Satgen314  Millennia Software Checks  by GM4IHJ              1st April 95
BID of this msg is SGEN314  Use this BID if you retransmit msg on packet

As reported in Satgen313 , a sizeable proportion of Amateur Radio
Satellite software will produce errors when used on or after 1st Jan 2000
To check the two main problems expected to arise, you can test software
now , by using two simple test proceedures :-

No1  Check of Year date number.
Run your program on any chosen satellite . Enter day of month, month, and
year number as required by the program to get data starting at 2356 utc on
31st December 1999. Run the program and get a display of several minutes
of Azimuth and elevation data into the next day 1st Jan 2000, as in the
following example 
RS15             utc    AZ   EL   Lat  Long     utc   AZ   EL   Lat   Long
start 2356 onto  0000   84   25   52   327      0002  71   22   56    320
Note... This data is actually run through midnight so what you see here is
tracking for 1st Jan 2000 . The satellite need not be in range , negative
elevation figure are acceptable for this check . Your figures will be
different from these,depending on sat chosen and station location. What
matters is that you get a set of figures based on an input in 1999.
Next run the program again for the same satellite but starting this time
from 0000 utc with an input date of 1st January 2000. If your program is
OK you will get the same figures as above. But if as many programs do it
asks for year input as 99 for year 1999, and so when you get to the second
part of the check you enter 00 for year 2000, you may get a set of wrong
figures for the 1st Jan 2000 input. If you get clear evidence of this
error in your software you can cheat. You can try inputing day 1 month 1
year 100 for Jan 1st 2000 data and you will probably get a correct set of
figures but this is hardly usable in the long term.

No2 Check of Keplerian Element Epoch Year
Here you need to compare two sets of keplerian elements. First enter a
test set of keplerian elements for your chosen sat.The test set can use
any old data you have, but you must alter the Epoch to 99366.0 , which you
will recognise as the existing form of NASA year designation but cheating
to change it to 1st Jan 2000 by adding an extra day over a full year. From
this false test set of Keps you get say the following track data when you
run the program for 1st Jan 2000 at 0000 utc.
UTC   AZ   EL   Lat   Long       UTC   AZ   EL   Lat   Long
0000  195  -15  0     16         0002  193  -11  5     14
Next alter the Keplerian elements to Epoch 00001.00 for 0000 utc 1st Jan
2000 ( where the first 00 represents how NASA MAY !!!! display Keps in
year 2000). Run the program again for 1st Jan 2000 tracking and you should
get the same result as before , thereby proving that your software will
work provided that this is the way NASA displays 21st century Keps.
Again if it does not work you may have to use the same cheat as before and
enter the Epoch year as 100 ie to get 100001.0 for 1st Jan 2000 Epoch.
Note... it is appreciated that the keps used in these tests are not likely
to be anything like the actual ones appearing in year 2000. This does not
invalidate the usefulness of these tests, but we still do not know exactly
how NASA will display Keps in year 2000. Hopefully they will say what they
plan to do soon.  73 de GM4IHJ @ GB7SAN  or, gm4ihj@delphi.com