Click to visit Home page

Google Adsense Privacy Policy

The Trindade Island UFO Photos (1958)

The Rouen UFO Photo (1954)

The Hawaii UFO Photo (1989)

Identified Flying Objects

The Nellis UFO Video of 1994:


Part 1 (Location)

Part 2 (Appearance)

Part 3 (Performance)

S-30 Transcript

Mobile Site






by Martin J. Powell

In the first part of this study I examined the nature of the instrument used to film the Nellis UFO and identified the probable location of the filming site 'S-30'. The footage was most likely filmed from a threat-radar trailer located just outside the Tonopah Test Range, deep within the Nellis Range Complex.1 In the second article I constructed a model of the UFO and studied its behavioural characteristics. In this final part of the study, I will concentrate on the UFO's flight dynamics and consider its possible function and origin.

In addition to its appearance and behaviour, some remarkable claims have been made about the object's performance capabilities, by both actual witnesses and later researchers. Among these are that the object "travelled 13 miles in six seconds", that "it moved very, very slow and then it moved very, very fast" and that "it did right angles [and] straight, horizontal climbs".2 This article sets out to determine whether such claims can be verified by the radar data which appears on the camera displays in both the S-30 and S-13 footages.

"That's why it's so slow…"

In the early part of the S-30 footage the object appears as a dark blob in the distance, moving slowly in front of Black Mountain, at the Western side of the range (see Part 1). Between 23:20:45 and 23:20:53 (UTC) it descends slightly, then it levels out and starts climbing (Figure 1). Its velocity appears fairly constant throughout this period, although this is difficult to determine because the object's relative approach angle to the camera is uncertain. The range display over the early phase shows the distance increasing, however the object's apparent size increases slightly, suggesting it is getting closer to the station. It is most likely performing a wide right-hand turn, adjusting its flight path in order to head towards the tracking station.

Montage showing the Nellis UFO flying in front of the distant mountain range at the start of the S-30 footage


Figure 1  A montage of the Nellis UFO's movement in front of the distant mountain range during the early part of the S-30 footage. The UFO's position is shown at 2 second intervals.

Sideways Motion

Two sections of the early footage give an indication of the object's manoeuvrability. Judging by the position of the dark 'frontal unit' and the discernible 'tail unit' (see Part 2) the object appears to move sideways on two occasions; from 23:20:56 to 23:21:18 and from 23:21:21 to 23:21:29. The second period is of greater importance because of the object's better clarity, and a montage of its movement is shown in Figure 2. The object initially travels front forward, but it rotates to starboard (right) of track at 23:21:20. It then rises (at least, relative to the line of sight) before moving sideways (to port) for five seconds after 23:21:23. This motion can clearly be seen against the dark layers of cloud. It then sways back to port and regains forward-facing flight. Evidently the object is capable of rotating (yawing) during forward flight, and of moving sideways; techniques it seems to use in order to adjust its flight path.

Montage showing the Nellis UFO's apparent sideways motion (click for larger image)


Figure 2  Montage showing the UFO's second period of sideways motion (click for a larger image)

 Range Data and Radar Return

At 23:21:35 the range display freezes at 9135 metres.3 Previous researchers have concluded from this that the UFO kept a fixed distance from the camera from this point onwards, whilst continuing to move forward at high speed. If this were correct, the object's ground track during this period would be perfectly circular, and its flight profile would be very peculiar. Figure 3 (upper) shows the cumulative flight profile of the object based on the range data shown in the footage. The first section of the profile covers a very large distance and it shows a gentle climb followed by a gradual descent. According to the range data, the object would have turned abruptly on many occasions during this part of the plot, covering very large distances in short periods of time. When the range finally settles on 9135m, i.e. when the object commences its wide arc, it is seen to climb sharply, gradually levelling off as it passes the station.

Two flight profiles of the UFO according to the range data displayed in the S-30 footage


Figure 3  (Upper) The flight profile of the UFO according to the range data displayed in the S-30 footage, and (Lower) the flight profile of the UFO according to the author's straight-line ground track hypothesis. Note the vastly differing distances for each profile.

Having studied the footage closely, I believe the entire upper profile in Figure 3 is incorrect, for the following reasons. Firstly, the fixed range figure does not seem to reflect the increasing apparent size of the object, which is clearly getting closer to the station. The camera zooms in on the object at one point (23:21:51), but this alone does not explain the object’s apparent increase in size towards the end of the clip. Secondly, when the range display is studied alongside the behaviour of the radar return, the two appear to be closely related. The radar return is highly erratic early in the S-30 footage, but it 'flat lines' suddenly at 23:21:21. The range then increases rapidly from 11705m to 32305m in only a matter of seconds (the claim that the object covered '13 miles in six seconds' originates from the erratic data during this phase). The range then drops suddenly to 9135m and it remains at this value until the end of the footage, when the message ‘TRANSMITTERS NOT RADIATING’ appears at the bottom of the screen. This implies that the radar (i.e. the source of the range data) has stopped transmitting.

The radar appears to stop providing reliable range data after the radar return flattens, in which case the object probably approaches the camera at a much closer distance than 9135 metres. If the object's approach track could be estimated beyond the point where the range becomes fixed, it would give a better indication of its closest distance to the station.

Object Ground Track

Any attempt to estimate the object's ground track must unite all factors of object azimuth, elevation, apparent size and distance. The easiest approach is to assume that the ground track is a straight-line. Whilst this might seem simplistic, any variations in the ground track away from a straight-line are likely to be small, and they are not obvious in the footage. In the absence of accurate range data, it is essentially impossible to determine a precise ground track.

As a starting point, I plotted the object's ground track based on the range data between 23:21:05 (range 12795m) and 23:21:19 (range 11705m). During this time the object's heading varies between North-west and North-east. Beyond this point the range data becomes unreliable - even registering 32305m on one occasion - before settling permanently on 9135m. Examination of the ground track shows that in all probability, the most reliable stretch of range data occurs between 23:21:08 and 23:21:13, when it reduces steadily, within a 5 metre accuracy, from 12785m to 12705m (Figure 4). The object heads Northward during this period, and this seems to be verified by one of the range controllers on the soundtrack who mentions that the object is "heading North at this time".

The hypothetical model (straight-line model) assumes that the displayed range data from 12785m to 12705m are accurate and true values. The starting azimuth of the object for the hypothetical track is taken to be the value coinciding with a range of 12785m, i.e. 221º.2, and the starting elevation is 1º.6.

Ground track of the Nellis UFO in the early part of the S-30 footage (click for larger image)


Figure 4  Plot of the UFO's ground track according to a section of the range data shown in the early part of the S-30 footage (click for a larger image). This is probably the most reliable section of the entire range display.  

The yellow curve in Figure 5 plots the object's apparent size (height) as measured on a 14-inch TV screen throughout the duration of the S-30 footage. Any deviations in the ground track are no doubt hidden in the zigzag nature of this plot. The zigzags are primarily caused by the object's constantly changing viewing aspect and elevation, and also measurement errors because of the film's poor resolution, which puts a limit on the accuracy of measurements taken from the television screen.

The general shape of the yellow curve is consistent with an object approaching from a distance to a point close to the camera. If the object were to maintain a fixed distance from the camera beyond any particular point, it would appear the same size thereafter, so the curve would abruptly level off beyond that point in time. In fact, the curve does not level off after the point when the range becomes fixed (indicated on Figure 5), and it only does so much later - as expected - when the object makes its close fly-past of the station.

I wrote a BASIC computer program which plotted the apparent size of a 10 metre high object according to the hypothetical straight-line track (the height of the object used in the plot does not matter because the shape of the curve would be the same regardless of the chosen object size). The angle of the straight-line track relative to the station was increased in small increments (from zero degrees) until a value was reached where the hypothetical curve (red) best matched that of the actual size curve (yellow). The best match was found to be at an approach angle of 18º.1, measured relative to the station. The hypothetical curve matches the actual curve very well, even when assuming a straight-line ground track.

Plots of the UFO's size (height) on a 14-inch TV screen over the duration of the S-30 footage


Figure 5   Plot of the UFO's height on a TV screen over the duration of the S-30 footage (click for a larger image). The yellow line is a plot of the UFO's size (height) on a 14" TV screen over the duration of the S-30 footage. The red line represents the equivalent size the object would appear according to the author's straight-line ground track hypothesis.

Small adjustments to the hypothetical curve were required at two points in time. These were at 23:21:43, when camera contrast adjustment causes the object to suddenly appear larger as it passes in front of a dark cloud layer, and at 23:21:51, when the camera zooms in on the object. These adjustments are indicated on the red curve.

The hypothetical model suggests that at closest approach, the object's ground distance from the tracking station is 3981m. This equates to a line-of-sight distance of 5130m, or just over half the distance indicated by the camera's range readout. The object would be heading approximately 23º (i.e. to the NNE). From the viewpoint of the tracking station, it would make its closest approach at an azimuth of around 293º, i.e. at 23:23:12. If it were to continue on a straight-line track, the object would eventually disappear from view over the NNE horizon.

Velocity and Rate of Climb

A profile of the object's flight path for the period appropriate to the hypothetical model is shown in the lower part of Figure 3. It shows the cumulative distance flown by the object against its altitude above ground level (agl). When compared to the profile plotted on the basis of the range data shown in the footage (Figure 3 upper) the results are significantly different. In my view, the hypothetical plot is probably a more accurate representation of the object's flight profile.

Over the course of the hypothetical plot (2 minutes and 8 seconds) the object climbs from an altitude of 1226ft agl to a maximum altitude of 10,617ft agl, giving an average rate of climb of 4,402 feet per minute. The object climbs for much of the time, appearing to reach a 'target' altitude of around 10,000 ft agl before levelling off.

The prominent 'bump' in the profile at 23:22:41 may be a genuine movement on the object's part (this is when it leaps out of shot as it turns to face the camera) or it could be a result of the camera attempting to keep the object in view. Irregular 'bumps' in the profile beyond this point are primarily artefacts of the method used to determine the azimuth and elevation from the television screen. A 'line of best fit' across the profile after 23:22:43 shows a slight climbing trend.

The flight profile was divided into sections so that the object's rate of climb and velocity could be determined in each case. The rate of climb varies considerably throughout the profile. The highest rate of climb occurs in the four seconds after 23:21:49, when the object attains a rate of 14,475 fpm. This is when the female radar controller on the soundtrack is heard to say "[It's] goin', like…straight up!" The ground speed and air speed are found to be reasonably consistent throughout. The average ground speed across the entire profile is 189 knots, and the average airspeed is 200 knots. During the accurate stretch of range data (i.e. between 23:21:08 and 23:21:13) the object's speed is around 43 knots, explaining why the operators think it is "so slow" at that time. It accelerates gradually as it approaches the tracking station, but particularly so after it levels out after the climb at around 23:22:30. Its 'cruising' air speed probably lies between 173 and 207 knots. There is no clear evidence that the UFO achieves an exceptionally high speed at any point in the S-30 footage.

Object Size

Any attempts to estimate the object's size are complicated because of the uncertainty of its distance throughout most of the S-30 footage. However, for the purposes of this exercise it is reasonable to assume that the range data is accurate when the object is between 12795m and 11705m from the tracking station. Since this occurs before the camera zooms in, and the width of one degree in azimuth is known, the object's height can be estimated by triangulation, and is found to be 8.7m ± 2.5. The object does not present a clear side or front view during the early period of the footage, so these dimensions were estimated from carefully selected still shots, using the height as a starting point. Its length was found to be 11.6m ± 3.5 and its width was 12.3m ± 3.7.

Object Viewing Angle

Figure 6 is a scale plan of the object's ground track, according to the hypothetical model, as it passes the tracking station. For the sake of clarity the plan commences at 23:21:43, i.e. 35 seconds after the start of the hypothetical model. The plan also shows the direction in which the black frontal unit faces as the object moves forward (green arrows), based on a comparison of footage stills with the author's UFO model. This also gives some indication of how the object might deviate from the straight-line ground track hypothesis, if it were assumed that the object was moving continuously front forward.

Simplified model showing the Nellis UFO's ground track as it passes the S-30 tracking station (click for larger image)


Figure 6  The object's ground track, drawn to scale, according to the author's straight-line model as it approaches tracking station S-30 (click for larger image). The green arrows along the track indicate the direction in which the UFO's black 'frontal unit' faces at the specified times. Points where the UFO turns or rotates are also indicated.

For much of the initial period the frontal unit faces towards the tracking station, after which it faces slightly to port (left) for a time. After the object turns at 23:22:41, the unit (now 'shielded') faces more or less continuously towards the tracking station. During its "surveillance period", the object continues along its flight path, moving sideways for much of the time, and even backwards for eight seconds after 23:22:47. It is interesting to note that much of the object's attention appears to be focussed on the region to the South of S-30, as well as on S-30 itself. After closest approach to the tracking station, the object turns to port and it is still turning as the footage ends.

S-13 Footage

I only have possession of a ten second clip of the S-13 footage.4 It would obviously be unwise to draw too many conclusions on the basis of such a brief clip, however it is worth commenting about the behaviour of the radar return and the object's flight path during this period.

Throughout the S-13 footage the object appears as a small white dot near the centre of the screen, too distant to see any detail.5 The radar return shows a small but significant peak in the centre of an otherwise flat line, and the letter 'A' appears next to the azimuth, elevation and range displays. As suggested in Part 1, this probably indicates that the radar is set to automatically track the object (since it now has a slow velocity and, presumably, a stable flight path).

Assuming it was filmed on the same day, the time difference between the end of the S-30 footage and the commencement of the S-13 clip in my possession is a little over 45 minutes. If the object had retained the velocity it had when it passed station S-30 (around 200 knots), it would have covered a distance of some 174 miles (280 km) in this time span. Evidently the object must have reduced speed - or even hovered - for some considerable time after passing S-30. Certainly, by the time it is filmed from site S-13, it has slowed down considerably. In the soundtrack, one of the controllers is heard to say, "It appears to be going outbound real slow. There's hardly any range velocity." In fact, the object is moving at around 73 knots, which is certainly slow in comparison with aircraft speeds more familiar to the range controllers.

In my clip, the azimuth of the object increases from 21º to 23º (i.e. to the NNE) and the elevation remains fixed at around (or just above) 18º. The range increases steadily, in 5 or 10 metre increments, from 10935m to 11020m. This equates to the object having an altitude of around 11,155ft agl, or about 540ft higher than the maximum height reached at station S-30.

Using the time, azimuth and range values, I drew up a plan of the object's ground track throughout the ten-second period. The ground track was fairly close to a straight-line, on a heading of approximately 100º, or just south of east. It would therefore appear to have turned right through some 77º since passing station S-30. Towards the end of the S-13 footage, the object apparently accelerates and leaves the area.6

"Where's it come from..?"

Although some claims about the object's performance capabilities appear to have been based on unreliable range data, the Nellis object is certainly capable of performing some bizarre manoeuvres, namely moving sideways, backwards and rotating during flight.

Of course, helicopters and other VTOL craft are able to achieve such manoeuvres, but it is the method used by the Nellis object to perform them which is difficult to determine, despite the visual evidence. It seems to use a revolutionary form of in-flight manoeuvrability. It can also attain high rates of climb over brief periods of time, although its 'cruise' speed is a reasonable 200 kts. The propulsion system is not obvious in the footage. But taking into consideration all factors of location, appearance, behaviour and flight characteristics, I would say there is little doubt that the Nellis UFO is a man-made machine.

So where did the Nellis UFO originate, and what is its purpose?

Occurring deep within the Test Range, perhaps it is a top-secret experimental project which has been flown over the radar operators' heads either as a test of radar detection or of operational procedure. Although this suggestion may sound implausible, the effectiveness of Stealth technology was once tested in a similar manner.7 If it is a classified project, however, it is difficult to account for reported sightings of similar objects over Central and South America.8

The object's behavioural characteristics leave no doubt that it is intelligently controlled. Perhaps it is a new generation of Unmanned Aerial Vehicle (UAV), whose purpose is primarily reconnaissance, battlefield surveillance or target acquisition. It may be operated remotely (by a pilot in a control room many miles distant) or it may be pre-programmed to perform a specific activity over a specific area. Like most UAVs, it would navigate using GPS, INS or similar navigation systems. Equipped with infra-red and electro-optical sensors, it would record intelligence information using photography or real-time TV pictures. The UFO's large frontal unit may be packed with such equipment.

The Global Hawk UAV

The DarkStar UAV

The Bombardier CL-327 Guardian VTOL UAV

Global Hawk is a high-altitude, long endurance unmanned aerial vehicle (UAV). It provides high resolution, near-real-time imagery of large geographic areas (U.S. Air Force Photo)

DarkStar is a high-altitude, low-observable endurance UAV optimized for reconnaissance in highly defended areas. It contains high-resolution synthetic aperture radar (SAR) and electro-optical (EO) payloads (U.S. Air Force Photo)

The Bombardier CL-327 Guardian is an advanced VTOL UAV developed from the CL-227 Sentinel, and is powered by two contra-rotating rotor blades. It can operate from both ships and rugged terrain and can function autonomously (Photo from

The manner in which the Nellis UFO is seen to manoeuvre with 'confidence' suggests that it may have the ability to operate autonomously, i.e. with a built-in level of 'intelligence'. It may even have wandered out of bounds by its 'own' mistake.  John Velez likened the Nellis UFO's behaviour to that of a fly, and this is certainly a good comparison when one examines its movement and the way it appears to respond to its environment.9 The physical and behavioural similarity to an insect has much in common with the new generation of 'micro-air robots', some of which are based on the principles of insect flight.10

American military UAV projects developed in recent years include Predator, Dark Star and Global Hawk and they are now well known, although they were developed in great secrecy.11 Whilst they are certainly weird looking, they still retain the basic structural components of a typical aircraft - wings, fuselage, etc. Other UAVs, such as the Sikorsky Cypher and the Canadair Sentinel, are unconventional in appearance but are made of conventional materials. The structure of the Nellis object, on the other hand, is far removed from any known aircraft flying today. Its component parts appear flexible and they can activate very rapidly. Put in conventional terms, this 'aerial robot' appears to combine lighter-than-air technology with a unique form of vectored thrust.12

Recent articles have claimed that research is currently underway (primarily at Groom Lake) into many new and bizarre sounding technologies. Two such examples are electrochromatic panelling (rendering the object invisible to ground observers by special use of camera projections) and holographic projection (causing distortion of its apparent size to a ground observer).13 The strange, almost luminous nature of the Nellis object suggests that it may involve these or any other combination of new technologies.

According to an official statement by the Pentagon in 1996, "…Some specific activities and operations conducted on the Nellis Range both past and present remain classified and cannot be discussed." If the history of black projects is anything to go by, the truth about the Nellis UFO may not be known for some 20 years or more.


1.  The complete S-30 clip is included in the video compilation 'UFOs: The Footage Archives (Part 3)' by Michael Hesemann and Nadia Zahradnikova, 2000 Film Productions, Düsseldorf (1998). The date assigned to the clip (30.5.95) is, however, incorrect.  See Part 1 for a list of websites which include clips from the Nellis UFO video.

2.  'Unidentified over the Nellis Test Range' at 'The Anomalies Network' website.

3.  I have a near-complete copy of the S-30 footage, lasting 2 minutes and 38 seconds. Unfortunately there is no indication in this footage as to the unit of measurement used for the range data display. Whilst most researchers have assumed the range to be in metres (as has the author) there are good grounds to consider the possibility that the units may be in feet. If this should be the case, then all sizes, speeds and rates of climb given in the text will be a third of the stated value or, more precisely, the value in metres divided by 3.281. [Note added April 2007: The author has now acquired complete copies of the S-30 and S-13 videos. The soundtrack of S-13 confirms that the range units are in metres.]

4.  Broadcast in the UK as a brief item in the programme 'Strange But True? Live' (Granada Television) on 27 June 1997.

5.  'The Nellis Range UFO Video' by David Rudiak (1997).  Rudiak's updated article can be seen at his Roswell Proof website.

6.  Susan Michaels, 'Invader on the Range' in 'Sightings: UFOs', Simon & Schuster, New York (1997), pg. 122.

7.  Ibid., pp. 124-5.

8.  Opinion is divided among researchers as to whether black project tests are taking place outside of US borders.  According to Spanish Inexplicata editor Scott Corrales, "advanced technology aircraft ... are often tested far beyond the U.S. border to circumvent domestic aviation restrictions and with the consent of foreign governments."  See 'Spy Planes or Saucers?' by Scott Corrales, originally published in FATE magazine (September 2000) and reproduced in UFO Magazine (UK), Jan/Feb 2001, pgs. 54-9.  For an opposing view, see 'Theories of Origin' in Rudiak (ref. 5).

9.  John Velez, in the Virtually Strange 'UFO Updates Mailing List', April 1997, here.

10.  For numerous links to sites discussing Micro Air Vehicles, see the Hargrave website.  These flying machines have design specifications of 6 x 6 inches in size and just 4 ounces in weight.

11.  A comprehensive listing of past and present UAVs worldwide can be found at Wikipedia.

12.  However, according to the Association for Unmanned Vehicle Systems Inc., Aerial Robot developers have usually abandoned blimp or balloon type designs in favour of heavier-than-air vehicles. Blimps and balloons tend to drift in light breezes, which overpower their small propulsion units.

13.  Norio Hayakawa, "AREA 51: Emerging Technologies for Global Governance".

^ Back to Top

Copyright  Martin J Powell  May 2002 with minor amendments (to external links) January 2012

The Nellis UFO Video of 1994:


Part 1 (Location)

Part 2 (Appearance)

Part 3 (Performance)

S-30 Transcript

Mobile Site


The Trindade Island

UFO Photos of 1958

The Rouen UFO

Photo of 1954

The Hawaii UFO

Photo of 1989


Flying Objects

Site hosted by  TSOHost