AVweb

« Back to Full Story

Probable Cause #55: Now What?

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A
This article originally appeared in IFR Refresher, May 2007.

Probable Cause

The VOR/DME-A approach into Weatherford, Texas, contains a specific missed-approach procedure, as do all instrument approaches. The concept is simple: When you get to the missed-approach point (MAP), the plane should be at the minimum descent altitude (MDA) and -- if the airport is not visible or the aircraft is not in a position to make a safe descent to the runway -- the missed approach must be flown. Yet on Jan. 15, 2004, the pilot of a Beech B36TC Bonanza failed to do so. In fact, he told ATC that he was "unable to finish the approach ... Need climb out instructions." The airplane crashed shortly after the Fort Worth Center controller cleared him to fly directly to Odessa, Texas. The pilot, the sole occupant of the aircraft, was killed in the crash.

Changing Weather

Beech B36TC Bonanza

The Bonanza departed Houston's Hobby Airport at 4:20 p.m. on a 220-nm flight to the Parker County Airport (KWEA) near Weatherford, a small community just outside the 30-nm veil that surrounds the Dallas/Fort Worth International Airport. The aircraft had been at Hobby for maintenance and the owner was picking it up to take it back to Odessa, where the airplane was based. The stop at Weatherford was planned so the pilot could pick up a dog. Earlier that day, at 9 a.m., the pilot had called the San Angelo Automated Flight Service Station for a weather briefing for the flight. The briefer advised the pilot that an AIRMET for icing for the entire route of flight had been issued a few minutes earlier. It called for the conditions to continue beyond 9 p.m. that evening. The prognostic chart showed an area of high pressure in the Gulf of Mexico that was drawing moisture across the state, promising marginal VFR and IFR conditions across the route of flight. A stationary front in the area between Midland and Lufkin was expected to result in light rain and IFR conditions later in the day. For the departure, Houston was forecasting ceilings of 800 broken with two miles visibility in light rain and mist, with the surface wind from 100 degrees at 7 knots. En route, the ceilings were forecast to be between 600 and 1,000 feet overcast with visibilities between three and five miles, and occasionally down to two miles in light rain and mist. For the stop at Weatherford, the Fort Worth forecast indicated ceilings of 2,500 broken with an occasional ceiling of 1,500 broken with three to five miles in light rain and mist until 7 p.m. To an experienced instrument pilot, the forecasts would not cause undue anxiety. But experience also provides that whenever the weather is marginal around sunset, expect anything to happen, especially if there is moisture in the atmosphere. More on that later. The Bonanza pilot then told the briefer that he would call back before departure to file his IFR flight plans, stating that all of his preflight planning data was in the aircraft and that he wasn't sure he would be making the stop at Weatherford but that he would know when he called back. The pilot did so at 1:45 p.m., and filed his flight plan to Weatherford with direct routing at 12,000 feet, listing Odessa as his alternate. However, he did not ask the briefer for an update on the weather, even though by now nearly five hours had passed since he had first received his weather briefing.

After The MAP

The flight proceeded normally and there were no communications difficulties with the Bonanza. At 5:52 p.m., the pilot was instructed to descend at his discretion and maintain 3,000 feet. That instruction was acknowledged and at 5:53 p.m. the pilot was cleared by Regional Approach Control for the VOR/DME-A approach into Weatherford. That approach aligns the aircraft on the 077-degree radial from the Millsap VOR. The approach requires that an aircraft should maintain at or above 3,000 feet until crossing the PANTR final approach fix (FAF), 11 miles east of Millsap. At that point, the aircraft can descend to the circling altitude of 1,680 feet. The MAP is at the 16-nm DME fix from Millsap and the missed-approach procedure calls for the pilot to climb to 1,900 feet, then continue in a climbing right turn to 3,000 feet via a 290-degree heading and the Millsap 077-degree radial to PANTR to hold. When the controller cleared the pilot for the approach, the transmission was properly acknowledged, but then the pilot added, "I'm only gonna make one attempt and at that time ... I want climb-out instructions direct to [Odessa]." There is nothing unusual about requesting a missed-approach procedure from ATC that is different than what is published, especially if it makes operational sense. But the new instructions are typically issued before the approach is begun. Also, in this case, deviating from the published missed didn't make much sense, because it required a climb to the west, in the general direction of Odessa. The controller confirmed the pilot's intention to go to Odessa if he didn't get in but did not issue any instructions as it regarded to the missed approach. The controller did provide the current weather at Mineral Wells, approximately 19 nm west of Weatherford, which did not have weather reporting capabilities. The visibility at Mineral Wells was half a mile in fog, with a broken layer at 200 feet and an overcast layer at 1,200 feet. The pilot acknowledged the report and left the frequency for his approach. At 6:07 p.m. the controller noticed that the aircraft had begun a climb out of the Weatherford area. Shortly thereafter, communication was reestablished with the aircraft and the pilot told the controller, "I am unable to finish the approach. I need some climb-out instructions direct to [Odessa]." The controller cleared the Bonanza via a left turn direct to Odessa and told the pilot to maintain 3,000 feet. The pilot acknowledged the clearance, then asked the controller what heading he should fly. The controller asked the pilot if he was capable of flying direct, to which the pilot answered that he was and that he would do so. The controller then coordinated with a nearby sector for the aircraft to climb to 6,000 feet, but when he called the Bonanza pilot to issue the clearance, there was no response. No further transmissions were received from the aircraft.

Few Clues

Excerpt from Weatherford, Texas, (KWEA) VOR/DME-A approach chart. Click here for full chart.

The aircraft crashed approximately two miles northwest of the Weatherford Airport near the town of Willow Park. There was no post-impact fire. The airplane came to rest in a ditch, which was a roughed out road for a future subdivision, at an elevation of approximately 840 feet. The landing gear was found in the extended position, the flaps were extended to 15 degrees and flight control continuity was established by investigators. No evidence was found of any pre-impact system failures that might have contributed to the accident. There were two witnesses who lived near the crash scene. One was in his home when he heard the sound of a loud, accelerating engine noise coming from an area directly north of the house. He heard the sound of the impact, called 911 and then left the house to search for the accident site. The second witness said it was raining at the time of the accident and that the fog "was so bad you couldn't see anything." The pilot held a Private pilot certificate for single-engine land and a rating for instrument airplane. His last medical certification was during October 2002. He had accumulated a total of 997 flight hours, 161 hours of night flight and 130 total hours in actual instrument conditions. He had 700 hours in the same make and model aircraft he was flying. A radar-data review indicated that the target initially tracked along the inbound course of the VOR/DME-A instrument approach at 3,000 feet. After crossing PANTR, it began a descent to 1,700 feet before climbing back to 1,800 feet and initiating a left turn to the north-northeast at the MAP. The last 1-1/2 minutes of flight data revealed that after the target turned to the north-northeast it maintained an altitude of 1,800 feet for about 25 seconds before it made a right turn toward the east, followed by a turn to the north-northwest. The data ended shortly after that turn. The NTSB determined the cause of the accident was the pilot's loss of control while maneuvering, adding that the dark night and rain were contributing factors. What Went Wrong? The task of investigating a GA accident is an arduous one, especially when the cause is not immediately obvious. In accidents where human error is suspected, the lack of a cockpit voice recorder or flight data recorder makes finding the cause even more difficult, if not impossible. Without any hard evidence, any plausible scenario that could have led to the accident is just going to be speculative. Combine that with the limited resources and manpower that the NTSB and FAA have to work with, and you end up with very broad and generic determinations, such as the obvious "the pilot lost control while maneuvering," but very little else. So if we want to learn from this accident, all we can do is speculate, based on the scant evidence that is available in the NTSB report. Keep in mind that we are just discussing what could have happened in this accident, none of which may have had any bearing on the true events of what happened that January night. Among the evidence that investigators found in the wreckage was a glucose monitor, syringes, lancets and a vial of human insulin. When investigators asked the pilot's family about this, they confirmed that he was a diabetic. However, the pilot never disclosed this condition to the FAA. If he had, he would have been medically grounded, unable to qualify for the third-class medical certificate that he had in his possession. Hypo- or hyperglycemia -- low or high blood sugar -- is a serious medical condition that can result in altered mental status, unconsciousness and even death. The onset can be accelerated by the stresses you'd find flying an instrument approach in lousy weather. However, there is no evidence that the pilot's diabetes had any affect on his flying that night. Investigators created a radar plot that showed the aircraft's track in relation to the final approach course. The plot showed that he was slightly right of the course for most of the approach, but that can easily be explained by the inherent errors that exist in the VOR receiver and the OBS. The plot also showed that the pilot maintained 3,000 feet until reaching PANTR before beginning a 1,200 fpm descent to around 1,700 feet. Subsequent plots show the airplane at around 1,800 feet but the discrepancy is likely due to normal radar and transponder errors. So based on the plot, the pilot flew a normal approach until he reached the MAP, at which time things started to go seriously wrong. Instead of initiating a climb to 1,900 feet and then a climbing right turn to 3,000 feet as the procedure required, the pilot began a turn to the left and never gained altitude. Neither did he raise the flaps or landing gear, hampering the aircraft's climb capability. The question, then, is why did he completely botch the missed approach after flying a perfectly normal approach all the way to the MAP? One possible clue is the pilot's request, before he even began the approach, for climb-out instructions for a direct course to Odessa. Did he even look at the missed-approach procedure? Did he even have the instrument procedure with him? For whatever reason, it appears that he was unfamiliar with what the published missed-approach procedure was. This is where the accident report is lacking. We know the airplane was built in 1985, but the report does not say what types of avionics were on board or under what suffix the airplane was filed. It did not have a GPS when it rolled off the factory line, but one could have been added later. Or perhaps the pilot had a handheld GPS. The fact that the pilot told the controller he could go direct to Odessa indicates that he had some form of RNAV capability. With the little that we know, we can come up with a few hypotheses. One possibility is that during the course of setting up the approach, he dropped the chart and was unable to retrieve it. Armed with the rudiments of the approach -- like the MDA -- and backed up by GPS, he decided to press on, hoping that he would break out in time to see the airport. However, when he reached the MAP, he was at a loss since he didn't have the missed-approach procedure. Things unraveled, perhaps he panicked, and he subsequently lost control of the airplane. The decision to press on may have been driven by the weather forecast he had received earlier that day, which called for better conditions than actually existed. There is no evidence that he contacted flight service en route for an updated briefing, although the Fort Worth controller did give him the Mineral Wells weather before he began the approach into Weatherford. Another scenario, although not as likely, plays out along a similar theme. What if the pilot didn't have the approach plate for KWEA? He wouldn't be the first pilot to try to sneak into an airport with just GPS guidance. It should be noted, too, that the NACO version of the VOR/DME-A approach to KWEA is found in the SC-2 volume, while Odessa and Hobby approach are found in SC-3 and SC-5, respectively. If he used NACO plates (the NTSB report doesn't specify), it is possible he wasn't carrying the volume with him or did not have access to it. This would be less likely if he was a Jeppesen subscriber. This theory also raises another question. If he didn't have access to the approach chart, how did he know the MDA? What if he had just added 800 feet to the airport elevation, with the knowledge that terrain is not an issue? Granted, that's a big "if," but not out of the realm of possibility. Again, he could have been bolstered by the belief that the weather was better than it really was and he would break out and complete the approach visually. The truth is that we'll never know. This accident raises more questions than answers. But it's through discourse like this that we can reduce the chances of repeating a similar event. And while much of it is conjecture, we can learn from accidents like this. Preparation is the key for a successful instrument flight, especially when we are flying single-pilot. We know from a comment the pilot made to the Flight Service briefer that he had never been to Weatherford before, so he was not familiar with the approach or the airport. A complete self-briefing on the approach is necessary. That briefing should include the details of the approach itself as well as the missed approach. Once a pilot stops flying the airplane, he is only along for the ride. He will go wherever the airplane goes, not where he wants to go. Riding in an airplane that you are responsible for piloting is not conducive to a safe landing.
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.

« Back to Full Story