June 28, 2007
If your flying is limited to Part 91, you may not realize that operators who fly for hire, like Part 135 and Part 121 carriers, must follow different rules when it comes to flying approaches.
One of those is that a pilot flying an aircraft for hire cannot commence the final segment of an instrument approach into an airport if that airport does not have an approved weather reporting station (certain exceptions exist) or if the weather reported at the facility is below the minimum for the approach. So, if you were operating an aircraft under Part 135 and the minimum visibility for the approach you were planning on is 1/2-mile, it would be a violation to fly the approach if the airport's approved weather station is reporting a 1/4-mile visibility. This limitation does not exist under Part 91, where you could fly the approach all the way down and land if you determine at decision height (DH) or the missed approach point (MAP) that the visibility meets the approach's minimum requirements.
You would think that a professional pilot or flight crew would be more qualified to fly an approach to minimums than a private pilot who flies his own airplane occasionally on personal business or pleasure. But that is not the issue. The higher standards that are in place for those who charge for transporting passengers or cargo are there to protect the consumer. In other words, the FAA allows the average Part 91 operator to take more risks because there aren't any unwary paying passengers on board.
While the regulations on this matter for Part 135 and Part 121 are clear, one Caravan pilot strayed from the rules, with fatal results.
|Cessna 208 Caravan
On Dec. 1, 2001, at 1:43 a.m., a Cessna 208B Caravan crashed at the Bessemer, Ala., Airport (KEKY) while attempting to fly the ILS to Runway 5. The airplane was owned by a cargo company and was operating under Part 135 on a non-scheduled cargo flight from Little Rock, Ark., to Bessemer, using the call sign "Fast Check 600." The airplane had departed Little Rock a little less than two hours earlier.
The weather that morning was mostly clear across Arkansas and Mississippi, but there was patchy fog over Alabama that brought marginal VFR and IFR conditions to some locations. Just before the pilot departed Little Rock, the weather at Bessemer was reported as 500 feet overcast with 1/2-mile visibility. The temperature/dew point was 5 C and 2 C, respectively, and the wind was from 230 degrees at three kts.
A few hours earlier, the National Weather Service had amended the forecast for the Birmingham International Airport (KBHM) -- located about 17 miles northeast of Bessemer -- calling for scattered clouds and six miles in mist from the time of issue until midnight followed by clear skies and six miles in mist until 3:00 a.m. After that, the sky was to remain clear and the visibility was forecast to drop down to two miles.
The forecast for KBHM proved to be fairly accurate. A few minutes after the accident occurred, Birmingham was reporting a few clouds at 100 feet with six miles visibility in mist. Until that report the skies at Birmingham had been clear.
The same could not be said for Bessemer, however, where a dense patch of fog had settled over the airport, reducing the visibility to a quarter mile, sometimes less.
Before departure the pilot probably decided that he had a good alternate at Birmingham should the weather at Bessemer be below minimums upon arrival. Or perhaps he wasn't even thinking in those terms, based on the KBHM forecast and its proximity to KEKY.
The flight proceeded normally from Little Rock, cruising along uneventfully at 9,000 feet. At 1:22 a.m. the flight was transferred from a Memphis Center controller to the control tower at Birmingham. After some initial communications, the Caravan pilot asked the tower controller for vectors to the ILS at Bessemer. The controller responded, "Tell you what, Fast Check 600, I've got all this up in the tower cab now, and this scope is so small it's hard to vector to Bessemer with the lines they've got on it. Maintain 3,000 till Brookwood, cleared ILS Runway Five Approach to Bessemer."
The pilot acknowledged the instructions and at 1:37 a.m. he told the controller that he was intercepting the localizer. The controller cleared the airplane to leave the frequency with instructions that he was to report his cancellation on the ground at Bessemer via the clearance delivery frequency. The pilot acknowledged that transmission and was not heard from again.
Missed Gone Awry
|Bessemer, Ala., (KEKY) ILS Runway 5 approach chart. The chart in effect at the time of this accident was slightly different. (Click here for larger version - 110 KB.)
The airplane crashed 0.37 nm from the end of the runway nearly on the centerline. Investigators found that it was in a 24 degree left bank when it impacted trees about 30 feet above ground level. The descent angle from the trees to the ground was calculated to be 22 degrees.
There was no fire and rescuers noted a strong smell of Jet A at the site, indicating that there was fuel on the airplane. The pilot and his pilot-rated passenger, a fellow company pilot, were killed on impact. Investigators spent much time and effort looking at the airplane and its systems, yet they could find nothing that indicated there was a failure that could have caused or contributed to the accident.
A courier who was waiting for the airplane to arrive at the airport heard what he described as a tapping sound on the outside speaker that was tuned to the airport's CTAF. What he probably heard was the pilot turning the runway lights on or attempting to be certain that they were on and as bright as possible.
The courier, who worked for a regional bank, reported that at the time he was waiting for the airplane to arrive, the fog was the thickest he had seen it since he began coming to the airport a year earlier. He told investigators that he heard an engine sound and about two minutes later a noise that he associated with a heavy gauge shotgun. He said it was a sharp noise, followed by silence.
Another company pilot was on a flight from St. Petersburg, Fla., to KBHM, landing there about two minutes before the accident occurred. He told investigators that since he was scheduled to continue on to Bessemer, he checked the AWOS at KEKY when he was about 65 miles from Birmingham and again when he was about 30 miles out. The first time the visibility was less than 1/4 mile with a 100-foot indefinite ceiling. The second time it was 1/4 mile with a 100-foot indefinite ceiling. He said he flew a visual approach to Birmingham's Runway 36, and while he was inbound he observed widespread dense fog and noted he could not see the rotating beacon at Bessemer.
The Caravan pilot held a commercial pilot certificate with airplane single and multi-engine land and instrument ratings. He was 62 years old and had 5,773 total flying hours with 390 hours in the 90 days prior to the accident, 990 hours in the Caravan and an unknown amount of instrument time. He had been employed flying Cessna 206 and 210 aircraft since October 2000 and he had checked out in the Caravan in February 2001.
A review of the approach plate in use at the time of the accident reveals that the inbound heading on the ILS Runway 5 Approach was 050 degrees. An inbound pilot would maintain 2,400 feet until intercepting the glide slope just outside the MEATA Intersection, 6.1 miles from the Bessemer DME. DH for the approach was 900 feet, which was 200 feet above the touchdown zone and 1.1 miles from the Bessemer DME. The visibility requirement for the approach was 3/4 mile.
Radar data indicated that between 1:38:47 and 1:42:11 the airplane was flying on a northeasterly heading and descended from 2,400 feet MSL to 900 feet MSL. At 1:42:11 the airplane was located 0.43 nm from the approach end of Runway 5. The last radar hit on the airplane showed it at 1,000 feet, 0.2 nm from the approach end of the runway. It's not known how accurate the radar positioning is that low to the ground.
We don't know what happened in the cockpit of the Caravan that night that led to the accident, but there are several facts we do know. First, contrary to the FARs that governed the flight, the pilot elected to fly the approach when he should have known the visibility was below landing minimums. Of course, that doesn't mean the airplane should have crashed just because the pilot elected to fly the approach.
In fact, the evidence indicates that the accident may have happened as the result of his attempt to make a missed approach that resulted in the pilot's failure to maintain control of the aircraft. The radar plot shows the aircraft at minimums about where he should have been had he flown the glide slope properly. Then, the radar indicates the airplane started to climb. The post-crash analysis of the aircraft also shows that the flap actuator was found in a position that suggests the flaps were nearly fully retracted.
The missed approach procedure requires a pilot to climb straight ahead to 1,400 feet, then to commence a climbing right turn to 3,000 feet while flying direct to the Brookwood (OKW) VOR. Assuming the pilot began the climb, something must have happened to cause the loss of control. What could it be?
Once the decision to go around has been made there is no reason to look out the window any further to see if the runway comes into view. But many pilots do that anyway. Perhaps they figure that if they see the runway as they pass over it, that it might be worth another approach. But the missed approach requires that you reconfigure the airplane for a climb and concentrate on the instrumentation to insure that it climbs safely away from the earth's hard surface. If you are looking out the window at that point instead of at the instrument panel, it is very easy to become disoriented, even for a seasoned pilot.
Fog creates illusions, especially during nighttime, that hammer away at our senses and leave us disoriented. That may be what happened to the Caravan pilot. Or, he might have seen the runway as he began the missed approach and simply looked outside the airplane too long. Whatever happened, the airplane descended into the ground when it should have been safely climbing above it.
Know The Limits
How do you prevent a similar kind of accident? There is no point in beginning an approach when you are reasonably certain you have no chance of making it in. If you find that your destination airport is below minimums, go to your original alternate or find another close-in airport that has landing minimums or better.
If you fly under Part 91, you can begin an approach even if the airport is below minimums. But unless there is some reason to believe that it will be successful, you might as well save time and fuel and head for the alternate. In this instance, you can see that at Birmingham the airplanes were landing using visual approaches while at Bessemer even the ILS approach was not sufficient to allow the pilot to locate the runway.
Each pilot should have personal limitations that dictate the conditions under which he or she may begin an approach. Certainly, they should dictate that -- if the reported weather is below the minimums for the approach to be used -- the airplane should be flown on to an alternate.
Some will suggest that there is nothing wrong by trying an approach under those circumstances to see if the runway environment might be visible. However, there are too many things that can go wrong during any flight, such as an electrical or avionics failure, landing gear that won't extend properly, or an engine failure, to name a few. You are familiar with all of the potential problems that we train for. Why take a chance on encountering something like that when you know that there is virtually no way you will be able to make a safe landing? Should your engine fail under those conditions, you will be totally at the mercy of the terrain below, because you likely will not see the ground before the airplane strikes it.
Some pilots who fly under Part 135 ignore the rule that says they must have official weather and know that the airport is at or above minimums before they begin an approach. It seems that they feel this is one rule that the FAA looks the other way on. As long as nothing happens, that is.
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.