This project needs a small telescope. The chart is two degrees in height. The starting point is lambda Sagittarii, the star representing the point of the bow - or the arrowhead - between the two quadrilaterals of the constellation. This is about magnitude 2.8. XZ is about 1.5 degrees west of lambda, about 0.5 degrees SSE of a red star of mag 6.2. There is another orange star of 6.6 a little farther west.
The star itself is normally at magnitude 8.82 and visible in good binoculars. But during eclipse, it will fade to 10.93, or five times fainter. Since the eclipses last for about 105 minutes, this represents a change of a magnitude every 25 minutes or so. Suitable comparison stars are marked, all lying to the south of the star.
Eclipses can tell us a good deal about a star. First we time the moment the eclipse begins, or first contact. Then we time the moment that the dimming stops, or second contact. Third contact is when it begins to brighten again and fourth contact occurs when everything is back to normal. By measuring a spectrum of the star we can see how fast each of the components is travelling around one another. This then allows us to determine the diameter of each component. The time from first to second contact, or from third to fourth, multiplied by the speed gives us the diameter of the smaller component. First to third, or second to fourth, similary treated, provides the diameter of the larger star.
Apart from the experience you may wish to time eclipses. This is done by measuring the brightness at fixed intervals, say 5 or 8 minutes in this case. These are then plotted using magnitudes against time and lines dran through the observed points. Where the lines of ingress and egress intersect is mid-eclipse.
There's an interesting problem here. In its orbit about the Sun the Earth is sometimes nearer the star, sometimes farther. So we must translate all periodic phenomena of this type to the centre of the Solar System by applying a heliocentric correction which, since Sagittarius is close to the Ecliptic, may be as much as 8 minutes in either direction.
We can go past this, however. We add colours to the light curve and we can determine the temperatures. A star like this will usually comprise a hot component, the 'blue' star , and a cool component, the 'red' star. (The colours may not be quite like this but it's a convention). During the total eclipse of the hotter star we're seeing only the light from the larger and cooler star and can determine a temperature from this. At other times, we're seeing the light from both but it's simple enough to use another relationship: the brightness is proportional to the second power of the radius or the fourth power of the temperature (Brightness = R2*T4).
In 2008 eclipses will be seen on 28 August, 8 46 UT, 10 September, 11 13 UT, 20 September, 7 43 UT, 3 October, 13 40 UT. These can be calculated using the ephemeris JD 2441890.6201 + 3.275555E. Mid-eclipse is probably about 5 minutes earlier than the quoted times due to the Earth being nearer the star than the Sun.
The dates and times given above by Stan were determined from the GCVS IV catalogue. Data from the Mt. Suhora Astronomical Observatory Cracow Pedagogical University gives different times of primary minima. eg 29 August 2008, 11:10 UT, 11 September 2008, 13:38 UT, 21 September 2008, 9:29 UT, 04 October 2008, 10:05 UT
About the difference in times Stan comments:
I got these values from GCVS IV which is the last hard copy edition. I'm surprised that the period value is so far out but maybe there has been a change. Or were the original measures poor? GCVS quotes a number of period changes but these are from early measures. With a designator like XZ there will be a lot of early photographic and visual epochs. In any case use the modern values which I see change the dates around a lot. The comparisons I used were all from Guide 8 and all to the south of XZ. From the left they are 112, 92, 105, 112, 90 (directly south), 86 & 106.
We would be interested in any observations of XZ Sgr during this period please send result to Brian Loader. Brian hopes to observe this star during this time so it would be interesting to combine results fo observations to see if we can determine which set of data given above provides the more accurate predictions.