Satgen 393 Keps 2000 AD Revisited by GM4IHJ (Bid SGEN393) 5 Oct 96 Having discussed the subject of how computer software for satellite tracking may have to deal with the moment when 2359 31 Dec 99 becomes 0000 1 Jan 2000 ( satgens 313 and 314 ), It seemed appropriate as the task of IHJ software conversion to VB4 Windows 95 was nearing completion , to check that it was millennium proof. Several surprises were encountered. Firstly Microsoft provide a function called Date which when called obliges by giving date in the form 30/09/96, and another function Now which obliges with 30/09/96 15:40:34 ie date and time . BUT BUT when the computer calendar is shifted to year 2000 a change of format occurs whereby Date produces 30/09/2000 and Now produces 30/09/2000 15:40:34. So anyone who starts a software program with this short cut will appreciate that amendment is necessary if 2000 is to dawn without problems from this Microsoft teaser. Secondly, as mentioned previously there is also a problem with the way NASA/NORAD present Keplerian elements of satellite orbits. Eg typical Epoch date time is quoted as 96253.125000 ie year 96 , day of year 253 , plus 1/8 day = 3 hours. At this time this writer does not know how NASA/NORAD will greet the millennium. Neither a change to Eg 00253.125 or a change to 2000253.125 will be accomodated by most amateur radio software, whether the input is manual or autoload. In fact a rather tortuous bunch of IF and ELSE statements will be necessary to accomodate any change, So hopefully NASA/NORAD will signal their intentions long before the fatal day. Meanwhile simply putting Mirs kep elements into the revised VB4 software produced a long term NASA chestnut whereby - Keps for Mir with a Mean Motion of 15.60238539 conflict with Keps for an Epoch less than 3 hours earlier for all the other vehicles which are firmly attached to Mir ( Kvant 1, Kvan2. Piroda, Kristall, Spektr and Progress M-32), all of which are given the Mean Motion of 15.62027093. Given simple arithmetic Mir would seem to be stretching its docking latches with the other vehicles by about 5 kms per day. That said, the difference probably arises because NORAD run a separate track algorithm and Data set for the more important Mir. But it does point up the message that one or two radio amateurs who quote acquisition times in seconds, should take care. Indeed NASA do say "Do not use these elements for calculations of high accuracy". To be fair to NASA 5kms per day is less than 1 second in elapsed time, but the super accurate amateurs cannot be excused. They should remember that the signal path from the satellite on their horizon is anything but straight ( watch the Sun distort at Sunset), and none of them have really accurate data regarding the actual elevation of their station horizon on each bearing . However, when all is said and done the super accurate specialists are slightly better than the poor G who meekly enquired " I have got Mir Keps, But can anyone give my the Shuttles Keps" . They were of course hard docked together at the time Hi.