Re: Zenits and payloads

Bart De Pontieu (BDP@MPE.MPE-GARCHING.MPG.DE)
Fri, 10 Nov 1995 19:47:23 +0100 (CET)

Walter Nissen writes :

>In a message from Bart and Tristan, on October 31, we read: 
>> Subject: Observing Program of the BWGS 
>> 94-074B: had a small acceleration, first Zenit to do so ! 
>Thanks to them for pointing out this interesting fact and also to the 
>unnamed observer whose observations detected this. 

Kurt Jonckheere and I looked at the flash period evolution (as a function
of time) of all Zenit rockets in the PPAS (the flash period database of
the BWGS). I've added three columns to Walter's original table.

>Some of us have watched the monotonic increase of some of these objects 
>for a considerable time.  

The three columns I've added have to do with the monotonic increase of
the flash period of these rockets. It turns out that the flash period of
these Zenit rockets has experienced an exponential growth with time. 
The 'char. time' column gives the characteristic time in days 
(determined from a linear regression analysis of the log(fp) vs. time 
curve), i.e. the time the flash period 'needs' to multiply itself with a 
factor of 2.73. The smaller the 'ct' is, the quicker the exponential
rise was.

The 'error ct' column is the estimated statistical error (in days) on the 
determined characteristic time (ct). A large error usually means the
growth was not a perfect exponential growth, mostly due to 'scatter' at
large periods which are more difficult to determine. 
This 'error ct' column is only a vague indication of how large the
uncertainty on the 'ct' is. Due to the concentration of the BWGS
observers around 40 to 50 northern latitude, the satellite's flash 
period history contains sizeable 'gaps', times during which the satellite
was invisible to the bulk of our observers. These gaps make the uncertainty
on the ct larger, so that it can become as large as 10 days or more.

Finally, the initial flash period is given in seconds. It is determined
by extrapolation of the best fit in the linear regression. It indicates
the flash period right after launch. 


			    vulgar	orbital   char. error   initial
cat #	id #	i    MM      name	plane     time  ct      fl. period
 
16182U 85 97B 70.99 14.15 C* 1697 r     A         156	2	3.6
17590U 87 27B 70.99 14.15 C* 1833 r     B 	  -	-	-
17974U 87 41B 71.00 14.16 C* 1844 r     D  	  155	1	1.18
19120U 88 39B 71.00 14.17 C* 1943 r     C 	  142	1	0.57
19650U 88102B 70.99 14.14 C* 1980 r     A 	  104	1	0.4
20625U 90 46B 70.99 14.13 C* 2082 r     D' 	  194	1	0.45
22220U 92 76B 70.99 14.15 C* 2219 r     E 	  -	-	-
22285U 92 93B 71.01 14.13 C* 2227 r *** B 	  135	1	0.3
22566U 93 16B 71.00 14.13 C* 2237 r     C 	  144	1	0.4
22803U 93 59B 70.98 14.15 C* 2263 r 		  143	1	0.28
23088U 94 23B 70.99 14.13 C* 2278 r     E' 	  144	2	0.33
23343U 94 74B 97.99 14.70 Resurs 1-3 r 		  170	5	1.03
23405U 94 77B 70.98 14.14 C* 2297 r     C' 	  175	7	0.39
23705U 95 58B 71.01 14.14 C* 2322 r     B' 	  -	- 	-
  
>*** The launch size indicated by these dimensions may be suspected 
>    because this object experienced an explosion (see Mike McCants' 
>    analysis of the elsets of the many fragments). 

Looking at the table now, there have been 14 Zenit launches so far 
(there have been more, but these 14 are the only ones still in orbit),
13 of which were in almost identical orbits. The second Zenit (87- 27 B)
rocket never showed any flashing behaviour that we know of. Since
then all Zenit rockets were fast flashers right after launch, except
for 92- 76 B which only showed high periods of about 2 minutes.
We have no data yet about 95-58 B, I think it's visible in the
morning these days. An evening visibility period is coming up early
December.

Of the 10 Zenit rockets in identical orbit of which we have ct-data, 
7 have a characteristic time of about 145 +- 10. Three satellites had
a significantly different characteristic time, namely 88-102 B (104 days),
90- 46 B (194 days) and 94- 77 B (175 days). Assuming that all rockets
have identical construction, and knowing that they are in identical orbits,
one would perhaps intuitively expect similar ct's. However, if we look at
the magnetic damping theory to explain the exponential decay of the rotational
energy (see e.g. http://www.rzg.mpg.de/~bdp/tumble/rottheory.html),
we see that it is possible that the characteristic times are not equal under
the conditions sketched in the above. The magnetic damping depends, among
others things, on the angle between the rotation axis and the Earth's
magnetic field, which is not necessarily a constant. The theoretical
bounds of the 'ct' for Zenit rockets turn out to be 104 and 211 days. All
determined ct's fit into that range. It is however unlikely that a rocket 
will have an extreme orientation over longer periods of time. This
means that the flash period development of 88-102 B can probably
not be explained by the magnetic damping theory *alone*, i.e. some other
forces have probably acted on it.

It may come as a surprise that 92- 93 B and 93- 16 B behave the way they do,
i.e. *so* similar to the other Zenit rockets, even though they are
reported to be 'desintegrated'. From the flash period behaviour and
the flash period evolution with time, I think the 'B' objects are still
very similar to the other (intact) rocket bodies, i.e. only small pieces
have broken off. This seems to be corroborated by the RCS values of both
rockets in the latest SSR. They are not distinctively different from the RCS 
values of the intact Zenits.

The magnet damping theory can also be used to determine certain physical pa-
rameters of the rocket. Using the experimental data, and assuming that the 
rocket's hull is made out of Aluminum, we can deduce that the rocket's hull 
is about 8 mm thick. Using the same method, we have (in the past) calculated 
the Cosmos rocket's hull to be significantly less thick (about 1 mm). 
These values depend critically on the electrical conductivity of the hull,
there is a lot of uncertainty in the determination because of this.
I think it's safe to say that, from the flash period measurements, the 
Zenit rocket is quite different in construction from the Cosmos rocket,
either in thickness or in electrical conductivity. Does anyone have
access to technical information about these parameters? I would be very 
interested to receive them!

Turning over to the third column, I should first say that the determination
of the initial flash period is not straightforward since we usually don't
have observations right after launch, so we have to extrapolate from later
observations. Quite a few rockets suffer from fuel leaks right after
launch, usually leading to atypical changes of the flash period (rapid
acceleration or deceleration), which makes the extrapolation cumbersome.
Nevertheless, it is quite apparent that all recent Zenit rockets had
an initial period of 0.4 +- 0.1 seconds, except for the Resurs launch.
This common initial period range is probably due to a standard method
of decoupling the payload from the rocket. Or, less probably so, due to
a similar way of venting left-over fuel to avoid explosion of the rocket.
The explosion of 92- 93 B and 93- 16 B, and the lack of recorded fuel-leak 
accelerations right after launch, make this seem improbable.

>Leaving aside the vulgar name, which presumably 
>reflects the intended purpose of the payloads, we see that Resurs 1-3 r 
>distinguishes itself in inclination and mean motion, as well as flashing 
>history.  As an observer, I have been impressed with the brightness of 
>Resurs 1-3 r, which may likely be strongly influenced by its lower 
>height and shorter range, which in this table is reflected in the faster 
>MM. 
 
It seems the Resyrs 1-3 r has a slightly different ct (so far), and a
very different initial flash period. The former is probably due to the
different inclination and height (both parameters are pivotal in
magnetic damping). The latter is probably due to having a different
payload and thus a different decoupling mechanism, though it may be
early to say, since only one launch in this category has been observed so 
far.
 
>As a personal note, let me say that I well remember watching C* 2082 r 
>when its period was perhaps 8 or 12 seconds and thinking "Boy, this junk 
>sure puts on quite a show". 

90- 46 B is the current record holder in the PPAS, it has been measured
by BWGS observers over 800 times since its launch ! The Zenit rockets have
been providing considerable consolation for the fact that the Russians
have stopped launching so-called 'J-objects' (octet launchers). 92- 30 J
is the last one in the series. Of course, at the rate these objects are
accelerating, we don't need new ones :-)

Sorry for the length of this reply, hope it was of some interest.

Cheers,
   Bart De Pontieu, bdp@mpe.mpe-garching.mpg.de
   http://www.rzg.mpg.de/~bdp/satintro.html