Timings are rarely inconsistent - a "mature" flasher (no recent collision or leak) has a (slowly increasing) period around one axis. But the required PPAS way of giving total time divided by an integer count often fails - if the object has more than one reflecting surface, you can find two or more series separated by a "phase shift" step, whose length depends on the angle between the surfaces. Sometimes the series are intermixed, with the first gradually fading as the next grows in brightness. One example is Superbird A, where the phase shift is ½ period. My graphical method of plotting ALL flash timings relative to an assumed fixed period usually reveals these patterns, and the true period. A minor problem is the synodic effect (which suggested the name of my graphing program) - the apparent motion of the satellite can increase or decrease the flash period. The amount depends on the varying angular speed and obs-sat-Sun angle. For a cylinder, the effect is due to the reflecting area gradually moving around the surface, but it can be quite abrupt if the rotation axis passes close to the observer. 65-34A was moving 0.12 d/s, but the flash beam 136 d/s (?), so the synodic effect is probably < 0.002 s on the period. /Björn ----- Original Message ----- .... 65-34A shown in "hot zone" of when it consistently flashed; I discarded earlier, less consistent timings. ... ------------------------------------------------------------------------- Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive: http://www.satobs.org/seesat/seesatindex.html
This archive was generated by hypermail 2b29 : Tue Jun 12 2007 - 03:00:06 EDT