Home Page || Flight 587 Page || Your Comments @ U.S.Read's Web Log || Send E-Mail to the Editor || Search U.S.Read or the Web ||

U.S.Read's Flight 587 Preliminary Report

Part 5:  The Radar Data

October 13, 2004

by Brett Hoffstadt and Victor Trombettas

In this part of our preliminary report we present the radar data which indicate that the tail and engine separations on Flight 587 (FL587) occurred later than established by the NTSB's timeline.  This would mean, among other things, that the crew lost control of the aircraft before the tail departed. Therefore, the NTSB has not identified the real cause of the loss of control –– nor have they identified the initiating event.

The primary radar data from JFK airport showed the following (other key events in the timeline are added for reference):

9:15:58.50 a.m.:  After several seconds of aggressive control inputs and three calls for maximum power by the pilot, the "loud bang" is heard on the Cockpit Voice Recorder (CVR).  The NTSB assumes the tail has just separated with the rudder breaking up into at least 12 pieces. The aircraft is now 0.82 NM (nautical miles) north of the crash site and traveling southerly at an altitude of 2,535 feet (see Figure 1).

Figure 1

9:16:01.20 a.m.:  All Digital Flight Data Recorder (DFDR) data abruptly ends yet the DFDR is still powered for several seconds beyond this point. The aircraft is at an altitude of 2,486 feet.

Why has the DFDR lost the entire data stream?  One possible explanation is that an onboard fire/explosion has taken out the main data line that runs under the cabin floor on the right side of the aircraft. The NTSB has stated on more than one occasion that the engines (which power the DFDR) had both separated by this time and that was the reason for the loss of all data.  But in late 2002, the NTSB revealed that the DFDR was indeed powered beyond the point where it lost the data stream. They have yet to explain how and why the data stream was completely lost –– 13 seconds before impact and before the aircraft had even begun its descent.  This is one of the more substantial clues that suggests a significant event on board created electrical disturbances, and could well explain the pilot's loss of rudder control and the failure of the wing spoilers.

U.S.Read is in complete agreement with the NTSB that FL587 is out of control at this time.  Pilot Sten Molin even tells us that himself –– as we hear him say in a stressed voice, "Losing Control", on the Air Traffic Control (ATC), Local Control tape.

At this point in the timeline (and at this point over Jamaica Bay), if the NTSB's conclusions about how and when the aircraft broke apart are correct, we should have multiple pieces of aircraft debris that appear on the JFK radar in the range of 0.60 to 0.82 NM (nautical miles) north of the crash site.  The tail and rudder debris should be closer to 0.82 NM, and there should be some evidence of both engine separations near 0.60 NM.

What the JFK radar data shows
9:16:01.94 a.m.:  Only one piece of debris, 0.68 NM north of the crash site, is seen on the JFK radar data. The aircraft itself is now 0.58 NM north of the crash site (Figure 2) at an altitude of 2,398 feet.

Figure 2

According to the NTSB's timeline, we would expect to see some of the more than one dozen pieces of airborne debris on the radar data.  But we do not –– we only see one.

If the NTSB believes this farthest piece of debris is the tail, then the implication is that in the 3.44 seconds between tail separation (9:15:58.50) and the radar return at 9:16:01.94, the tail has traveled forward 0.14 NM (850 feet –– almost the length of three football fields) in those 3.44 seconds. That would mean the tail flew forward (in the same direction as the plane) at an average speed of 146 knots over those 3.44 seconds. 

Such a feat is aerodynamically impossible according to our contributing experts.  It is also contrary to all witnesses who saw the tail separate, and reported that the tail immediately fell away from the airplane. That is why some witnesses described the tail separation as "a large piece flying off to the north" (as the plane was traveling south).

As illustrated in Part 1 of our Preliminary Report, the in-flight separation of an airfoil from an F/A-117 stealth bomber in 1997 showed exactly the same phenomenon –– the airfoil will quickly bleed off any forward momentum and fall straight down:

You can view this video at:

If this debris seen on radar at 9:16:01.94 is the vertical tail, it broke off later than the NTSB has declared in order to show up where it did on radar. This would change the nature of this crash, and prove that the tail separation came after the aircraft was already out of control.

If this debris is not the tail, then even more so this challenges the NTSB's timeline because it means something else separated before the tail did. We believe this is a critical point, and our assessment is that this piece on radar at this time is not the tail.  We take this position because  the NTSB believes that when the tail broke off it created an airborne debris field of more than a dozen pieces (we agree with them on this point –– this is consistent with the eyewitness statements and the debris field).  More than one of those pieces would be visible on radar. This does indeed occur –– but not until later in the crash sequence –– as we will outline here. Also, some eyewitnesses have stated the tail was not the first debris to separate (some described debris from the fuselage or wings striking the vertical tail), and that the tail separated after they noticed the aircraft on fire, and lower in the sky than normal –– which would also mean the aircraft had begun it's descent before the tail separated. At this point in the timeline, at 9:16:01.94, the aircraft has yet to begin descending.  In fact, it won't begin losing altitude for another four seconds. 

9:16:06.20:  A bright flash and a huge smoke/mist trail emerge from the aircraft (visible on the tollbooth video). It is at this time the aircraft begins to lose altitude in a nose-up and/or horizontal attitude (Figure 3 below).  The aircraft is at an altitude of 2,428 feet (from the NTSB's Tollbooth Video Study).

Figure 3 - from tollbooth video (video frame 91653 frame 22) –– just after the "flash"
(video frame times are not the same as Air Traffic Control times)

9:16:06.43:  FL587's transponder does not reply to the JFK radar.  The transponder reports back to the radar facilities on the altitude of the airplane. Why wasn't the transponder functioning? The NTSB hasn't explained this significant malfunction, and the plane is just beginning its descent (it is still at an altitude of 2,400 feet), still in a horizontal attitude, as confirmed by the tollbooth video in Figure 3 above.   Therefore, it cannot be argued that the aircraft was wildly out of control at 9:16:06, and in an unusual attitude, and therefore could not detect the radar signal and reply. Along with the cessation of DFDR data, the failure of the transponder suggests something catastrophic had occurred on board which negatively affected critical systems. 

9:16:06.35 to 9:16:07.11:  These JFK radar returns show the farthest item is 0.63 NM north of the crash site and there is only one item at that range. However, the radar shows two items at 0.33 and 0.40 NM north of the crash site (Figure 4). 

Figure 4

Why is there more debris closer to the crash site and still no debris near where the NTSB says the tail separated, at 0.80 NM? That is where most of the debris should be (if the NTSB timeline is correct). 

9:16:11.04:  FL587's transponder does not reply to the JFK radar again. Though the aircraft is descending, it is still in a mostly horizontal attitude as shown on the tollbooth video (Figure 5). 

Figure 5 - from tollbooth video (video frame 91657 frame 20)

9:16:11.06 to 9:16:11.51:  Aside from the aircraft itself, these JFK radar returns show four airborne objects behind FL587 (Figure 6).  Flight 587 is at an altitude of approximately 1,600 feet.

Figure 6

Three of the four are 0.09 to 0.46 NM north of the crash site –– a significant distance from where the NTSB claims the tail and engines separated.  The absence of debris at 0.80 NM supports what the majority of witnesses have told U.S.Read about when the tail separated over Jamaica Bay and where the majority of rudder debris was in fact recovered –– much closer to the Bay seawall than the NTSB has reported. 

9:16:14.78:  End of Cockpit Voice Recorder (CVR) recording (assumed time of aircraft impact).

Taken alone, the JFK radar data suggests that tail separation and engine separations occurred after the flash and smoke/mist trail seen on the tollbooth video at 9:16:06.2.  That is because the radar sees only one piece of debris before this time.  Several pieces of debris weren't seen until 9:16:11.06 –– more than twelve seconds after the NTSB claims the tail departed

The tail, rudder, and engines were the largest items that separated in-flight and were therefore the most likely objects to be seen on radar.  But this cluster of objects on radar appeared much later than the NTSB claims the tail and engines fell from the aircraft.  And if this cluster of debris seen by radar after 9:16:06 is the tail and engines, as we have concluded, what was the debris seen at 9:16:01?

As with all the evidence, a close look at the radar data creates more challenges for the NTSB in their search for answers –– the radar data certainly does not support the NTSB's conclusions. 

As we have been outlining in our multi-part Preliminary Report, our conclusion is that all the evidence suggests the tail departed after the crew had lost control of the aircraft –– that tail separation was not the cause of the crash.

In Part 6 of our report, we will examine the eyewitness accounts.

We welcome your comments and feedback on this article at U.S.Read's new web log at: or you can send us an email by clicking here.

Home Page || Flight 587 Page || Your Comments @ U.S.Read's Web Log || Send E-Mail to the Editor || Search U.S.Read or the Web ||