Probable Cause #45: Out of Position

A Skylane pilot crashes during a foggy ILS approach. Is it possible his HSI was showing data from the wrong navigation source?

This article originally appeared in IFR Refresher, Dec. 2006.

It's not too difficult to understand the theory of the ILS approach. The pilot is required to fly a horizontal radio beam while referring to a cockpit indicator that tells him whether he is left or right of the course. At the same time he must fly a vertical radio beam by looking at a second needle in the same instrument that will inform him of his position above or below it.When the weather is good and there is plenty of outside reference to the ground it can be very easy to fly. But take the weather down to IFR minimums in fog and the pilot's workload increases dramatically. Now, his eyes must dance across the instrument panel taking pictures of what each of the flight and navigation instruments is telling him. His brain must meld all of the images into a coherent mental reference of where the airplane is relative to the approach he is flying. Then, he must do it again and again and again.Flying a good approach requires constant practice and the pilot must be current and proficient at all phases of instrument flight. After all, flying an approach to minimums is the most demanding instrument flying a pilot is likely to do in normal, everyday operations. If he is lacking any of the basic skills, he will have problems with the ILS.

The Set Up

300

Near the end of September 2003, a Cessna 182T crashed while attempting to make a missed approach following an ILS to Runway 11 at Hanscom Field (KBED) in Bedford, Mass. The private pilot and his passenger were killed in the crash.The airplane had departed the Oswego County Airport at Fulton, N.Y., at approximately 9:10 that morning (all times EDT) in VFR conditions. The pilot had filed an IFR flight plan and, once airborne, he contacted Syracuse Approach Control for his clearance. The departure and en route phases of the flight were uneventful.As the aircraft neared its destination, the pilot was turned over to Boston Approach Control. At 10:49 a.m., the Boston Approach controller acknowledged the pilot's call that he was at 6,000 feet and asked if he had received the Bedford weather. The pilot reported that he had.At 10:20 a.m., the Bedford ASOS reported an overcast sky at 400 feet with 3/4-mile visibility. The winds were from 100 degrees at 8 knots, temperature and dewpoint were 18 degrees C and the altimeter setting was 30.11 inches.At 10:50 a.m. the pilot was told to descend to 5,000 feet and four minutes later he was cleared for the ILS approach to Runway 11. At 10:58 a.m. the controller advised the pilot that radar service was terminated and he was instructed to contact the Hanscom Tower.

Unraveling

250

An accident sequence can be thought of as a series of strands on a string that begin to unravel. The accident occurs at the point when the string breaks. The difficult part is determining exactly when the unraveling begins to take place, but there can be signs.The transcript of the communications between the pilot and the tower controller certainly contains signs that events were beginning to unravel for the pilot as he was setting up for the approach. When the pilot contacted the tower controller, he was flying at 2,700 feet. The controller asked that the pilot report crossing the outer marker and advised him that the Runway 11 RVR was more than 6,000 feet.The pilot asked for a repeat of the transmission, with which the controller complied. A missed call in itself is no big deal, but when it becomes a pattern, as the transcript shows, it points to bigger problems ahead.At 10:59 a.m. the pilot reported that he was at the outer marker. The controller cleared the aircraft to land and advised the pilot that the last two aircraft broke out at minimums. This time, the pilot did not immediately reply, another indication things were far from perfect in the cockpit, as evident by the following exchanges. (Pilot transmissions are in italic, ATC transmissions are in bold):11:00:45 -- "Nine Six Three Lima Pop …" [The aircraft's call sign]11:00:50 -- "Calling tower. Say again." There was no response.The glideslope intercept altitude is 1,700 feet MSL and, once established, one would cross the outer marker at 1,456 feet. The Cessna pilot, however, crossed the outer marker at 2,000 feet, 544 feet above the glideslope. He then descended to 700 feet in about 40 seconds, going below the glideslope, something that did no go unnoticed by the controller.11:00:57 -- "Nine Six Three Lima Papa, check your altitude. Altimeter is 30.12. Minimum descent altitude [is] 383 feet. Check your altitude immediately."Again, there was no immediate reply from the pilot until 11:01:24, when he transmitted, "... and Six Lima Pop ..."The controller then told the pilot that he was on a two-mile final and asked him his heading, a request he repeated at 11:01:41. This time, the pilot reported he was on a heading for 330 degrees.At that point the tower controller contacted the Boston Approach controller. He advised that the aircraft was off course and he was unsure as to what the pilot was doing. Boston Approach advised him that a low-altitude alert was observed and instructed the tower controller to issue the pilot an immediate climb to 2,000 feet.11:02:02 -- "Cessna Three Lima Papa, climb and maintain 2,000 feet, execute a missed approach and contact Boston Approach now [on] frequency 124.4."11:02:06 -- "Missed approach. Can you give me the heading?"11:02:09 -- "What heading are you now on, sir?" 11:02:13 -- "Heading 180." 11:02:16 -- "Alright. Just climb to 2,000. Contact Boston Approach now [on] 124.4." However, the pilot read back 124.0. The controller repeated the correct frequency but the pilot never responded and was not heard from again.

Into The Woods

300

The airplane crashed into a wooded area 5,700 feet north of the localizer approach course and 2.62 miles from the approach end of Runway 11. The debris field, which measured 250 feet long, was oriented on a 130-degree bearing.Several witnesses observed the aircraft prior to the accident. One witness was fishing from his boat in the Concord River. He was used to hearing airplanes on approach and he told investigators that he heard an airplane that morning that at first sounded normal. He then heard an application of power and a few seconds later he saw a white, high-wing airplane operating just above the tops of the trees, heading away from the airport. He did not hear the accident, adding that there was fog near the tops of the trees.Another witness reported that he heard an airplane turning where they don't normally turn. He never saw the airplane, but he did say that he heard it until it impacted the ground. According to the witness, the engine was not sputtering or missing. He then walked into the woods and found the wreckage, reporting that there was a strong fuel smell at the site.Investigators interviewed several pilots who were flying in the area around the time of the accident. None revealed any problems with the localizer or glideslope for Runway 11.The pilot earned his private pilot certificate in 1978 and added an instrument rating three years later. On his second-class FAA airman medical certificate obtained on March 20, 2003, he reported that he had a total of 2,600 hours in flight, 70 hours of which were in the past six months. Investigators estimated that he had flown 30 hours in the 90 days preceding the accident and about 10 hours in the last 30 days. It was estimated that he had 210 hours in the accident aircraft.The last entry in the pilot's logbook was dated August 29, 2002. It was a satisfactory flight review and instrument competency check in the accident airplane that was administered by an FAA operations inspector.

Off Course

300

At 10:56 a.m. the recorded weather at KBED Field included visibility of 3/4 miles, a ceiling of 400 feet overcast and winds from 090 degrees at 8 knots. Pilots flying through the area that morning indicated that the cloud tops were at 7,000 feet, and that they extended west to the Hudson River. The base of the clouds was about 450 feet. Local witnesses reported that clouds were at the tops of the trees.The ILS Runway 11 approach course is 113 degrees. The minimums for a straight-in approach are 383 feet (250 feet AGL), and the visibility requirement is an RVR of at least 5,000 feet. The published missed-approach procedure calls for a climb to 2,000 feet while proceeding straight ahead to the Shaker Hills NDB, which is about six miles east of the airport.A study of recorded radar data shows that the Cessna was above the glideslope and descending as it approached the outer marker. Once passed the marker, it began its descent through the glideslope to 700 feet. Soon after, the aircraft began a turn to the left and started to climb. While climbing, the aircraft proceeded to the northwest and, after a short time, began another turn to the left. While in the turn, radar data showed that the airplane's altitude was inconsistent and the radius of the turn began to decrease.The airplane completed a turn of more than 360 degrees, but as the radius of the turn decreased, the radar contacts began to appear too close together, making it impossible to determine the exact path of the aircraft during the last few seconds of flight. The last radar contact with the flight was just before 11:03 a.m. The indicated altitude was 500 feet and the airplane's position was about 245 feet northwest of the accident site.Investigators sifted through the wreckage in an effort to determine if there was any mechanical or electrical failure before the accident. No evidence of either was found.An examination of the navigation radios showed that one was tuned to the ILS frequency at Bedford while the other was damaged so severely that it was impossible to determine its frequency setting.

GPS/NAV

The Cessna was delivered new in 2001 and therefore came with a new and modern avionics package, including a GPS and an HSI. The configuration included a two-position, push-button switch mounted on the instrument panel below the glare shield that allowed the pilot to select whether GPS or nav data was depicted on the HSI. The setup does have a serious drawback: Unlike some other GPS/NAV configurations, selecting an ILS or localizer frequency in the active window of the nav radio does not override the GPS/NAV switch.The functionality of the switch and its possible effect on the accident received extra scrutiny by the investigators. A flight test was conducted with a similarly equipped C-182 to test the function of the GPS/NAV switch. It was determined that the glideslope pointer on the HSI would remain out of sight unless a glideslope signal was being received. That meant that the switch had to be selected to the NAV function and an ILS frequency had to be placed in the nav radio's active window to display a glideslope indication.Unfortunately, the NTSB report does not say if it was able to determine from which source the HSI was getting its information.As is typical with accidents of this type, the NTSB concluded that the pilot became spatially disoriented. The radar track of the aircraft tends to confirm that, but we have to question why a seemingly experienced pilot who had been flying regularly would lose control of his aircraft in the manner that the Cessna 182 pilot did.Any time a pilot becomes distracted from his primary duty -- flying the airplane -- he is risking losing control. Shortly after contacting the Hanscom tower controller, the pilot had to ask for a repeat of the RVR. Later, the pilot did not acknowledge the controller's transmission telling the pilot that the two aircraft in front of him broke out at minimums. When the pilot was told to check his altitude immediately and given the altimeter setting, it took him almost 30 seconds to respond. The pilot did not respond when the controller asked what heading he was flying when the aircraft was on a two-mile final. The controller asked a second time and, 12 seconds later, he responded. Finally, when the controller instructed the pilot to begin the missed approach procedure and climb to 2,000 feet, the pilot asked for a heading. It seems that either he had not studied the missed approach procedure or that he had forgotten it.It would seem that the pilot was distracted from the time he came on the tower frequency until he lost control of the aircraft. Some might argue that he was not, that he was doing what he was supposed to be doing, which was to fly the airplane.I would agree except the pattern of not answering the controller continues throughout the entire time he was on the tower frequency. Failure to acknowledge one transmission might be acceptable, but a series of unacknowledged calls or very slow acknowledgement seems to indicate that something else was going on in the cockpit.Then there's the potential issue of the GPS/NAV switch. Is it possible that the pilot did not engage the switch properly and that the HSI was set to GPS instead of the ILS? If he had built the approach into the GPS or selected the airport waypoint from a position close to the inbound course, the HSI's depiction would have looked the same or very similar to the localizer course. The pilot may then have become distracted trying to figure out why he had no glideslope information. It's important to note that the Cessna pilot did not tell the controllers that he was not receiving the glideslope.We have no idea how current or proficient the Cessna pilot was since there were no entries in his personal logbook after his review with the FAA inspector. It may be that he was lacking recent IFR time in spite of the recent hours he had flown.Spatial disorientation while flying on the gauges is a killer. Make sure you are current and proficient on the instruments before committing to any IFR flight. Watch the weather carefully so that you don't wind up facing conditions that you are not prepared for. And don't let inconsequential things distract you from your job of flying the aircraft.


More accident analyses are available in AVweb's Probable Cause Index. And for monthly articles about IFR flying including accident reports like this one, subscribe to AVweb's sister publication, IFR Refresher.