Re: Different flash period for 95 58B???
Mon, 22 Jan 1996 18:31:20 +0100

>> Ron Lee writes :

>>I just finished observing 95 58 B  (23705) on 16 Jan 96 UT on a 53  =20
>>pass from the NNW to SE from Falcon, CO.  The approximate period while  =20
at 0
>>deg azimuth, 25 degree elevation was 0.49 sec (49.11 sec for 100  =20
>>starting about 109 UT.  Just after passing Capella, azimuth 70 degrees, =20=
>>degrees elevation, the period was about 1.01 sec (50.26 sec for 50  =20
>> Given that I may have miscounted the number of flashes, the period was
>>double when abeam me compared to when due north.  Why the apparent

and    Bart De Pontieu answered  ( forgive me this long quote,
but I think both the possibility for making mistakes, and
the great work are done by Bart, Tristan and the rest of
the group, is well worth reading about ! )

>Presumably, you were counting secondary flashes during your first
>measurement at 25 degree elevation. A cylindrical body like a rocket  =20
>usually give off 2 bright flashes per rotation period, due to reflection =20=
>the sunlight on the body of the cylinder (one every 180 degrees). If the
>rocket has reflective structures at the (usually more flat) *ends* of  =20
>cylinder, secondary flashes may appear, under favourable conditions. As  =20
>illumination geometry changes during the pass, those secondary flashes  =20
>disappear, which is probably happened during the pass you observed.

>I would guess that you miscounted during the second measurement, there  =20
>probably 51 flashes instead of 50, with a flash period of 0.985 sec.  =20
>is, within the bounds of the synodic effect, the double of
>your first measurement. Miscounting on your first measurement would mean =20=

>counting 2 flashes too much. Not impossible, but less probable, I guess.

>The maximum possible synodic effect is 0.002 sec for a 1 sec flash  =20
period, so
>it's unlikely that the synodic effect is to blame for the discrepancy.

>Thanks for reporting your observation, Ron, this gives me the  =20
opportunity to
>plug the activities of the Belgian Working Group Satellites, which is
>collecting and archiving flash period observations, for research  =20
>The archive is available to anyone interested (e.g. at the  =20
seesat-archive or
>the oxford ftp-site), and regular articles with analysis of the archive  =20
>in the bimonthly newsletter Flash. I am also preparing an article in a
>refereed journal. There is some interesting science in there!

>So, if you're getting tired of just looking at satellites and want do  =20
>fun observations that are useful, send your observations to Tristan  =20
><> who is our official 'observations collector'. You  =20
>ask him or me (or any of the other active people from the BWGS that are  =20
>on seesat-l like Kurt Jonckheere, Leo Barhorst, Bram Dorreman,...) about
>flash period observing. Or, check out our Web-pages at :

I would like to take this opportunity to warn about the risk
in just reporting a total time and number of flashes ( or the
computed period ) In the DRA Project, all flash timings ( but
not necessarily the times of all flashes in the observed time )
are reported. a lot more information can be obtained from
such reports. As these mails show, it is possible to miss a flash,
and the geometry of observer-sat-Sun may cause the reflections
from one surface to disappear, more or less, and make one of the
periods longer. There are stopwatches that can store 20-100
lap timings, that are very convenient for this.

On the other hand, I don't think that you necessarily mis-counted.
I timed two early passes of 94-77 B #23405, which is a similar
Zenith-2 flasher, with 46 and 56 timings every 10th flash over
4 minute periods. Graphing the deviations from a constant average
interval showed gradual changes in flash period, which were too
accurate and smooth to be ignored (or caused by miscounts), but
too large to be caused by synodic effect. I have found no
explanation for these results.

 --    +46-8-7428086  (home)  --
 --                +46-8-59095739 (office)--
 --(      59.22371 N, 18.22857 E --