ALPA's letter concerning San Luis Obispo midair

The following is in the Matter of...

Investigation of Accident Involving
Wings West Airlines, Inc., Beech
C99, N6399U, and Aesthetech, Inc.,
Rockwell Commander 112TC, N112SM, near
San Luis Obispo, California, on
August 24, 1984

On March 1, 1985, Harold Marthinsen, Director, Accident Investigation Department of ALPA, wrote to the NTSB stating ALPA's findings in relation to the San Luis Obispo midair collision. On page 6 of that correspondence, under 1.15 "Test and Research," it stated...

[Return to "Reasonable Doubt" Home]

The investigators observed a Wings West Beech C-99 check ride which included 
operations in a scenario similar to the accident.
...
A departure commenced on Runway 29 and tracked the runway 11 localizer course 
outbound.  Los Angeles Center was contacted to determine when a target was 
observed by the controllers.  Radar contact was established at approximately 
2,200 MSL at about the outer marker.  Several VFR traffic advisories were 
pointed out by ATC to the flight crew....

As you recall from my slide show and my two papers available here on my web site, once a controller starts a track on an aircraft and it has a Full Data Block, supplemental tracking would then be available. Therefore, even if that aircraft then flew into sort box 1114, as long as PRB was designated supplemental, coverage for that tracked target would continue, uninterrupted, giving the illusion of total and uninterrupted radar coverage.

What follows is another portion of ALPA's findings. It begins on page 8...

1.16(a) DISCUSSION OF RADAR TARGETS AND DISPLAY ON CONTROLLER'S SCOPE

Testimony was heard at the public hearing from the four air traffic controllers 
who were working at or observing the radar scope for the Los Angeles ARTCC 
radar Sector R-15 on the day of the accident that no radar targets were seen in 
the vicinity of WW628 at the time of or just prior to the mid-air collision.  
Subsequent to the accident, a National Track Analysis Program, NTAP, computer 
printout was provided by the Los Angeles ARTCC which contained a printout of a 
VFR target which was identified as N112SM, the aircraft that collided with 
WW628.  This aircraft was identified on the NTAP printout as a VFR target, 
squawking the normal 1200 code and having an altitude encoding transponder.  
The Assistant Manager for Automation at the Los Angeles ARTCC, provided 
testimony describing the flow of the radar data from the radar antenna through 
the ARTCC computers to final display on the controllers radar scope or Plan 
View Display, PVD.  Figure 1 represents a simple flow chart which summarizes 
the witness' description of how the radar data proceeds to the PVD.  The 
witness stated that the data saved on the Systems Analysis Recording, SAR, tape 
from which the NTAP generates the above mentioned radar printout is retrieved 
prior to the point where that data is received at the Central Display Channel,

Continued on page 9...

CDC.  The CDC contains software which further manipulates the data and passes 
it for display on the PVD.  Unfortunately, the witness was unable to provide 
information on what type of data manipulation takes place during the time the 
data passes through the CDC and is displayed on the PVD because these two 
systems, the CDC and the PVD are maintained by the Airway Facility personnel 
and are not under his supervision.  The witness did state, however, that to his 
knowledge there is no reason why the data which is contained on the NTAP 
printout would be manipulated by the CDC and the PVD in a way which would cause 
it to not be displayed for the controller on the radar scope.  Therefore, we 
are presented with inconsistent information to aid in determining why N112SM 
was not seen by the controllers.  From all indications, this VFR aircraft 
should have been visible for approximately 7 minutes on the Sector R-15 radar 
scope, but none of the controllers who were monitoring this position remembered 
seeing this target.  The Air Line Pilots Association believes that the 
elimination of these inconcistancies must be accomplished before a correct 
determination of the possible cause of this accident can be determined.

In attempting to eliminate the inconsistancies, ALPA looked at two 
possibilities; first that it was possible that the target of N112SM, although 
appearing on the NTAP radar data printout, was somehow erased from the Sector R-
15 radar scope, and second, that the target was on the radar scope and the 
controllers just missed seeing it due to inattention.  Each of these two 
hypothesis will be discussed individually below.

The first hypothesis would involve a malfunction within the computers which 
control and assist the flow of the radar data to the PVD.  As was stated 
before, the specialist of the Los Angeles ARTCC testified that if a radar 
target is represented on the NTAP printout then this target will also make its 
way through the CDC and to the PVD for display.  This testimony was reiterated 
by the Manager of the FAA Enroute Automation Branch in a deposition held 
January 29, 1985.  This witness, who is responsible among other things for the 
software and hardware contained in the CDC and the PVD, presented testimony 
which confirmed that there is no malfunction within the ARTCC radar computer 
system which will cause a target which appears on the NTAP printout to somehow 
be deleted prior to its presentation on the PVD.  In his testimony, however, 
this witness stated that the controller has the capability, using specific 
controls on the PVD, to inhibit targets from the PVD over certain altitude 
ranges.  These keys allow the controller to inhibit certain transponder codes 
within specific altitude ranges.  The witness testified that the only reason 
why a VFR target, squawking 1200, interpreted and processed by the NTAP program 
would not appear on the PVD is if the controller had deselected the 1200 codes 
within that corresponding altitude range.  Therefore, using the testimony of 
the two FAA specialists in addition to the fact that the track of N112SM was 
presented on the NTAP printout, it must be concluded that the radar target for 
this aircraft was available for display on the Sector R-15 radar scope just 
prior to the mid-air collision.

The four controllers testified at the public hearing that there were no targets 
in the vicinity of WW628 prior to the loss of WW628 transponder signal.  The 
developmental controller testified that on the day of the accident, he had his 
PVD set up on the following way:

    1. Altitude filter set from 0 to 24, 200 feet;
    2. Non Mode-C targets selected; and
    3. Limited data blocks not selected.

Continued on page 10...

With the PVD set up in this manner, the radar scope would be displaying targets 
from the ground to 24,200 feet altitude and would be displaying all aircraft 
with functioning transponders regardless of whether they were displaying 
altitudes or not.  Also, the full data block for each target would be 
displayed.  It cannot be absolutely confirmed what the PVD was set up for at 
the time of the accident because the key positions are not recorded, but some 
strong conclusions can be drawn using other available information.

At 1816:09 the Sector R-15 controller gave the following traffic advisory, "Sun 
Air nine twenty-two traffic twelve o'clock two miles southwest bound VFR 
showing nine thousand six hundred unverified."  This statement confirms that 
approximately one and a half minutes prior to the accident that the Sector R-15 
PVD was displaying a VFR target which was squawking 1200 and displaying a Mode-
C altitude of 9600 feet.  The witness from the Enroute Automations Branch of 
the FAA testified that a controller using the altitude filter keys can inhibit 
targets from 0 to 7000 feet altitude using one key, 7000 to 14,000 feet with 
another key independently.  At 1746:28 on the expanded ATC transcript, the 
Sector R-15 controller gave the following traffic advisory, "Bonanza two one 
bravo charlie, traffic eleven o'clock, eastbound, he's eleven o'clock and three 
miles eastbound, VFR, out of six thousand one hundred unverified, appears to be 
climbing."  These traffic advisories mentioned above provide strong evidence 
to support the conclusion that the PVD was set for displaying, VFR Mode-C 
targets in the altitude range in which the accidednt took place, i.e., 0 to 
7000 feet.  The Sector R-15 position is responsible for low altitude traffic 
control in that area of Los Angeles ARTCC.  It is known that the controller, 
just prior to the accident was receiving a VFR, Mode-C target at an altitude of 
approximately 9600 feet because of the 1816:09 traffic advisory.  It is also 
known that approximately a half-hour prior to the accident the controller gave 
a traffic advisory at 6000 feet.  The fact that the Sector R-15 is a low 
altitude control position and that the controller testified that the PVD was 
set up to display VFR targets from 0 to 24,200 feet, leads the Air Line Pilots 
Association to believe that the VFR Mode-C targets in the altitude range of 0 
to 7000 feet were being displayed the day of the accident.  It does not make 
any sense for a low altitude ATC control position such as Sector R-15 not to be 
displaying VFR targets in the 0 to 7000 foot altitude range where the majority 
of VFR traffic would be expected to be located.  The developmental controller 
testified at the public hearing that he believed that the PVD for Sector R-15 
was set at the time of the accident to display the Mode-C, VFR targets from the 
ground level to 24,200 feet.  The related facts of this accident coincide with 
the controller's testimony, therefore, leading the Air Line Pilots Association 
to conclude that the position of the altitude filter keys on the Sector R-15 
PVD at the time of the accident had no effect on the display of the radar 
target for N112SM.

Therefore, considering the testimony from the FAA radar specialists, the 
developmental controller and the ATC transcript, Exhibit 3G, the Air Line 
Pilots Association concludes that the target for the Rockwell aircraft N112SM 
which was presented on the NTAP printout, was in fact displayed on the Sector R-
15 radar scope just prior to the mid-air collision.  With the elimination of
the hypothesis that the target of N112SM was not displayed for the Sector R-15 
controller on the PVD, the only other conclusion which can be drawn is that the 
target for N112SM was not seen by the controllers due to inattention.

The actual reason why the target for N112SM was not identified by the 
controllers working the Sector R-15 is unknown.  The controller stated that 
just prior to the time the Wings West flight contacted him he had been involved

Continued on page 11...

in a non-standard traffic sequencing exercise between a United Flight 1265 and 
a slower Sonic Airlines Flight 766 which were both arriving into Santa 
Barbara.  This sequencing problem involved the faster United Jet overtaking the 
slower Sonic Airlines aircraft as they both proceeded into Santa Barbara.  The 
Sector R-15 controller was at the time receiving his check-out on Sector R-15 
by another controller, a full proficiency controller.  The sequencing of these 
two aircraft was deemed a good test for the Sector R-15 developmental 
controller and was the reason that the on-duty supervisor decided to observe 
the check-out at that time.  This sequencing problem however had been resolved 
approximately one minute prior to the accident since the Sonic Airlines 
aircraft had been handed off to Santa Barbara approach at 1816:31 and was no 
longer the responsibility of the Sector R-15 controller.

Just after the Sonic aircraft had been handed off, 1816:40, the Sector R-15 
controller acknowledged a previous radio call from Wings West 628 which 
occurred approximately 40 seconds earlier.  Wings West requested an IFR flight 
plan to San Francisco and was given a discrete transponder code.  Approximately 
16 seconds later, the controller radar identified WW628 six miles northwest of 
San Luis Obispo Airport and verified the flight's altitude.  At this same time, 
using the NTAP printout, N112SM was approximately 2 miles from WW628 traveling 
in the opposite direction at 3700 feet.  It is difficult to understand how the 
controller could have missed the target of N112SM at the time 1817:16 when he 
stated that he had radar contact with WW628.  It is assumed that at the time 
when radar contact was made that the controller was looking in the vicinity of 
WW628 and thus would have observed the N112SM target if it was there.  It is 
inconceivable that a controller with his attention focused on an IFR target 
would not see another target only two miles away on a converging track and 
displaying a Mode-C altitude which is close to the altitude of the IFR target.  
It is possible that the developmental controller, the full proficiency 
controller, and the suppervisor were so preoccupied with following the two 
aircraft involved in the sequencing problem into Santa Barbara discussed 
previously, that they did not pay sufficient enough attention to the remainder 
of the scope.  In their effort to train the developmental controller, the 
supervisor and the full proficiency controller possibly were discussing the 
sequencing problem further with the trainee.  When WW628 called requesting 
their IFR clearance, insufficient attention was directed to this part of the 
radar scope.  The target of N112SM would have been a required safety advisory 
by the Sector R-15 controller to WW628, but it was not made.  Therefore, the 
only conclusion that can be drawn is that the controller had his attention 
improperly focused elsewhere on his radar scope and did not observe N112SM.

The Air Line Pilots Association feels that the facts surrounding the 
circumstances of this accident indicate that the controllers due to inattention 
to the radar scope, missed the target of N112SM and therefore did not provide a 
safety advisory which might have averted this accident.  Since it has been 
testified to by FAA expert witnesses that the target of N112SM had to be 
displayed on the Sector R-15 PVD prior to the accident, no other conclusion can 
be reached after analyzing the additional information surrounding this accident...

As you notice from ALPA's investigation, they did not learn about the methods in which radar data is processed, even though these methods have been in the specs since the original mosaic radar data processing model was accepted.

[Return to "Reasonable Doubt" Home]

The above text was obtained from microfiche as obtained from General Microfilm of Wheaton, MD.

This text was converted from microfiche, to paper, then scanned for Optical Character Recognition (OCR) processing, followed by a great deal of manual editing by Tom Lusch for correct presentation on "Lusch's Midair Collision Investigations" web site <http://home.columbus.rr.com/lusch>. Any errors or omissions resulting from this microfiche to web page process are solely the responsibility of Tom Lusch. Every effort was made to make this an accurate representation of this deposition transcript, right down to mis-spelled and/or incorrect words.

This page created Feb 25, 2000