Stealth Satellites -- More on AFP-731 & K-2

John Pike (
Thu, 27 Jun 1996 08:24:44 -0400

I am still trying to puzzle through what might have happened to our old 
friend AFP-731 [the Feb 90 STS "flasher"], in anticipation of the 
possibility that the upcoming Titan-4 K-2 launch might be a replay of 
this most puzzling mission.

As you may recall, initial March 1990 events, what with the Soviets 
publicly claiming that the thing had blown up, certainly confirmed 
either that the AFP-731 survivability elements that were tested at that 
point in time either left the Soviets actively bewildered, or at least 
convinced that it was in their interest to convince the US that they 
were actively bewildered [in this wilderness of mirrors it is always 
difficult to know how many layers of perception are being managed]. I am 
guessing in any event that the stuff used in the March display was some 
of the residual hardware from the *losing* contractor in the program, 
and certainly not the whole ball of wax, as little would be gained by 
staging the same show twice, for the repetition of history is folly.

OTOH, the October disappearing act seems to have been much more 
convincing [or at least our Russian colleagues have decided that it is 
in their interest to attempt to persuade us that they found it 
convincing, as their FBIS traffic generally ignores the existence of 
this bird], and upon reflection it is not hard to comprehend why this 
would have been the case:

It would not be neccessary to make AFP-731 entirely invisible, merely to 
reduce its radar and optical signature to the point that it would blend 
in with the background debris population -- if one could achieve a one 
to two order of magnitude RCS and visible magnitude reduction [one might 
as an exercise attempt to calculate just whate reduction would be 
required and what would be needed to do this], you would transform 
something the size of a city bus [which was pretty clearly KH] into 
something the size of a refrigerator, which would be pretty hard to 
distinguish from a couple of thousand other objects, if one even took 
the time to try to do so.

I am still having a bit of difficulty precisely comprehending just how 
one would go about developing and deploying a stealth applique package 
for a Keyhole. There is not one whole heck of a lotta rattle space 
inside either the STS or T-4 payload compartment, so the thing would 
probably have to be deployable, which probably means inflatable. Naively 
one could just shrink wrap the whole spacecraft, though I am kinda 
imagining this more along the lines of an inflatable life raft, or 
something, but I think that we are probably going to have to sit down 
and actually try to draw the thing in order to comprehend a not 
implausible way of developing such a configuration.

The real trick seems to be twofold:

1 - Really low signature for the initial get-away maneuver [probably 
have to turn the optics aperture away from the Earth, and have all the 
dorsal radiators covered or stowed];

2 - An operational configuration that had an acceptably low nadir RCS -- 
ie, radar return from the optics aperture and not much else, along with 
some difficult but not impossible efforts to *redirect* the IR signature 
away from the belly of the spacecraft to radiate it off the spacecraft's 

I am kinda gettin lost here, as my current model has the back of the SC 
being the stealthy part during the evasive maneuver, and the belly of 
the SC being the stealthy side during normal ops, but IMHO I am having 
an easier time making the back stealthy against radar but not IR and the
belly stealthy against IR but not radar, and this fits with totally 
evading radar during the disappearing act and evading IR and fooling 
though not entirely hiding from radar during normal MEO ops.

In any event, I am increasingly suspecting that this whole sat.track 
game is a lot less perfected than one might have imagined, in that I 
think that the elset crowd has less-than-perfect knowledge of what is 
what -- NORAD's two-day bewilderment as to the fate of Landsat-7 was 
either an active attempt on their part to convince the Sovs of our 
incompetence, or a rather shocking demonstration of incompetence [I tend 
toward the later, though cannot exclude the former].

Based on the dismal Landsat-7 performance, I could easily imagine that a 
spacetrack system with limited capabilities and limited resources would 
have a limited appetite for precisely characterizing every last object 
that ever generated a track file in their database, and that it would 
not be *ovewhelmingly* difficult to hide a KH in the debris population 
if suitable steps were taken to reduce and modulate the signature to 
approximate that of debris.

It would be an interesting exercise, currently beyond the scope of my 
resources [though I would be interested to hear if anyone has any 
thoughts on this matter], to contemplate whether there exists a debris 
population in the 1000x5000x63* neighborhood in which one could hide 
such an object -- that is, is there a preexisting category of rocket 
body debris from molniya orbit injections into which one might insert an 
additional object. Indeed, was there another launch coincident with the 
~~October 1990 disappearence of AFP-731 that this spacecraft could have 
used to mask its own maneuver??? Is there any possibility that some of 
the NOSS-2 debris might in fact be our ole friend AFP-731???

John Pike
Federation of American Scientists
 CyberStrategy Project   
 Intelligence Reform Project
 Military Analysis Network
 Space Policy Project    

Indeed I tremble for my country when I reflect that God is just.
  - Jefferson