parallax of 17.22+/-0.31mas is more accurate than the HIP2 parallax 16.09+/-0.65mas. The stars are slightly evolved off the main sequence, hence the masses in Table 3 are somewhat over-estimated. After submission of this paper, we became aware of the work by Docobo et al. (2017MNRAS.469.1096D) who proposed a similar visual orbit. However, they have not computed the combined orbit and apparently used a different weighting scheme. Their orbit is consequently less accurate. While the combined orbit gives a=0.167''+/-0.002 and i=69.0{deg}+/-0.5{deg}, the orbit reported by Docobo et al. gives a=0.161''+/-0.002 and i=66.4{deg}+/-0.5{deg}; the difference exceeds the formal errors.
04422+0259 The updated 58 year orbit of A 2424 is now well defined by the speckle data alone; only the first visual measure is used to constrain the period.
05086-1810 The first orbit of WSI 72 with a remarkably short (for a resolved binary) period of 1 year (Figure 1) is well defined by the 11 speckle measures, leaving the rms residuals of only 1.4 mas (the latest 2016.96 measure at the diffraction limit deviates more, so it was given a low weight). The dynamical parallax of 118.5 mas appears more accurate than the HIP2 parallax of 108 mas, which is obviously biased by the photocenter motion. The deduced masses of 0.3M_{Sun}_ are normal for the M5V dwarfs.
05103-0736 The nearly circular 37.6 year orbit of HIP 24076 by Hartkopf et al. 2012 (Cat. J/AJ/143/42) fits the measures well, although it predicts a too small mass sum of 1.2M_{Sun}_. Double lines were noted by Nordstroem et al. 2004 (Cat. V/117; obsoleted by Cat. V/130). Spectroscopic monitoring by N. Gorynya (2017, private communication) did not show any substantial velocity variability during four years, meaning that the system is not triple. Therefore, the orbit of the visual binary is likely eccentric, with half the period. The new 18.9 year orbit with e=0.82 fits the interferometric data very well (rms residuals 2 mas). It yields the mass sum of 2.7M_{Sun}_, which is appropriate for a couple of F8V stars. Observations around the next periastron in 2018.6 are expected to confirm the eccentric orbit and to refute the circular one. The RV measurements near the periastron can yield accurate masses.
05525-0217 The 12 year orbit of HDS 787, updated from (Tokovinin et al. 2014, Cat. J/AJ/147/123), is now definitive, with a good coverage. Four speckle measures made at the 3.5m WYIN telescope in 1999-2004 were assigned low weights, although the reason why they are discrepant is not clear. The Hipparcos magnitude difference {Delta}Hp=0.87mag is underestimated (in fact {Delta}y=1.6mag), which is normal for such a close (97mas) pair, well below the diffraction limit of the 30cm Hipparcos aperture.
06454-3148 This is a nearby triple system GJ 245.1 where the faint pair of low-mass dwarfs Ba,Bb at 1.4'' from the main star HIP 32366 was discovered in 2005 by Ehrenreich et al. (2010A&A...523A..73E). Apart from the discovery paper, the only published observations are those made at SOAR. The magnitude difference between the F7V primary and the late-M component Ba is {Delta}I=6.57+/-0.47mag, meaning that the faint pair Ba,Bb is just above the detection limit at SOAR. The 6.9 year orbit of Ba, Bb is slightly updated with respect to Tokovinin et al. 2015 (Cat. J/AJ/150/50). With the Gaia parallax of 39.61mas, it corresponds to the mass sum of 0.54M_{Sun}_. The mass of Ba is therefore close to 0.30M_{Sun}_ and implies an M3.5V dwarf. Its magnitude I=11.9mag, according to our photometry, roughly matches the standard relations and the distance modulus of 2.0mag. Given the lack of the V-band photometry, Table 3 adopts the combined magnitude and color appropriate for the masses, making the dynamical parallax meaningless.
06533-1902 The first 36 year orbit of CHR 169 is reasonably well constrained, despite the large gap in its coverage between 1996 and 2014. The pair is actually going through the periastron.
07269+2015 The orbit of CHR 26 by Olevic & Jovanovic (1998SerAJ.158...73O) with P=14.17 years did not match the latest measures; it is revised to P=8.6 years.
08122+1739 This is {zeta} Cnc C, which is a member of the hierarchical multiple system. The 59 year visual orbit of A,B and the crude visual orbit of AB,C (period 1115 years) have been computed previously. Here the preliminary visual orbit of Ca,Cb by Riddle et al. 2015 (Cat. J/ApJ/799/4) is replaced by the combined 17 year orbit using the RVs from Griffin (2000Obs...120....1G). Our orbit roughly agrees with the astrometric orbit computed by Heintz (1996AJ....111..408H). However, only a small sector of the visual orbit is covered by accurate measures, whereas the remaining measures have large errors and fit the orbit poorly. The HIP2 parallax of 39.9mas leads to the large mass sum of 2.45M_{Sun}_ and supports the conclusion of Hutchings et al. (2000PASP..112..833H) that the masses of Ca and Cb are comparable, with Cb being a close pair of M2 dwarfs. The unusually red color of Cb ({Delta}y=4.3, {Delta}I=2.7, {Delta}K=0.2mag, see Figure 2) is explained by its binarity. To reconcile the RV amplitude with the mass sum, we fixed the inclination to 150{deg}. The RV amplitude and the inclination in the combined orbit correspond to the Cb mass of 1.25M_{Sun}_ if the mass of Ca is 1.2M_{Sun}_. The dynamical masses in Table 3 do not account for the binarity of Cb and are therefore incorrect. Further speckle monitoring of this interesting pair is obviously needed.
08403+1921 This is an A9V triple system in the Praesepe cluster (NGC 2635), where the visual secondary B is a double-lined spectroscopic binary with a period of 48.7 days. The outer orbit reported by Olevic (2002IAUDS.146....1O) with P=13.2 years corresponds to the very large mass sum of 54M_{Sun}_, using the Gaia parallax of 4.71mas; it does not match the latest measures. The available data can be represented by a very eccentric (e=0.84) orbit with P=10.6 years or by a nearly circular orbit with P=21.6 years, both with the inclination of 90{deg}. However, neither of those orbits match the separation of 115mas measured in 1982.25 by lunar occultations. Therefore, we prefer the third orbit with a longer period of 35.5 years, which is compatible with the occultation measure, although it fits the speckle data slightly worse than the two shorter-period orbits. The mass sum in the 35 year orbit is reduced to 21M_{Sun}_ with the Gaia parallax or to 6M_{Sun}_ with the HIP2 parallax of 7.15mas. The HIP2 parallax places the primary component on the main-sequence turnoff (Figure 2). The problem with this binary is the lack of speckle monitoring between 1997 and 2016; during this period, only one measure was made. Abt & Willmarth 1999 (Cat. J/ApJ/521/682) measured the RVs of the primary broad-lined component that shows a positive trend during the two year time span of their data. The trend is compatible with all visual orbits and does not help to choose between them. The RVs of the center of mass of the secondary computed from their 48 day spectroscopic orbit have a large scatter and throw some doubt on that orbit, which the authors themselves call tentative.
08447-4238 This is a very well covered orbit with P=2.26 years. The HIP2 parallax is likely biased by the short period.
09252-1258 The 27 year orbit of WSI 73 by Tokovinin et al. 2015 (Cat. J/AJ/150/50) is radically revised here to P=13 years after this pair of K1V dwarfs became unresolved at SOAR in 2015 and opened up in 2016. It will close again in 2019. The new orbit appears secure.
10121-0241 DEL 3 is a pair of nearby M0V dwarfs, for which we propose the first, and still preliminary, edge-on orbit with a short 6.4 year period. The eccentricity is fixed to tune the mass sum.
10294+1211 The orbit reported by Cvetkovic et al. (2016AJ....151...83C) (P=23.36 year) strongly disagrees with the data and is radically revised to P=15.6 years. Although the pair has made nearly two revolutions since its discovery in 1991, the coverage remains scarce.
10529-1717 The updated 15 year orbit of HDS 1556 is now definitive, with both extremities covered. The pair is closing down and will pass through the periastron in 2018.55.
11420-1701 This visual triple system was discovered at SOAR in 2013. The first preliminary 21 year orbit of the inner pair TOK 384 is proposed here (obviously, it still lacks coverage), while the estimated period of the outer 1.1'' companion is about 300 years.
12485-1543 The orbit of the 2.6 year inner pair Aa,Ab, first computed by Tokovinin et al. 2014 (Cat. J/AJ/147/123), is updated using new speckle measures and the unpublished RVs of both components measured by D. Latham (2012, private communication). The outer component in this nearby triple system, at 2.6'' separation, has an estimated period of 400 years. Horch et al. (2017AJ....153..212H) published recently a similar visual orbit, apparently unaware of the paper by Tokovinin et al. 2014 (Cat. J/AJ/147/123). Compared to Horch et al., the coverage is extended here by two years and the RVs are added, making the new orbit more accurate. The combined orbit yields the masses of 0.96+/-0.08 and 0.75+/-0.07M_{Sun}_ and the orbital parallax of 36.7+/-1.1mas.
13137-6248 The 25.6 year period of HDS 1852 is well defined because it is now passing through the same part of the orbit where it was discovered in 1991. Other orbital elements are still preliminary.
14330-4224 The pair of A7V stars HDS 2054 passed through the periastron around 1998 without being observed and is now slowly opening. Its 66 year orbit is preliminary, with a fixed eccentricity. The mass sum of 3.3M_{Sun}_ is less than expected from the spectral type. If the Hipparcos measure is flipped, then the observed motion looks almost like a straight line, although its curvature is still significant. A rough circular orbit with P=222 year and a=0.33'' can be fitted to this arc.
15245-1322 The preliminary 82 year orbit of K7V dwarfs appears reasonably well constrained by the measures at SOAR and by those of Horch et al. (2017AJ....153..212H).
15339-1700 The first 60 year orbit of HDS 2185 by Tokovinin 2012 (Cat. J/AJ/144/56) is updated here, as half of it is now covered.
17176+1025 The orbit reported by Cvetkovic (2013, IAUDS, 181, 1) is revised to P=46 years. This chromospherically active star V2369 Oph is possibly young; it contains an eclipsing subsystem with a period of 0.655^d^.
17195-5004 The measures of FIN 356 with nearly equal components, allowing arbitrary quadrant changes, can be fitted by several different orbits. However, only the 12.9 year eccentric orbit gives a reasonably large mass sum of 2M_{Sun}_. Its major axis is oriented toward us, the inclination is 99{deg}, hence the apparent separation is always less than the true semimajor axis. The elements e and i are strongly correlated. Interferometric and RV monitoring of the next periastron in 2023.6 is needed. The orbit predicts the RV difference of ~30km/s near the periastron.
17447-4244 We reprocessed the observations of FIN 341 made at SOAR in 2008 and 2009 to verify that the quadrant was opposite to that of 2015-2017, which is in agreement with the proposed 15 year orbit. One of the two archival speckle measures made in 1989.3 fits the orbit nicely, while the separation of 0.15'' measured on 1991.39 strongly contradicts both the orbit (which predicts the separation of 20mas) and the non-resolution by Hipparcos. We discard the 1991.39 speckle datum and use the interferometric measures by W. Finsen made in 1959 and 1963. This object belongs to the {lambda} Bootis class of chemically peculiar stars (Paunzen et al. 1997, Cat. J/A+AS/123/93), making the measurement of its mass particularly interesting. The masses of 1.9 and 1.5M_{Sun}_ deduced from the orbit are slightly less than expected for normal A2V stars.
18040+0150 This is HD 165045, a double-lined spectroscopic binary resolved at SOAR. The RVs measured by D. Latham (2012, private communication) are used together with the speckle measures in the combined orbit with P=1.6 year. Its small inclination prevents an accurate measurement of the masses; the orbital parallax is 29.2+/-3.4mas.
18520-5418 The Hipparcos astrometric binary was first resolved at Gemini in 2012 by Tokovinin et al. (2013AJ....146....8T) and then followed by SOAR. The period of this preliminary orbit is fixed to 8 years. One can not help noting the discrepancy between the Gaia parallax of 13.77mas, the HIP2 parallax of 19.5mas, and the dynamical parallax of 21.5mas. The Gaia parallax leads to the unrealistically large mas sum of 6.2M_{Sun}_ and is therefore suspect. With a physical companion at 146'', the system is triple.
19377-4128 Both extremities of the 55 year circular orbit of VOU 34 are covered by accurate speckle measures. The pair has made 1.5 revolutions since its first resolution in 1936. The moderate {Delta}V~0.4mag and the opposite quadrants in 1991 (Hipparcos) and 2015 exclude the alternative eccentric orbit with half the period.
19598-0957 The combined 5 year orbit computed by Pourbaix (2000A&AS..145..215P) had large residuals to the SOAR measures, prompting its revision. The new combined orbit uses the published RVs and is well defined (grade 2). Its small inclination and correspondingly small RV amplitudes do not help accurate measurement of the masses and orbital parallax, which is 40.0+/-5.8mas.
20212-5147 The Hipparcos binary HDS 2097 is quite fast, with P=10 years and a well-defined first orbit. The star is on the Magellan program of exoplanet search (Arrigada 2011, Cat. J/ApJ/734/70).
21099-2424 The first 6.9 year orbit of HIP 104476 is well defined by the SOAR data. SIMBAD calls this star "pre-main sequence"; it is an X-ray source. The components are located on the main sequence, however.
21368-3043 The preliminary orbit of VOU 35 by Tokovinin et al. 2014 (Cat. J/AJ/147/123) is revised here using recent observations. The pair has a relatively close companion C, discovered at SOAR at 0.45'' from A; it shows rapid orbital motion (the estimated period of AB,C is ~100 years).
21400-5222 The SOAR data cover the same part of the 21 year orbit where the pair was resolved by Hipparcos one revolution earlier. The new orbit is preliminary. The components are nearly equal; both are located on the subgiant branch in the CMD in Figure 2, in agreement with the F7III spectral type.
21504-5818 The first 17 year orbit of HIP 107806 is well covered. The mass sum computed with the Hipparcos parallax of 24.1mas is 1.4M_{Sun}_, which is appropriate for the G6V spectral type. The dynamical parallax in Table 3 is smaller and the derived masses are correspondingly larger.
22228-2937 HDS 3172 is a fast pair of G6V dwarfs with P=10.5 years. Its first orbit (Figure 1) is well constrained. The resolution in 2016.9, at periastron (38 mas), is tentative. Now the pair opens up again.
22535-1137 This is the bright triple system HR 8704 (74 Aqr) where the primary component of the interferometric pair A,B (MCA 73) is a 3.4 day double-lined spectroscopic binary (Catanzaro & Leto 2004A&A...416..661C). The 19 year outer orbit was first computed in 1993 and updated several times since, the last one by Mason et al. (2010AJ....140..735M). This is a typical case where the measurements can be modeled either by a highly inclined orbit with moderate eccentricity (the 19 year orbit) or by a very eccentric orbit with half the period. The ambiguity can be resolved when the true quadrants are known, which is not the case for binaries with small {Delta}m and for classical speckle observations. The quadrants of MCA 73 with {Delta}m~0.6mag can be determined from the speckle image reconstruction, which has been used at SOAR since 2015. We found that in 2015 and 2016, the quadrants differed from the orbit prediction, whereas they matched in 2008.76. This archival observation was reprocessed to confirm the quadrant; also see the adaptive optics measurement by Schoeller et al. (2010A&A...522A..85S). Therefore, the 19 year orbit is wrong, as the pair is always resolved in the same quadrant. The 9.5 year orbit with e=0.86 is computed here. Catanzaro & Leto (2004A&A...416..661C) give the RVs of the center of mass of the close pair (component A) and one measurement of the secondary B. These RVs support the eccentric orbit (Figure 3), which predicts a large RV variation near the periastron. The RV amplitudes are K_1_=10.0km/s and K_2_=19.6km/s; the systemic velocity is -2.7km/s. The RV amplitudes and inclination correspond to the mass sum of ~15M_{Sun}_;however, the large error of the inclination, i=30{deg}+/-17{deg}, makes the spectroscopic masses quite uncertain. The HIP2 parallax of 3.95mas, also not very precise, leads to the mass sum of 16.6M_{Sun}_. The orbit thus roughly matches the estimated mass sum of three B9V stars, ~11M_{Sun}_.