The F-15 ASAT story

By Gregory Karambelas, edited by Sven Grahn

[Gregory Karambelas' text has normal font, Sven Grahn's text is in italics]

Note: Orbital analysis presented at this web page was generated using Two-Line Element sets obtained from the NASA OIG web service and not the newer Space-Track site.

I am an ex Air Force Captain.  My first four-year assignment was on the F-15 ASAT test program!  I was stationed at Vandenberg Air Force Base and had several responsibilities including being an analyst for both the Miniature Vehicle (MV) interceptor made by LTV (Vought) and for the Instrumented Target Vehicle (ITV) made by the AVCO corporation.  I made several trips to the factories of both companies and saw up close and touched these vehicles many times.  I was intimately familiar with their design, construction, and operation. These are my brief recollections

Target vehicle

In short, the ITV balloon, had a grid of break-wires on the surface.  In case the interceptor did hit the ITV, we would be able to tell where on the surface we hit it.  The ITV was sending telemetry in real time once it was inflated on the final orbital pass.  We were guaranteed a few frames of data before the vehicle was destroyed. The ITV had magnetometers on board to use in determining the attitude of the ITV.  This allowed us to know with very good precision the location of the Hit Position Indicator wires and the IR signature in the field of view of the interceptor.

In case the interceptor did not impact the ITV, the ITV was equipped with a Miss Distance Indicator (MDI).  This was a continuous wave radar transceiver.  It sent out a continuous signal, and received the doppler shifted returns from anything that flew by it.  It did not have very long range, but it would give us a very accurate estimate of the scalar miss distance.  I remember having a few good technical papers at the time detailing how we reduced the data.  Those CW MIDIs are fairly standard.  One just looks for the change in the slope of the doppler returned frequency and you know when the closest approach was.  I think a few iterations and analysis of the slope and rate of change of the frequency is needed too, but the process was fairly straightforward.  Xontec incorporated, a well-known company specialized in Test Range tracking and Best Estimate Trajectory analysis did the calculations for us.  But of course we did not shoot at the ITVs so all we had is one run of simulated data that I believe AVCO had generated against a howitzer cannon rail test shot.

The ITV Miss Distance Indicator was not high power and was basically omnidirectional.  It was designed to help out post test analysis of flights where we got close, inside the tracking resolution of the ground based tracking radars, but missed.  The range tracking radars at the time could only reconstruct the trajectory and distances with an accuracy of 20-50 feet, depending on whose Best Estimate of Trajectory (BET) analysis you believed.  So the Miss Distance Indicator was designed only to operate inside that range.  If we missed by more than 100 feet, the range radars would certainly tell us that.  I can't recall the frequencies that RF Miss Distance Indicator used.  That was not important to my analysis so it is not coming to mind.    I assume it was not in the same frequency band as the test range C-band tracking transponder radars to avoid signal interference.

Obviously, if we hit the ITV, we knew the trajectories coincided to within 6 feet, the diameter of the balloon.   I remember seeing the canisters on the lab bench and the kevlar laying around.  I calculated that infrared signature so many times getting ready for the actual intercept.  I should remember that number as I needed the cross section area exposed to the interceptor!

Those wires on the surface would tell us where on the ITV we hit or the Miss Distance Vector.  That was helpful for the guidance analysis of the interceptor.  When one is trying for a hit-to-kill, 3 feet or a 1 foot miss distance is important, and the location on the balloon was supposed to help the terminal guidance people figure out exactly what the interceptor did.

Yes, we used to joke about the ITV's Miss Distance Indicator being used as a proximity fuse, especially when we heard that the test data tape of one howitzer shot from a test range was used to check out the Miss Distance Indicator.  But it was not used as a proximity fuse, just a closest approach determiner.

The ITV was "inflated".  It had a hydrazine catalyst generator that produced hot gases from a rheuthenium catalyst decomposition driven device.  It was designed to heat and maintain the Kevlar balloon at a temperature that would emit the desired infrared signature in a controlled manner for our Interceptor to see.  Post test telemetry analysis was used to calculate the IR signature of the balloon.  There were many temperature sensors embedded in the kevlar in known locations.

Homing vehicle guidance

One must remember that the Miniature Vehicle interceptor had no "guidance" as most people think. It had a ring laser gyroscope for spin rate determination, and for creating its own reference before it left the missile upper stage, but it had no gyroscope in the normal sense.  It did not know its own altitude, or attitude.  All it knew was the position of IR targets in its own field of view.  It tried to home in on that.   It had no idea of the range to target, it's altitude, etc.    It had a C-band transponder for radar tracking. The interceptor was developed from an old US Army program that was tested against tanks!  I remember seeing some photos of it.

The interceptor was called the Miniature Vehicle (MV) It was spun up to approximately 30 revolutions per second just prior to being deployed from the Upper Stage.  It did have an infrared sensor on board.  This was before the days of Charge Coupled Device arrays so it had "strip" detectors on it.  I should be able to remember the composition we used.   It was not your usual Mercury Cadmium Telluride.  I think it was an Indium Bismuth strip. Hughes Research Corporation AKA Santa Barbara Research Corporation was the manufacturer.   It had four strips arranged in a square and four spiral curves.  We could determine the object's position in the focal plane by measuring the time of detections on the strips.  The object detection was by a simple centroid detector as the IR "blob" crossed the strips. (See figure below)

The guidance of the MV was simple Direct Proportional Line of Sight.  The MV had 56 full charge solid propellant rockets arranged around the circumference, and 8 half charge solid rocket motors for a "bang-bang" control system.  The 56 motors were called divert motors, and the 8 other ones were supposed to be used in the end game phase of the intercept where the needed positional changes would supposedly be less.

To control or dampen coning or wobbling of the interceptor, the back end of the interceptor had four pods of attitude control rocket motors.  They were little tiny squibs that would detect the wobbling or off central rotation of the interceptor by logic that would look at the strip detector data.  I can't recall exactly how many charges were in each pod, but the number was not that large.

Sven Grahn's comments: I have been thinking about the intercept geometry of the F-15 ASAT. As I see it the two-stage booster could not have reached more than maybe 4-5 km/s at burnout, which is not enough to catch up with a satellite in orbit moving at 7-8 km/s. Therefore the intercept must have been made "head-to-head", i.e a head-on approach. Also, this would have made it necessary to launch the interceptor more or less into the orbital plane of the target. Of course the F-15 could fly to a point in the orbital plane of the target...within certain limits, of course. Also, the trajectory must have been chosen to make the necessary field-of-view of the IR telescope reasonably small, i.e. the angle between the interceptor and target trajectories must have been as small as possible. In this way the target moved little relative to the intrecptor's flight path vector.. I have tried to sketch this in the attached picture.

Infrared sensor design

The interceptor had a folded Gregorian Telescope for an optical system which as you know is similar to a Cassegrain design.  It had a Honeywell ring laser gyro onboard to maintain an inertial timing reference.  We "initialize" this just prior to deploying the interceptor from the upper stage.  This allowed us to reconstruct the attitude of the interceptor in real space.  But the interceptor really had no idea where it was or how fast it was going or how far away the target was.  It would happily have sat there and tried to maneuver and null out any Line-Of-Sight changes even if it was 10,000 miles from the target and moving away from it!  It did not know.   If the aircraft/missile did not put the interceptor in the right general place where it could maneuver to get in the way, no intercept could possibly occur.  I am sure the total divert capability was classified, and I am not sure I even knew what it was.  The solid rocket divert motors were quite a development problem.  To minimize the IR contaminants and debris, they used some specially designed organic propellant.  Unfortunately, as I recall we had many problems with voids and propellant slumping.  Our IR sensor was so sensitive, I'm not sure if all the trouble of trying to design special propellants was worth it.  I did a short paper for my Master's degree calculating the IR signature of those little burned propellant particles and it is pretty amazing how "bright" they are and how long they stay bright with radiation being the only means of heat transfer in space.

The interceptor detector was cooled by liquid helium prior to release.  Yes, 4 degrees kelvin helium.  We had a large helium dewar on the ground that was about the size of the Robot on the old Lost in Space TV show.  We then had a large helium dewar on the F-15.  We removed the ammunition drum and the back seat in the F-15 with a large tape recorder and the helium dewar.  This allowed us a fair amount of time to do pre-mission checks, perform the flight from Edwards AFB to Vandenberg AFB.  The upper stage of the missile also had a helium dewar that obviously was much smaller than the one on the aircraft. This dewar was connected to the interceptor by a supply and return line.

In space, after the missile was launched, these cryo lines were retracted, and then two spin motors located on the spin bearing assembly of the interceptor were fired and the interceptor was spun up.  It is important to note that the interceptor did not work while not spinning.  This was not a staring array.  I have attached a crude drawing of the strip detector layout.  Those spiral lines were symmetrical logarithmic spirals (spirals of archimedes as I recall).  Some simple geometry and trigonometry would enable us by recording the time of the detector crossing, calculate the objects location in the sensor field of view.

The spin bearing was quite a development problem too.  Once the interceptor was spun up, we did not have all day to deploy it, and spin-down was a problem as the interceptor had to operate within a certain revolutions-per-second range.  Since my BS was in mechanical engineering I would stick my nose into those meetings quite a bit.

Flight operations

I also worked briefly on coordinating the "Pass Plan" with the Air Force Satellite Control Network headquartered at what was then called "Sunnyvale Air Force Station" before it was renamed to Onizuka. The way we planned the engagement shot, how we would contact the ITV as it passed over Hawaii, "inflated" or "erected" the balloon, etc.  was actually a very tight timeline.  We based the ASAT missile and F-15 at Edwards AFB, but we launched the ASAT missile off the coast of California.  We had to have the ITV up and running and all heated up as soon as the interceptor was able to see it.  There were many arguments over who should commit first.  Do you waste a $100 million dollar interceptor, or do you waste a very expensive ITV that had to be launched into space many months in advance?  Who gives the go, no-go?  The ITV could basically only be used once if it was fully heated. It did not have a lot of hydrazine on board for heating purposes.  I think we could shut it down and reuse it once again if it was briefly puffed up and not using much hydrazine for long, but nobody trusted the corrosive effects of the catalyzed hydrazine gas products once they flowed all over the balloon and electronics.

Sven Grahn's comments: The following flight tests were conducted with the system

Nr Date Purpose, result
1 21 Jan 1984 Successful: missile tested without miniature vehicle
2 13 Nov 1984 Failed: directed at a star with miniature vehicle
3 13 Sept 1985 Successful: destroyed NLR satellite P78-1 Solwind (79-17A, Sat Cat Nr 11278)
4 22 Aug 1986 Successful: directed at a star
5 29 Sept 1986 Successful: directed at a star

" ....By September 1985, all was finally ready for a test against an orbiting satellite. On Sept. 13, Maj. Wilbert D. "Doug" Pearson, the director of the F-15 ASAT CTF, took off on a crucial mission that required him to fly an extraordinarily exacting profile in order to arrive at a precise firing location at exactly the right time. Flying at Mach 1.22 some 200 miles west of Vandenberg Air Force Base, he executed a 3.8g pull-up to a climb angle of 65 degrees. The missile automatically launched itself at 38,100 ft. Minutes later, orbiting peacefully 345 miles above the Pacific Ocean, an obsolete satellite named P78-1 was suddenly shattered into pieces. Pearson had become the world's first pilot ever to shoot down a satellite. To this day, now Maj. Gen. Doug Pearson remains, as Air Force Materiel Command Commander Gen. Lester Lyles recently observed, the first and only "space ace ..." (1)

President Reagan gave the go-ahead for the test against a real target in space on 20 August 1985 (4). The test was originally scheduled for 4 September, but because the 15 days notice had not been given to Congress it was delayed 9 days (3). The target was Solwind P78-1, a gamma ray spectroscopy satellite weighing 850 kg (2) that had been launched in February 1979 into an initial orbit at 563-602 km at 97.6 degrees inclination. P78-1 was in a noon-midnight, Sun-synchronous orbit. The identity of the target was actually leaked before the test (3).

The most probable time of intercept is at around 2040 UT on 13 Sept. 1985 when the Solwind passed off the US West Coast from south to north. This time of intercept is also given in (2). The local time in the Pacific time zone was then 1240, i.e. the target satellite was illuminated making the surface of the spacecraft warm and radiating infrared radiation. The altitude of the target satellite at the probable time of intercept was 530 km.

Target vehicles finally launched

Sven Grahn's comments: The two target vehicles (85-114 A Sat Cat Nr 16328, 85-114 B Sat Cat Nr 16329) were actually launched on 13 December 1985 at about 0230 UT from Wallops Island into orbits at 315-772 km and 319-768 km and 37.05 deg inclination. The mass of each vehicle was 81.6 kg (2). A Scout rocket was used and it carried both ITV's on top of the final stage. It is interesting to note the quite elliptical orbit. Was this done on purpose to be able simulate "adversary" targets at altitudes ranging from those of photo-reconnaissance satellites to those of other Soviet space assets such as navigation satellites? However, it also seems that the Scout launch vehicle underperformed, so the elliptical orbit may be unintentional.

It is interesting to plot the trajectory of an ITV when passing straight over the ground station at Hawaii, where Greg Karambelas says the inflation command would be issued. It turns out that the inclination of the orbit seems to be deliberately chosen so that the target satellite would head straight for the southern California coast, indeed directly in the direction of Vandenberg and Edwards AFB where the F-15 was based (see figure below).

One can also see that the timeline was tight. After AOS at Hawaii it would take a few minutes to command and verify that the balloon had deployed properly, say five minutes. Then there would only be ten minutes before the target vehicle crossed the California coast. The F-15 would hardly be able to scramble and reach launch altitude in such a short time.

What happened to the target vehicles?

According to Gregory Karambelas the target vehicles were battery-powered and were kept dormant and uninflated in orbit following the ban on further testing of the ASAT system. Finally, permission was granted to test the inflation system. Evidence of this can be gathered from an analysis of the orbital period of the taget vehicles. This analysis shows that the ITV-2 (Sat Cat Nr 16329) target was actually inflated late on 17 December 1986 or early on 18 December, see figure on the right. The decay rate of the spacecraft suddenly increases as the cross section suddenly increases.

No corresponding sudden increase in decay rate can be observed for the ITV-1 (Sat Cat Nr 16328) vehicle. The ITV-2 target decayed after 604 days (on 9 Aug 1987) and the ITV-1 target decayed after 1245 days (on 11 May 1989) (2).

References and Notes

  1. Raymond L. Puffer, "The Death of a Satellite"
  2. D.G. King-Hele et al, The RAE Table of Earth Satellites 1957-1989, The Royal Aerospace Establishment, 1990
  3. G.E. Perry, day-to-day log, entry for 5 Sep 1985.
  4. G.E. Perry, day-to-day log, entry for 21 Aug 1985.

Back to Space History Notes