COMMISSION 27 OF THE I. A. U.
         INFORMATION BULLETIN ON VARIABLE STARS
                      Number 3390

                                         Konkoly Observatory
                                         Budapest
                                         25 October 1989

                                         HU ISSN 0374 - 0676


 RECENT MINIMA OF THE SUBDWARF ECLIPSING BINARY BD -7d 3477 (=HW Vir)


During a photoelectric UBV survey of UV-bright objects, Menzies & Marang (1986)
discovered the B-type subdwarf BD -7d3477 (=HW Vir) to be an eclipsing binary
with a very short period of 2h48m. In a preliminary analysis, Menzies &
Marang (1986) determined the temperature of the primary to be 26000K and the
secondary probably ~4500K, the radii of both components to be near 20% of their
separation, and plausible masses of 0.25 and 0.12 solar masses. This model
suggests that the system could have passed through a "common envelope" stage
and is probably similar in evolutionary status to AA Dor (e.g. Hilditch &
Kilkenny 1980; Paczynski 1980; Kilkenny 1986).

Because of the possibility of eventually detecting the effects of gravitational
radiation (only on timescales of a few hundred years) and, more realistically,
of seeing the effects of mass loss or mass exchange, it has seemed worthwhile
to establish accurate ephemerides for these evolved binaries by continued
monitoring.

Menzies & Marang (1986) used 27 timings of primary minima obtained in 1984-85
to establish a rather accurate ephemeris for the HW Vir system and in this note
we report four further timings from 1989. The data were obtained with the 0.5m
telescope and modular photometer of the South African Astronomical Observatory.
For each eclipse, a series of 20-second continuous observations with a Johnson
B filter was made, with occasional breaks for sky background measurement; a
typical primary eclipse result is shown in Figure 1. From these data, the
following times of primary minima were determined:

         HJD      Est.error    Cycle

2447684.32597   +-0.00003     16739
   7687.24396     0.00003     16764
   7688.29443     0.00005     16773
   7689.22817     0.00003     16781

where the cycle number is based on the Menzies d Marang (1986) ephemeris and
the observation at cycle 16773 is less accurate because the early part of the
descending branch of the eclipse was missed. A linear least squares solution,
including the Menzies & Marang data gives an ephemeris:

HJD (primary min) = 2445730.556074 + 0.1167196311*n
                        +-0.000014 +-0.0000000024


                               [FIGURE 1]

 Figure 1. Primary minimum of HW Vir on HJD 2447684 observed with 
   continuous 20-second integrations. Gaps occur where sky background
   measurements were made.


Formally, a quadratic solution with a decreasing period term is a slightly
better fit to the data, however this is a very weak result, probably not
significant, and further observations over one or more years will be required
to test this suggestion.


F. MARANG and D. KILKENNY
South African Astronomical Observatory
P.O. Box 9, Observatory 7935, S. Africa

References:

Hilditch, R.W. & Kilkenny, D., 1980. Mon. Not. R. astr. Soc., 192, 15. [BIBCODE 1980MNRAS.192P..15H ]
Kilkenny, D., 1986. Observatory, 106, 160. [BIBCODE 1986Obs...106..160K ]
Menzies, J.W & Marang, F., 1986. In :"Instrumentation and Research
   Programmes for Small Telescopes", Proc. IAU Symp. No. 118, eds.: J.B.
   Hearnshaw, P.L. Cottrell, Reidel, Dordrecht, p. 305. [BIBCODE 1986IAUS..118..305M ]
Paczynski, B., 1980. Acta Astr., 30, 113. [BIBCODE 1980AcA....30..113P ]