Jump to content

Approach Impossible


Recommended Posts

Not news I am sure to a lot of you but nevertheless may be of interest to younger crews.

Approach Impossible: 'Chair Flying' To Minimums Or Not At All

Apr 27, 2017James Albright | Business & Commercial Aviation

This article appears in the May 2017 issue of Business & Commercial Aviation with the title “Approach Impossible.”

Much of flying is an “act of faith.” You are placing your trust in those who designed and built the aircraft, in those who maintain it, and in those who trained you to defy gravity for a living. Your act of faith goes even further than you may realize, however.

Who, for example, ensures the instrument approach you are about to fly can be safely flown down to minimums without breaking anything? We assume the approach was designed correctly, tested in real world conditions, and has the seal of approval from the aviation authority of the host nation. In most cases, all of that is true.


A Jeppesen approach plate will often have the term “TERPS” or “PANS-OPS” printed on one side. In the first case, the approach was designed in accordance with the U.S. Standard for Terminal Instrument Procedures (TERPS), an FAA Order (currently numbered 8260.3C) in a constant state of revision. In the second case, the guidance came from the International Civil Aviation Organization (ICAO) Procedures for Air Navigation Services, Aircraft Operations (PANS-OPS), also known as ICAO Document 8168. With the TERPS or PANS-OPS “seal of approval,” you know the approach plate has been vetted. But in either case, can you assume the instrument approach is flyable down to minimums exactly as published?

Unfortunately, the answer is no. There are cases when the approach, while legal, is improbable because the terrain makes the required descent angles unsafe. Other approaches, while perfectly safe, are impractical due to airspace design or airport congestion. Finally, some approaches are impossible to fly because of poor design and will guarantee the need to execute a missed approach if attempted down to minimums. You can, however, discover these improbable, impractical and impossible approaches before leaving the ground. And that knowledge can help you come up with a “Plan B.”

The key is to “chair-fly” the approach by visualizing each step of the procedure while considering terrain, country-specific and other local restrictions, and aircraft descent and turning performance. In many cases advanced trigonometry is helpful but not required; a few basic math rules of thumb and a pocket calculator will suffice.

Mountainous Terrain — The Improbable Approach

If you’ve never flown into Eagle County, Colorado, Regional Airport (KEGE) and had only the publicly available instrument approaches available, you might think the published RNAV (GPS)-D weather minimums of 2,353 ft. and 3 sm would allow a comfortable and safe arrival with weather just above those figures. But that would be wrong thinking. There are special approach procedures requiring operator approval and specific training for EGE, but the RNAV (GPS)-D can be flown by any RNAV-capable aircraft and instrument rated pilot. Easy, right?


Pilots with at least one approach into this airport know that flying north of the procedure course down a valley in visual conditions is the better choice. They only begin the approach if they can spot the airport from waypoint POWRS at 12,000 ft. MSL, nearly 6,000 ft. above the runway. They have basically doubled the weather minimums. But what if you’ve never flown into Eagle and don’t know anyone who has?



Imagine yourself flying the approach in Instrument Meteorological Conditions (IMC) after successfully making the descent to 9,860 ft., your last step-down altitude prior to the missed approach point. While you began the day with a bit of concern, you breathed a sigh of relief when the ATIS reported the weather was 3,000 ft. and 4 sm. You still have a mile before you can leave the step-down altitude but start to make out what has to be the runway. You spot it! But then it hits you that even though the runway is 4 mi. away, you are still over 3,000 ft. above the landing surface. Too high! Now what? Can you circle? The surrounding Rocky Mountain terrain discourages that thought immediately. You have no choice but to go missed approach and think of a new way to get your passengers to their Vail ski chalet.


Thankfully, Garfield County Regional Airport in Rifle, Colorado (RIL) is just over 30 nm to the west and can fit you in on their crowded ramp. The FBO was out of rental cars and any available hangar space was already taken. As your passengers wait for their ground transportation to catch up, you are forced to revisit the decision-making that brought you to this point.

The Eagle County weather was well above minimums, but landing from the approach in that weather would have required a wildly unstable approach and been unsafe. It appears the Garfield ramp had already been consumed by other crews who knew better than to attempt an approach to Eagle with a 3,000-ft. ceiling and “only” 4 sm visibility. The more seasoned pilots unveil the Eagle County secret. “If I don’t see the runway before POWRS,” one pilot tells you, “I’m not descending any farther.” He goes on to tell you that even on a clear day, flying with the needles centered leaves you too high to land. “You have to fly down the valley to the north, otherwise you aren’t landing.” Well, now you know better! But how could you have known this without previous experience?

The key to flying an unfamiliar instrument approach correctly the first time is to mentally put yourself on the approach before you have to do it for real. You can do this from your dining room table, hence the seasoned veteran’s technique of “chair-flying,” but you need to be methodical about it. You need to think about the airplane’s ability to descend and turn along each segment of the approach.


Looking back at our RNAV (GPS)-D approach into Eagle County we understand immediately that the terrain imposes descent restrictions until at least the 9,860-ft. step-down altitude located 3.5 mi. from the runway. If the weather was good enough to spot the runway from this distance, what kind of descent rate is needed? We need to descend 9,860 – 6,547 = 3,313 feet, or 33 hundreds of feet. Using our descent rule of thumb, we find that our required descent rate will be 33 ÷ 3.5 = 9 deg. Under TERPS, the maximum glidepath angle for a precision approach is 3.1 deg. for Category D and E aircraft, 3.6 deg. for a Category C aircraft and 4.2 deg. for a Category B aircraft. While those numbers don’t restrict how you fly this non-precision approach, they offer you a good idea of what can be done safely. The 9-deg. descent angle is simply too steep.

Our chair-flying exercise reveals that flying this approach with the needles centered leaves you too high to make a stable approach to landing from instrument minimums. The terrain depiction on the instrument chart reveals a valley to the north of the approach course that would allow you to descend earlier and provides the added benefit of lengthening your flight path to give you more distance to descend. But will it be enough?

The first time I flew into Eagle, I took a paper terrain map and plotted a hypothetical ground track to determine the distance flown. These days there are free internet applications that can automate the process. Using a terrain mapping application such as Google Earth shows the valley route from waypoint POWRS to the runway is 16 nm long. Beginning our descent from POWRS means we have to lose 12,000 – 6,547 = 5,453 ft., just over 54 hundreds of feet. That reduces the required descent gradient to 54 ÷ 16 = 3.4 deg.

The terrain at many airports in mountainous areas makes landing from instrument approaches improbable because the required descent rates are too high while remaining precisely on course. Other approaches can be impractical because of national rules, air traffic density or other unusual circumstances.

Unusual Circumstances — The Impractical Approach

ICAO course reversal entry procedures are different than U.S. procedure turn entry rules and the difference can get you in trouble. The international procedures do a better job of ensuring you begin the approach on course but often require extra maneuvering prior to starting the approach.

Some airports can compound this confusion with local procedures needed to deal with high-density traffic. These local procedures are rarely published where a visiting international pilot can be forewarned.

The ILS or LOC Rwy 27 to Schiphol Airport (EHAM) in Amsterdam provides a classic example. Under PANS-OPS this type of course reversal is known as a base turn and must be begun from a specific entry sector. The entry sector is generally within 30 deg. of the outbound course. If outside the entry sector, the holding pattern must be used to get within that sector before starting the approach.

Our Schiphol example approach has two initial approach fixes for aircraft arriving from the west and one from the east. Only pilots entering from SUGOL are permitted to immediately begin the outbound segment of the approach. Pilots arriving from ARTIP and RIVER are expected to execute a turn in holding at the Schiphol VOR.


About a year ago I was arriving from the west and got the clearance, “cleared ARTIP ILS Runway 27.” Under U.S. procedures I could fly from ARTIP to SPL and then turn left to intercept the 116-deg. outbound radial. This would have earned me a violation under ICAO rules.

Because we chair-flew the arrivals into Schiphol as a crew, we were fully prepared to deal with having to reverse course twice. This “double” course reversal hardly makes sense for one of the world’s busiest airports, but these are the rules as published under ICAO PANS-OPS. If we had lost communications or air traffic control had lost radar, we would expect to fly the arrival precisely this way. But we knew it couldn’t end up this way since Schiphol is far too busy. Our chair-flying exercise included other options to arrive at each runway. There was also a VOR approach to Runway 27, though it is hardly anyone’s first choice of a procedure to use in actual instrument conditions.


The Jeppesen airport arrival briefing pages spelled out the lost communications scenario that included the double course reversal. But those pages also noted, “navigation in the initial and intermediate approach segment is primarily based on radar vectors by ATC.”

As we neared the airport our first clearance was “cleared ARTIP, ILS Runway 27.” We realized our hypothetical double course reversal was really possible but suspected a vector might shorten things considerably, so we began configuring early. Shortly after passing ARTIP we got a new clearance, “Direct Papa Alpha Mike, cleared the ILS Runway 27.”

Now we could have had a new problem: Where is Papa Alpha Mike? Fortunately, we had also reviewed the VOR Runway 27 approach, which is flown off of the PAM VOR.

There is no doubt the ICAO double course reversal can be impractical at times, but it also serves to remind us that many U.S. procedures are exceptions to ICAO PANS-OPS. We need to know the rules of the host country and keep a level of situational awareness to make an impractical approach usable.

Sometimes an approach can seem straightforward and quite practical, but a simple design error will make landing at the published minimums impossible.

Poor Design — The Impossible Approach

Approaches with specific tracks to fly can seem deceptively easy: You just need to follow the heavy black line. But these approaches can be built for the approach designer’s convenience, not the pilot’s. Chair-flying these approaches ahead of time can reveal minimums that are set too low.

The NDB Runway 07 into E. T. Joshua Airport, Kingstown, St. Vincent (TVSV) looks straightforward at first glance. You pick up a 283 deg. course for 3.5 min., turn left, and then turn left again when on runway centerline. The MDA is at 1,500 ft. and the minimums are 3,200 meters, about 2 sm (1.73 nm).

I first flew this approach in a Gulfstream V with a ragged ceiling between 1,500 and 2,000 ft. but good visibility outside the clouds. Our approach speed was just under 120 kt. and we planned on flying the entire procedure fully configured at that speed. We didn’t spot the runway until right on an extended centerline and by then we were too high to land. Fortunately, on the second try, the ragged ceiling allowed us to spot the runway earlier and descend comfortably to land. Our postflight critique began with one thought: “Why were we too high on the first try?”

Had we chair-flown the approach ahead of time, we would have realized landing at minimums would have been impossible. The distance needed to descend from the 1,500-ft. MDA to the near sea level runway exceeded the distance available along the 063-deg. extended runway centerline or within the distance of the visibility minimum. But you cannot predict your distance from the runway on that extended centerline without knowing your aircraft’s turn radius.


Since we flew the entire procedure at 120 kt., we were doing 2 nm per minute. (120 nm per hour divided by 60 min. in an hour.) That gave us a turn radius of 0.6 nm. Doubling that gives us our turn diameter and the answer to the question, how far south of the runway is the 103-deg. course? Answer: 1.3 nm.

But we will be flying the diagonal 063-deg. line, which gives us more distance to descend. But how much more distance? At this point, we have two options on determining the distance: Armed with our turn radius, we can plot our ground track on the approach chart or we can do the same mathematically.

The heavy black line on the approach plate may or may not be an accurate representation of the aircraft’s actual ground track, depending on the aircraft’s speed and environmental conditions. We can construct our own hand-drawn ruler by transferring the scale on the left of the Jeppesen chart onto the edge of an index card or other straight-edged paper. Using this makeshift ruler, we discover that the heavy black line traces an eastbound course that is about 1.5 nm south of the westbound course. Because we know our turn diameter will be 1.3 nm, we know our aircraft will actually fly inside the depicted track but will be close.


We then measure the distance from the eastbound track to the runway and see we will have less than 2.5 nm, because we will be inside the depicted course.

With a little knowledge about right triangles and a scientific calculator, we can find the distance between our turn to final and the runway more precisely. Instrument approaches are often made up of straight lines and semicircles that can be further broken down to a series of triangles. In the case of our Joshua NDB approach, the distance to descend along the 063-deg. course line is the hypotenuse of a right triangle for which we know the smallest angle because we turn left from 103 to 063 deg., a difference of 40 deg.



Our right triangle lengths decoder tells us the length of the (c) leg is equal to the length of the (a) leg divided by the sine of the angle (A). A calculator makes quick work of this: c = 1.33 ÷ sine (35) = 2.1 nm.

Whether you use the hand-drawn ruler or a scientific calculator, the chair-flying exercise reveals that we have less than 2.5 nm to descend 1,500 ft. Our earlier rule of thumb tells us that this will require a 15 ÷ 2.5 = 6-deg. descent rate. No wonder we were too high to land!

So the next question would be how much distance do you need to make that descent? Remembering that a 3-deg. glidepath takes 318 ft. per nm, our answer is 1,500 ÷ 318 = 4.7 nm. In terms of visibility, that equates to
5.4 sm.

Now we know the approach minimum of 3,200 meters (2 sm) does not provide enough distance to descend in a safe, stabilized manner. We had future trips to St. Vincent and realized we would need Visual Meteorological Conditions (VMC) to safely land.



An Instrument Approach Chair-Flown at 0 Kt.

If an instrument approach looks unusual at first glance, it will be worth a second or third examination. But analyzing an unusual instrument approach just minutes prior to beginning your descent doesn’t leave you a lot of time to consider if the approach is improbable, impractical or perhaps impossible. Chair-flying the approach before you leave the ground gives you the time to come up with other options, including not going in the first place.

The only real math skill needed is knowing how much airspace your airplane needs to turn. With a few rules of thumb, an approach plate drawn to scale and a sharp pencil, you can accurately predict your flight path and find out if you are looking at an impossible approach before you are committed to flying it. 


Link to comment
Share on other sites

  • 1 month later...

Poor Decisions: Unstabilized Approach, Vicious Winds Trigger Crash

May 24, 2017Richard N. Aarons | Business & Commercial Aviation

This article appears in the June 2017 issue of Business & Commercial Aviation with the title “Poor Decisions.”

A highly experienced Mexican crew with newly minted Bombardier Challenger type ratings attempted an approach into one of business aviation’s toughest airports with adverse winds on Jan. 5, 2014. The result was tragic. This month we’ll look at the NTSB’s investigation into the accident and the lessons it presents.


The CL-600-2B16 Challenger, N115WF, impacted the runway — nosewheel first — at 1222 MST during an attempt to land on Runway 15 at Colorado’s Aspen-Pitkin County Airport/Sardy Field (ASE). The copilot was fatally injured; the captain and the pilot-passenger in the cockpit jump-seat suffered serious injuries. The airplane, operated by Vineland Corporation Co. of Panama, was destroyed. The IFR Part 91 flight originated from Tucson International Airport (TUS) at 1004.


The departure and en route portions of the flight were uneventful, according to the safety board. Weather was VMC for the entire route; however, Aspen winds were strong and gusting, causing some inbound flights to go to alternate destinations.

As the accident flight neared Aspen, air traffic control (ATC) provided vectors to the LOC/DME-E approach to Runway 15. At 1210:04, the Aspen local controller informed the flight crew that the wind was from 290 deg. at 19 kt., with gusts to 25 kt. At 1211:18, the crew reported that they were executing a missed approach and then requested vectors for a second attempt. ATC vectored the airplane for a second LOC/DME-E approach. At 1220:35, the local controller informed the flight crew that the wind was from 330 deg. at 16 kt. and the 1-min. average wind was from 320 deg. at 14 kt., gusting to 25 kt. He then cleared the flight to land. (See “CVR Recording For Challenger N115WF” for the CVR record.)

Airport surveillance video of the runway, analyzed by the safety board, showed the following sequence of events during the incident:

  • The airplane above the runway in a slightly nose-down attitude.
  • A flash of light consistent with a runway strike.
  • The airplane in the air above the runway in a nose-down attitude.
  • The airplane impacting the runway in a nose-down attitude and being engulfed in light. About 4 sec. elapsed between the runway strike and the final impact.

The Challenger came to a stop inverted on the west side of Runway 15, halfway between Taxiways A5 and A6. The ASE aircraft rescue and firefighting (ARFF) station was located on the west side of Runway 15, about 0.3 mi. north of the accident site. ARFF personnel witnessed the crash and responded immediately, requesting clearance onto the runway about 50 sec. after the accident occurred. The rescuers extricated the crew and pilot-passenger and contained the fire. However, the copilot, seated in the right seat, died of blunt force trauma due to crushing of the right cockpit area.

Investigators found inverted jet bore visible fire damage on the fuselage and wings but no evidence of fire inside the cockpit and cabin. The airplane’s underbelly had markings consistent with ground scraping. The right upper cockpit structure was partially collapsed and structurally breached. The right wing was folded beginning about one-third of the wingspan from fuselage. The left wing was bent downward (relative to the airplane’s resting position) outboard of the outboard flap. The upper section of the vertical stabilizer, including the horizontal stabilizer, was detached from the main hull.


Both main landing gear were found in the extended position and connected only by their side stay actuators. The nose landing gear wheel-well structure was deflected upward, but did not contact any flight control beam assemblies under the cockpit floor. The nose landing gear was folded about 70 deg. aft and 30 deg. to the left. The right axle for the nose landing gear was severed, and the right nosewheel tire was missing. The left nosewheel was missing a portion of the inboard hub rim. The nose landing gear lower oleo strut had markings consistent with ground scraping on the axle jack point. A portion of the nose landing gear axle fracture surface had markings consistent with ground scraping. The flaps were found in a deployed position. Investigators counted the exposed actuator threads (26 were found) and determined the flaps had been at 45 deg.

The horizontal stabilizer trim actuator jackscrew was examined and found intact. The measurement from the gearbox upper surface to the upper gimbal lower surface was 4.85 in., indicating a trim setting of about 4.85 (trim indicator range is from 0 to 9; 0 is full nose-down and 9 is full nose-up; a measurement of 7.59 in. equates to about full nose-down).

The left and right main angle of attack (AOA) vanes were found intact with no visible damage. Both vanes moved normally when a finger force was applied. The right aux AOA vane had no visible damage; however, it did not move when a finger force was applied. The left and right engines showed no indications of pre-impact malfunction.

A visual inspection of the cockpit found the following: 

  • The flap handle was at 45 deg.
  • The engine power levers were in the SHUT OFF position.
  • The engine reverse thrust levers were in the STOWED position.
  • The landing gear handle was in the DOWN position.
  • The flight spoiler handle was in the RETRACT position.
  • The ground spoiler switch was in the ON position.
  • The right control column was bent to the left about 20 deg.
  • The left and right control yokes were deflected to the right about 20 deg. and appeared to be synched.
  • The pitch and roll disconnect handles were in their normally stowed position.
  • The enhanced ground proximity warning system (EGPWS) PBAs were in their normal out positions.
  • The pilot and copilot stall protection pusher switches were in the ON position.
  • The air-driven generator was in the DEPLOYED position.

The 52-year-old captain was a citizen of Mexico. He held a Mexican ATP certificate that included an Airbus A320 type rating and ratings for airplane multiengine land and instrument airplane. He also held an FAA temporary commercial airman certificate issued on Nov. 9, 2013. That certificate included a CL-600 type rating and ratings for airplane single-engine land, airplane multiengine land and instrument airplane.


After the accident, the FAA reviewed the certificate and determined that a limitation on the pilot acting as pilot in command (PIC) for the CL-600 should have been included; however, the limitation had been overlooked by the designated pilot examiner who issued the certificate. The limitation would have restricted the captain from serving as PIC in the Challenger with revenue passengers on board until he had acquired 25 hr. of actual flight time in the CL-600 with another qualified pilot.

The safety board said the captain had completed Challenger type-rating training on Nov. 8, 2013, at SimuFlite in Dallas, and that his check ride was “satisfactory.” The captain told investigators that he had no trouble during that training other than with use of the FMS. His flight experience in the Challenger at the time of the accident consisted of a ferry flight from Dallas to Toluca, Mexico, and a roundtrip flight from Toluca to Eagle County Airport, Colorado. He stated that his total flight time in the Challenger was 12 to 14 hr., which included his flight training at SimuFlite.

However, he reported having logged 8,000 hr. flying the Airbus A318, 319 and 320 before flying the Challenger and had about 17,000 hr. of total flight time. The Airbus time reported was completed under his Mexican flight certificate and did not transfer to his FAA-issued certificate.

The copilot, age 54, also was a citizen of Mexico. He held a Mexican ATP that included an Airbus A320 type rating and ratings for airplane multiengine land and instrument airplane. The copilot also held an FAA temporary airman certificate issued on Nov. 14, 2013. The temporary certificate included a CL-600 type rating and ratings for airplane single-engine land, airplane multiengine land and instrument airplane. The certificate was subject to a limitation for the CL-600, which restricted him from serving as PIC with revenue passengers on board until he had accrued 25 hr. of actual flight time in the type with another qualified pilot.

A limited first-class medical certificate was issued to the copilot on Dec. 13, 2012. At the time of the accident, 13 months after the time of issuance, the copilot’s medical certificate would have been equivalent to a third-class certificate. The copilot reported on his most-recent medical certificate application that he had accumulated 20,398 total flight hours, with 31 hr. in the previous six months. His logbook was not located during the investigation.

On Nov. 9, 2013, the copilot completed training for the CL-600 type rating at SimuFlite in Dallas. The safety board said the copilot received an “unsatisfactory” rating at the completion of the training check ride. He had failed to satisfactorily complete two tasks under the “missed approach” approach skills, including “from a non-precision approach” and “engine out.” On Nov. 14, 2013, he was retested and received a “satisfactory” rating at the completion of the second training check ride.

The safety board also took a close look at the 52-year-old pilot-passenger who rode the jump seat and was contributing to the cockpit decision-making, according to the CVR. He, too, was a Mexican citizen and held an FAA, foreign-based commercial pilot certificate with airplane single-engine land and airplane multiengine land ratings. No type rating for the CL-601 was included on the FAA-issued commercial certificate. The FAA certificate was not valid for the carriage of persons or property for compensation or hire. The accident captain told investigators that the passenger was his and the copilot’s friend. He added that the passenger was an experienced Challenger pilot and was invited to join them on the trip to “provide any recommendations” because of the “special conditions” at Aspen.

What Happened?

The safety board analyzed the approach in phases. It said the initial part of the airplane’s second approach was as expected for descent angle, flap setting and spoilers. However, during the final minute of flight, the engines were advanced and retarded five times, and the airplane’s airspeed varied between 135 kt. and 150 kt.

“The final portion of the approach to the runway was not consistent with a stabilized approach,” it reported. “The airplane stayed nose down during its final descent and initial contact with the runway. The vertical acceleration and pitch parameters were consistent with the airplane pitch oscillating above the runway for a number of seconds before a hard runway contact, a gain in altitude and a final impact into the runway at about 6 G.”

The weather at the time of the accident “was near or in exceedance of the airplane’s maximum tailwind and crosswind components for landing,  as published in the AFM,” said investigators.


Given the location of the airplane over the runway when the approach became unstabilized and the terrain limitations of ASE, performance calculations were completed to determine if the airplane could have successfully performed a go-around. “Assuming the crew had control of the airplane, and that the engines were advanced to the  appropriate climb setting, anti-ice was off and tailwinds were less than a sustained 25 kt., the airplane had the capability to complete a go-around, clearing the local obstacles along that path,” the investigators concluded.

The safety board also considered the pilots’ time in type. Neither flight crewmember would have met the minimum flight time requirement of 25 hr. to act as PIC under FAR Part 135, but since the accident flight was conducted under Part 91, the 25-hr. requirement did not apply to this portion of their trip.

“Nevertheless,” said the safety board, “the additional flight time would have increased the crew’s familiarity with the airplane and its limitations and likely improved their decision-making during the unstabilized approach. Further, the captain stated that he asked the passenger, an experienced CL-600-rated pilot, to accompany them on the trip to provide guidance during the approach to the destination airport. However, because the CL-600-rated pilot was in the jump-seat position and unable to reach the aircraft controls, he was unable to act as a qualified pilot-in-command.”

The safety board determined the probable cause(s) of this accident to be “the flight crew’s failure to maintain airplane control during landing following an unstabilized approach. Contributing to the accident were the flight crew’s decision to land with a tailwind above the airplane’s operating limitations and their failure not to conduct a go-around when the approach became unstabilized.”

Much of the story of this accident can be understood, I think, by a review of the CVR record in the accompanying sidebar. The reader gets the impression that the pilots at the controls were uncomfortable with the approach and never really on top of the situation. Look at the CVR information and see what you think. 

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

  • Create New...