Safe Landings: Windshear

The windshear saga in American aviation history reveals a complex and costly past. Windshear has existed for as long as aviators have taken to the skies and is largely responsible for several classic aviation losses. Notable U.S. aviation accidents include Eastern Flight 66 (1975), Pan American Flight 759 (1982), and Delta Flight 191 (1985).

Windshear remained unrecognized for years. It was not clearly understood until swept wing, jet aircraft encountered the phenomenon. Since 1975, windshear has been researched and studied, measured, defined, catalogued, and rightly vilified. Technology has been developed to identify and minimize the threats that it poses. Procedures have been implemented to aid pilots who experience windshear in flight and flight crews invest hours of simulator training practicing windshear escape maneuvers.

Even with progress to date, windshear continues to be a worthy adversary to aviation professionals. It requires respect and wisdom to defeat. Pilots often must make decisions regarding known or anticipated windshear, and the best practice is always avoidance.

This month, CALLBACK shares reported incidents that reveal some means and extremes of windshear experienced in modern aviation. Lessons to be gleaned are ripe, rich, and many.

Teasing a Toronto Tailwind

After encountering windshear that resulted in an unstabilized approach, this A319 Captain elected to continue to a landing. He noted his awareness of the current winds and trends as well as his personal preparedness to go around as reasons for continuing the approach.

After being delayed due to low ceilings in Toronto, we were finally descending…in heavy rain and moderate turbulence with clearance to 7,000 feet MSL. After a third 360 degree turn, we were…transferred to the Final Controller and proceeded inbound for the ILS RWY 05. The last several ATIS [reports] showed winds at approximately 090 to 100 [degrees] at 5 to 10 knots, and the Final Controller mentioned the same with an RVR of 6,000 plus feet for Runway 05. When cleared for the approach, we were at 3,000 feet MSL to intercept the glideslope, and I noticed the winds had picked up to a 50 knot direct tailwind. The First Officer was flying. We were assigned 160 knots and began to configure at approximately 2,000 feet AGL. At 1,500 feet the wind was a 30 knot direct tailwind and we had flaps 3. Indicated airspeed (IAS) had increased at this point [with] thrust at idle to 170-175 knots, prohibiting final flaps just yet. The First Officer did a great job aggressively trying to slow the aircraft, as we were concerned about getting a flaps 3 overspeed. As I knew from the ATIS and the Controllers (Tower now), the winds were to die off very soon to less than 10 knots. [Below] 1,000 feet we were just getting the airspeed to put in final flaps (full) and were finally stabilized and on speed between 500 to 800 feet. The winds were now at the reported 090 [degrees] at 8 knots or so [below] 500 feet. The total wind shift was approximately 90 degrees from direct tailwind to a right crosswind - losing 40 knots [of tailwind] in the space of 1,500 feet or so. The reasons I elected to continue the approach were:

1. We landed uneventfully in the touchdown zone and on speed…after breaking out before minimums.

2.[I saw] a positive trend in the wind.

3.  [I was] prepared for the missed approach (at 500 feet) IF the winds and IAS stayed as they were earlier in the approach.

We landed uneventfully in the touchdown zone and on speed…after breaking out before minimums.

Up and Down into Salt Lake City

While being vectored for an approach, this light twin transport Pilot encountered a vertical windshear that dramatically demonstrated the intensity, danger, and potential traffic conflict that a challenging vertical shear can present.

We had lined up for the ILS RWY 3 at Ogden, but at glideslope intercept, the weather had [deteriorated] to . mile visibility and a 400 foot ceiling. We broke off the approach,… requested an approach to land at Salt Lake City, and were vectored to the ILS RWY 34L. Approximately 10 miles downwind in solid IMC [with the] autopilot and altitude hold on and about to turn base, we hit a downdraft that dropped us approximately 2,000 feet. The horizon ball was all brown, the autopilot and altitude [hold function] were ineffective,

the loss of control set off the master warning system due to lack of fuel (at the time we had 750 pounds per side), and the terrain warning went off. Recovery was accomplished, but with a 2,000 foot gain (assigned altitude [had been] 10,000 feet; at the floor of the incident [the altitude was] approximately 8,000 feet; at the ceiling of the incident [the altitude was] approximately 12,000 feet). I was then routed back to the west and north on vectors for sequencing back to the ILS RWY 34L at Salt Lake City that was shot with a side‑step on final in VFR conditions to Runway 34R.

Shearing Situational Awareness

This Air Carrier Captain accomplished a successful windshear recovery while on final approach. He was surprised by the quickly changing environment and challenged by his diminished awareness as a result.

We were on final for Runway 8R in Houston and encountered windshear.… Tower started calling an approach wind loss of 20 knots that increased to 25 knots at a 3 mile final. The Copilot and I were discussing what constituted a microburst alert, which was 30 knots, so we elected to continue the approach. We were in moderate turbulence and the wind was currently a right quartering tailwind which would switch to a left crosswind on the runway. I asked the Copilot to increase our target speed to plus 20, which he did, and as we approached the outer marker, we were fully configured and on speed. At approximately 1,400 feet AGL, we received a “MONITOR RADAR DISPLAY.” I saw that the indication was ahead of us to the right of our course. Since we were still stable and fully configured [with the] autopilot and autothrottles on, we elected to continue.

Shortly we received the call, “GO AROUND, WINDSHEAR AHEAD.” I initiated the go-around and asked for flaps 15 and gear up. Very shortly after this, we received the call, “WINDSHEAR, WINDSHEAR, WINDSHEAR.” At that point I pushed the throttles to the stops, verified the spoilers were stowed, and selected Takeoff Go-Around (TOGA) again. The First Officer called ATC and said we were going around. I was so focused on flying the plane with regards to Radio Altimeter (RA) and trend, and verifying I was doing everything correctly, I did not hear what ATC replied back to us. Adding to the workload and task saturation was the plane on Runway 8L, which also went around, and then the two planes behind us on Runways 8L and 8R also went around.

The Copilot advised that ATC said to level off at 2,000 feet as we were passing through 2,000 feet with a high climb rate. I still had “WINDSHEAR” displayed on my ADI, and I told him I was not going to level off. He then had to try to talk to ATC again to get a new altitude. They gave us 3,000 feet. We were climbing rapidly, and I brought the throttles back to level off at 3,000 feet, but overshot it to approximately 3,200 feet and descended back to 3,000 feet. The landing gear horn immediately began to sound when I pulled the power back since we still had flaps 15. I made sure we were above flaps 15 retraction speed, and we completed a normal go-around at that point to clean maneuvering speed.

Everything happened so fast. ATC should not give a level off altitude of 2,000 feet since I now know it is possible to still be in windshear…at that altitude. If I were to fly this approach again, I would elect to abort the approach and wait for tower to stop calling a 20-25 knot loss at a 3 mile final.… We thought that since the planes ahead of us were landing, we would be able to [as well]. Obviously there is always a first flight that cannot land, and on this day, that was us.

The Final Authority – 14 CFR 91.3

This heavy transport Captain perceived a subtle suggestion to take off when weather that may have presented a windshear hazard was nearby. He exercised his authority with seasoned wisdom and sound judgment when he opted not to leverage the safety of his aircraft or crew.

As we were taxiing west on Runway 27, we could see a radar return of a strong storm which was depicted red on our screen. The storm was directly west of the…airport and appeared to be moving east toward us. As we turned south on Taxiway N, we could only see part of the storm to our right on the radar display. When we switched frequencies to Tower, we heard that there was windshear on a two mile final for our runway. As we approached the runway, we advised Tower that we would not take off. Tower reminded us that the windshear was two miles in the opposite direction from where we would be heading. It seemed like the cell was directly over the field at that time, possibly centered a little north.… The FOM guides us not to get within 5 miles of a cell below FL200. Tower instructed us to taxi out of the way so that several other aircraft could take off while we waited a few minutes for the storm to pass.

I feel that Tower was more concerned about getting airplanes on their way than waiting a few minutes until it was safe. I also think [there is an] air carrier culture pressure to get the job done even if there is an increased risk.

When one aircraft decides it is not safe to take off, perhaps Tower should inform the following aircraft that might not have been on frequency to get the same information. Although several aircraft took off away from the storm, they faced the possibility of getting a decreasing performance


Safe Landings: Controller Pilot Data Link Communications 

Controller Pilot Data Link Communication Departure Clearance Services (CPDLC-DCL) is one segment of the Future Air Navigation System (FANS) that has been recently implemented in the contiguous 48 states at local Tower Data Link Service (TDLS) equipped facilities to deliver departure clearances and revised departure clearances prior to takeoff.

As any new system is implemented, some “bugs” may be expected, and CPDLC-DCL is no exception. ASRS is receiving reports suggesting that crews are experiencing problems while using CPDLC-DCL for its intended objective. The problems that are experienced point to sources from system architecture, to precise meanings of specific words and formats used in the CPDLC-DCL syntax, to basic interpretation and understanding of the CPDLC-DCL information protocols and operational procedures.

This month, CALLBACK shares reported incidents of complications that arose from the crews’ use of CPDLC-DCL to obtain departure clearances and revised departure clearances. While CPDLC-DCL offers many improvements and advantages over voice and Pre-Departure Clearance (PDC), some issues remain as we transition to this new system. As these examples may hint, ideas will emanate from the cockpit and formal solutions will be devised.

Cautious Pilot Distrusts Link


This Air Carrier Crew clarified an initial question they had about a revised departure clearance. Curiosity over the revised SID and transition that had not been “properly” LOADED resulted in a route portion that was manually loaded but not included in the clearance. 

During preflight, we received a revised clearance via CPDLC. The change was from the TRALR6.DVC to the STAAV6.DVC. I verified [the] clearance and received a full-route clearance over the radio. When the LOAD feature was selected in CPDLC, the new revised route did not LOAD into the ROUTE page properly. It still showed [the] TRALR6.DVC, but now it had a discontinuity. At this point, I had to load the route manually. When I did load the STAAV SIX, however, I failed to select the DVC transition, [so the FMC] now had point STAAV direct to LAA in the LEGS page. When we did the route verification later, during the preflight, we both failed to detect the missing transition that included the points TRALR, NICLE, and DVC.

This went unnoticed until passing point STAAV on the departure. That is when ATC queried us if we were headed to point TRALR. We indicated to ATC that we were direct LAA. He re-cleared us to TRALR to resume the departure. There was nothing significant to report for the rest of the flight. 

Complications Perceiving Data Link  


After using CPDLC to obtain their clearance, this Air Carrier Crew saw no indications that their clearance had been revised. It appeared the same as the filed route, so they did not LOAD it. ATC soon called them off course.

Prior to departing SNA, we loaded the FMC using normal procedures… We received a ready prompt but did not get a clearance. Shortly before push, we still had not received a CPDLC clearance, so we requested a clearance via PDC. We got a PDC reply message stating to use CPDLC, and simultaneously a clearance was uploaded to the FMC ATC page. The clearance showed our departure and transition as filed, as well as the altitude restrictions, expected altitude, squawk, and departure frequencies as we expected to see. The ATC page did not state that it was a revised clearance or route. All obvious indications were [that] our clearance was unchanged from the filed route. A LOAD prompt and ACCEPT prompt were shown. We ACCEPTED the clearance, but because we had already loaded the flight plan, we did not LOAD the flight plan sent via CPDLC.

We departed as normal. Once airborne passing 10,000 feet, we received an ACARS flight progress printout that showed our originally filed course. After passing TRM, ATC stated they showed us off course. They gave us a revised route clearance. There was no printout of our clearance to reference, and since the CPDLC did not display the full-route clearance, we had difficulty tracking down whether or not there was actually a change to our originally filed route. We were able to find a LOAD prompt on page three of the ATC clearance page. When we selected LOAD, a new route was LOADED to the FMC, which was different from our originally filed route. We discovered our route had, in fact, been changed via CPDLC. We erased the change since we already had a new route assigned by ATC airborne and continued uneventfully to [our destination].

Complex Presentation — Desired

Learning Cumbersome

Non-intuitive wording and convoluted clearance procedures coupled with incomplete systems knowledge caused this Crew to misinterpret the message that their clearance had been revised. The result was another call off course.

We received our departure clearance via CPDLC. During my preflight, I loaded the FMC with the route we were given on our release paperwork (PORTT THREE DEPARTURE). Our release had the following FMS route: KEWR BIGGY PTW J48 BYRDD J230 SAAME STEVY HVQ UNCKL MAUDD4 KSDF. When it came time to log on to the CPDLC, the Captain and I received the following [departure clearance]: CLEARED ROUTE CLEARANCE ORIGIN: KEWR DEST:KSDF ARRIVAL: DARBY 5.UNCKL +LOAD NEW RTE TO KSDF+ EWR2, CLB VIA SID EXC MAINT 2500FT EXPECT FL340 10 MIN AFT DP, DPFRQ 119.2 SQUAWK 1534, CTC GROUND 121.8 FOR TAXI.

When we saw that this was a change, I manually changed the SID to the NEWARK TWO off of Runway 22R. I did not select the LOAD prompt, as I did not see any other change to our clearance. I believed our clearance was now the NEWARK TWO (flown in LNAV), and our first fix was still BIGGY then PTW, etc. The Captain and I agreed on this. I printed the CPDLC clearance, folded it, and laid it on the center console.

After takeoff, we were handed off to Departure Control. He cleared us to fly directly to a fix, which neither the Captain nor I saw on our route. He said, “Don’t you have the PARKE J6” on your routing? We said, “Negative, we have BIGGY PTW J48.” He then told us to fly a heading. He came back a short time after that and told us to fly directly to LRP and join J6 and expect a further clearance later. He did come back shortly thereafter, and told us to fly J6 to UNCKL, then the arrival.

We obviously knew we had been expecting something different than the Controller had been, so I picked up the “printed” copy of the CPDLC clearance we had received, and on it was also the following: PARKE J6 UNCKL Note that this was NOT on the original CPDLC clearance we looked at on the FMC. I don’t know if it had been truncated off due to space, or it had been inadvertently left off or what, but we both went back and looked and noticed this. That was why we thought the only change to our clearance was to the NEWARK TWO, [while keeping] the same fixes as we originally had on our paperwork (i.e. BIGGY PTW J48).

Contributing factors to this confusing situation are numerous. I now know that I am supposed to select the LOAD prompt when we have a change to our routing. The method we are transitioning into with respect to getting our clearances via CPDLC is very confusing. The wording… on the FMC is not intuitive, and the overall procedure… is very convoluted. We now receive our clearances in several different ways, at different airports, and in different airplanes, [which] all lead to a higher chance of mistakes.

Complicated Process Destines LoweredCognizance

This CPDLC message arrived at precisely the wrong time. The Crew’s attention was diverted, and their situational awareness suffered as they attempted to eliminate the confusion generated when they could not quickly resolve the revised clearance CPDLC message.

We were taxiing out of a very congested, weather-impacted, JFK airport that day. The ground frequency was non-stop due to long taxi routes because of 20-mile in-trail spacing for departing aircraft in all directions. A CPDLC message [reading] “THIS IS A REVISED CLEARANCE” appeared with no other information. No revised route [was] included. [We] spent [the] next five to 10 minutes heads down, while taxiing, trying to figure out what was happening, in addition to eventually calling Clearance Delivery on the radio, and Dispatch. [There was] no place to pull out of [the] way due to long taxiways with no exits. And, we were getting automatic ACARS messages [that we] must be airborne in 15 minutes due to [the] nine-hour flight time restriction. A simple printout of the revised clearance would have resolved the issue in a few moments and would have been much more intuitive.

The current system of having an ATC clearance, current or revised, stored on multiple, disjointed pages of the ACARS or FMS display is confusing and causes excessive heads-down time while taxiing. It will cause a gross navigational error, is a defective system, and is going to harm someone.

Common Precautions Demystify Link


This Air Carrier Crew noticed an ambiguity in their departure clearance. Although the syntax was confusing, Clearance “cleared up” their misunderstanding.

The First Officer loaded [the filed route] into the FMC before requesting a CPDLC clearance. The clearance came back, “CLEARED ROUTE CLEARANCE. FREE TEXT. POM9.GMN. FREE TEXT CLB VIA SID EXC MAINT 14,000FT.” The First Officer noticed a LOAD prompt, and [saw that] the new route [read], “DIRECT GMN DIRECT RGOOD RGOOD.EMZOH3.SKIZM.” Because we were now confused, we called Clearance to see if we were now filed direct to GMN, but they cleared up our confusion. We were still on the POM9.GMN.


Safe Landings: The Pursuit and Presumption of Balance 

Weight and balance has been a critical issue in aircraft operations since the beginning of aviation. Loading errors can go unnoticed and have potential to cause great harm. Clerical mistakes that account for cargo weight and location can be subtle and equally costly.

This month’s CALLBACK examines several reports that highlight weight and balance errors. In the following accounts, all the aircraft unknowingly departed with uncertain centers of gravity and most departed with an inaccurate gross weight that was assumed correct. Many of the mistakes were not discovered until the aircraft was airborne and some, not until the aircraft landed. Other similarities included unknown cargo weights and freight that was loaded in improper locations. These mistakes might have been prevented. The ASRS report excerpts reiterate the need for attentiveness and accuracy in every aspect of weight and balance procedures.

The first three reports describe incidents where cargo was loaded in the wrong location on the aircraft. The remaining accounts detail various other errors that were experienced in Air Carrier Operations. 

The Usual Suspects 

Cargo loaded into the wrong compartment and closeout paperwork that did not specify its location allowed this B737 Flight Crew to launch with an inaccurate Center of Gravity (CG) that was not discovered until after the aircraft landed. 

• The [destination station] Crew Chief came to the cockpit and inquired about how the aircraft handled during our flight… He then informed me that according to his paperwork, all cargo should have been loaded in the aft compartment, but when opened, he found it completely empty. Upon further inspection, he found that all cargo was loaded in the forward compartment. I checked my load planning paperwork and found the plan was for 1,900 pounds of cargo to be loaded in the aft compartment. Closeout paperwork showed 1,100 pounds of cargo with no indication whether forward or aft.

I then called Dispatch and was transferred to Load Planning. They checked the computer and said that all cargo should have been loaded in the aft compartment…, but that was not the case. Actual loading was in the forward compartment.

We had a light load of only 105 souls on board and a light cargo load. The Load Agent ran the numbers with the actual cargo in the forward compartment and found that we were still within safe CG limits. How much [misloaded] cargo weight would it have taken on this aircraft to create an unsafe situation? Would a full load of passengers have helped or hindered the situation? How about fuel burn on a long flight? Is it the Ground Crew’s habit to load cargo in the forward [compartment] on smaller aircraft? Did they fall back on habit or disregard loading documents? 

The load closeout we receive in the cockpit does not show forward or aft cargo weights. It just shows total weight and a breakdown for live animals and restricted articles. Maybe we should receive that information on closeout. Although that would not have helped in this situation since all the “paperwork” was correct.

Trust but Verify

Non-standard operations resulted in freight being placed in the wrong cargo compartment of this B737-800. The Flight Crew was unable to confirm compliance with loading instructions.

• After the parking brake was released for push back, the Ground Crew opened the forward cargo door twice without notifying the Captain. The Captain flew to our destination and other than noting that the aircraft was nose heavy on takeoff, the flight was uneventful. After we parked, the Crew Chief entered the cockpit as the passengers were deplaning. He explained that the cargo had been incorrectly loaded and pointed to his offload report. The report clearly showed that only one bag should have been placed in the forward cargo and the rest should have been in the aft cargo. The Crew Chief reported that the aft cargo was empty and all the bags were in the forward cargo. Obviously this was a very serious issue—one that could have caused aircraft controllability issues, or worse… Pilots should have the same paperwork used to load the aircraft, so we can double check with the load closeout and takeoff performance data and verify proper loading.

The Edge of the Envelope

This CRJ-700 Captain directed that ballast be added to the forward cargo compartment but got a post-flight surprise.

• Due to ACARS weight and balance, I directed the Ramp Lead to move the one and only bag from the aft cargo compartment to the front and to add 500 pounds of ballast to the front cargo compartment. On rotation we noticed a slight nose up pitch tendency but dismissed it as normal for the aft CG limit. On arrival, the First Officer discovered that the 500 pounds of ballast had been placed in the aft cargo compartment.

Late Arrivals

Conflicting load numbers that surfaced during preflight planning remained suspect into the flight, nurtured mistrust, and spawned a weight and balance error for this A319 Flight Crew.

• We received a flow release time from ATC that was 10 minutes from our scheduled push time. At push, we had not received the weights, so I sent an ACARS [message] because I wanted to make sure we had the weights to make our slot time. I received the response that weights were not available because the ramp had not completed the loading document. We continued to taxi to the active runway where we held for 10 minutes waiting for weights and missed our slot time. I called Station Operations, and they said they were talking to Load Planning about the weights. We waited another five minutes and received a Dispatch ACARS message stating our zero fuel weight had gone up 4,000 pounds with new [projected] fuel burn and fuel at touchdown numbers. We acknowledged the increase and accepted the numbers. The weight manifest printed, and it showed our weight below the weight I had used to calculate performance numbers. After we departed, we received another weight manifest with an even lower gross weight and numbers closer to the planned weights on the flight plan. While the numbers we were working with resulted in minimal changes in the CG, there was potential for a very serious error to occur.

Missing from the Manifest

Upon arrival, this Air Carrier Flight Crew noticed three tires being offloaded but had no paperwork or knowledge that they were even onboard during the flight.

From the First Officer’s report:

• The Captain and I, upon receiving the load sheet, asked the Ramp Agent if it was correct. We were told that it was. During the post-flight inspection, I noticed Ground Operations removing three main tires from our [aft] baggage compartment. I did not remember seeing this on the load sheet, so I went back up to the cockpit and took [another] look at the load sheet. To my surprise, there were no tires listed in the baggage compartment. We departed unaware that we had an extra 300 pounds of cargo in the back of the aircraft.

From the Captain’s report:

• During the post-flight walk around, the First Officer noticed that three tires were being removed from the [aft] cargo bin. He asked the Ramp Agent if those were on our flight, and he replied that they were. The First Officer got the cargo load report from the trash, and it showed no cargo [listed] on the airplane other than the standard bags, the heavy checked bag, and the gate claim items. Each tire weighs 100 pounds, so 300 pounds were missing from the cargo load report. We both agreed that missing items on the cargo load report was a safety of flight issue.

Who’s on First?

An ERJ-170 Flight Crew took off with an inaccurate cargo weight. The correct weight would have identified an out-of-balance condition and an exceeding of structural limitations.

• The ramp personnel asked the First Officer during his walk around if we could accommodate… freight weighing a total of approximately 2,000 pounds. He instructed them to wait on loading until he could confirm that the load could be safely accommodated. When the First Officer returned to the ramp, the cargo was already loaded in the aft compartment, and he was told it was approximately 1,000 pounds. When we received the cargo load report, it indicated a total load of 59 standard and five heavy bags in forward cargo and 1,000 pounds of freight loaded in the aft cargo compartment. We ran the reported load, and after reseating four passengers as a result, we received good takeoff performance numbers. After closing the door, the tug driver said they had made a mistake and that we should add one standard bag to the forward compartment and that the actual weight in the rear was 2,200 pounds. I asked twice to clarify these numbers, but I wasn’t confident in his count. We ran new numbers anyway and adjusted the passengers, once again, per the ACARS instruction.

I called Ops before taxiing to confirm the load numbers. The ramp manager told me that the second numbers I had received were, in fact, accurate. Just prior to reaching the runway, we received a message from Dispatch stating to once again add two bags to the forward cargo. After a normal takeoff and being airborne for approximately 30 minutes, Dispatch informed us that the load in the rear cargo compartment was actually 4,000 pounds. The cargo compartment’s weight limitation was exceeded… They [then] informed me that the CG was out of limits and… the decision was made to divert. After a 74,000 pound uneventful landing, Ramp personnel removed and weighed all cargo from both front and rear compartments. The actual contents of both compartments were: 62 standard and four heavy [bags] forward, and 3,600 pounds in the rear compartment.


Safe Landings: February 2015

Ground Loop Lessons

By definition, a ground loop is the rotation of a fixed-wing aircraft in the horizontal plane while on the ground. It is predominantly associated with aircraft that have conventional landing gear (taildraggers) due to the center of gravity being located aft of the main gear. If horizontal rotation is all that happens, the ground loop may only affect the landing gear or cause a runway excursion. Unfortunately, aerodynamic forces can cause the advancing wing to rise, which may then cause the other wingtip to contact the surface. A ground loop that progresses to this stage may result in extensive airframe and engine damage and even personal injury.

While often caused by an unfavorable wind component or adverse runway conditions, ground loops may be caused entirely by pilot error. 

To avoid a ground loop, the pilot must respond to any directional change immediately while sufficient control authority is available to counteract the unwanted movement. In order to respond quickly enough, taildragger pilots have to anticipate the need for corrective control input. This means keeping ground loop countermeasures in mind whenever the aircraft is moving. 

To reinforce the need for taildragger pilots to keep the nose ahead of the tail, this month’s CALLBACK looks at three ground “oops!” incidents. Note that while these reports emphasize the particular need for vigilance in training scenarios, the basic techniques noted apply to all taildragger operations.

Errant Cub Strikes PAPI

This J3 Cub instructor’s observation that, “we were comfortably in control right up to the point when it became clear…we were going to depart the runway” emphasizes the need for constant vigilance in a taildragger.

I was…flying from the front seat. An ATP rated pilot was the student for tailwheel training, flying from the rear seat. We did two landings and takeoffs from a small grass field several miles from our home airport. The day was clear with very light winds, essentially calm. We returned to our airport for our final landing. 

The student had done well with his earlier landings, and I felt comfortable having him make this landing also. I briefed that a pavement landing was more challenging than grass and required even more precise directional control… We had previously discussed the center of mass location relative to the main gear and how that causes a ground loop tendency in tailwheel aircraft if the aircraft is not aligned with the direction of travel or is drifting at the time of touchdown. The final approach segment was flown precisely on speed and on glide path. As we neared touchdown and were into the landing flare, I noticed that the airplane began drifting very slightly to the right. It was my impression that the degree of drift and the alignment of the aircraft for landing were within safe limits and therefore, I continued to monitor the landing, letting the student maintain full control.

After we touched down, just at stall with the stick full aft, the aircraft began to turn gradually left. I began to assist the student on the flight controls and then said, “I have it” as the rate of turn increased. Despite full right rudder and brake, the turn developed into a progressive swerve to the left. I do not recall if I added left aileron. I noticed a small amount of power still on, and I took this out. We left the runway between the runway lights and continued to roll onto the grass. The radius of the turn tightened, and I began to see the PAPI lights to our left… As the turn continued, we went past the first three lights and slowed, but the radius of the turn tightened despite all control inputs. We struck the fourth PAPI light.

We were moving so slowly at the time of impact that we did not feel a discernible force. I checked the brakes, bungees, and tailwheel. All seemed to be intact and functional. I initially wondered if there could have been a mechanical problem because the degree of side movement seemed to be in an acceptable range at touchdown, and I was surprised by the ground loop. We did subsequently note that the tailwheel springs and linkages were somewhat loose. I have made thousands of tailwheel landings and felt this time that we were comfortably in control right up to the point when it became clear the swerve was increasing and we were going to depart the runway. I have to conclude this was mostly pilot error for not fully recognizing that lateral limits had been exceeded, perhaps exacerbated by a somewhat loose tailwheel steering linkage.

“Never Relax Your Vigilance”

The type of aircraft was not given in this report, but the lessons given are good for any taildragger. Also, the importance of not overestimating a student pilot’s ability is good advice for instructors in any type aircraft.

This was the first flight of a tailwheel endorsement for a previously endorsed pilot who had lost his documentation. He had approximately 100 hours of tailwheel time… Two hours of ground school was accomplished covering tailwheel aircraft and model specific characteristics. The start and taxi, including control positioning, was normal. The takeoff was somewhat erratic in that the control yoke was “pumped” slightly; rudder control was erratic, but satisfactory. Slow flight at various flap settings and stalls were accomplished. On the first pattern, downwind to final was satisfactory, but he elected to use 30 flaps instead of 40. As the flare was initiated, he “pumped” the yoke initially, but quickly established a proper attitude. As the aircraft touched down he relaxed backpressure and over-controlled the rudder causing a minor heading change. He then reversed the rudder, adding backpressure and causing the aircraft to become airborne and change direction. At this point, I commanded him to hold the yoke with a nose up attitude and center the rudder; however, he relaxed backpressure, allowing the aircraft to touch down. His rudder input at this time was excessive (push and hold rather than the quick inputs required for a taildragger). 

I took control of the aircraft (at this time we were very slow), but I could not override his rudder input in a timely manner. The aircraft did a slow ground loop, exiting the runway. It was more of a quick turn than a classic ground loop. I reentered the runway and taxied back to the ramp to perform an inspection. There was nothing wrong with the aircraft or tail wheel assembly.

I have about 5,000 hours of instructor time with no incidents/accidents and have trained many pilots, but I committed a cardinal sin in having higher expectations for this pilot than warranted based upon his experience. Could this have caused me to relax my vigilance? It probably did… When the student started pumping the yoke at the initial round out, I should have taken the aircraft and performed a go-around. I also did not demo the first landing, which is usually my method of operation.

This event reiterated the fact that a demo is also appropriate for someone who has never flown a particular model and [I should] never fail to take timely control of the aircraft, even though someone has extensive experience. Never relax your vigilance.

Wayward WACO

Even a very experienced instructor pilot might not be able to overcome a student pilot’s error when it involves a critical action at a critical time. The situation is aggravated in an aircraft such as this WACO where the instructor was unable to see, and possibly anticipate, the student pilot’s actions.

The objective of the flight was to practice takeoffs and landings on a paved runway, which is more difficult and challenging than operations from a turf runway in a vintage aircraft of this type… The decision was made to practice at a nearby field where there is a 150-foot-wide runway. 

A key point in technique that had been stressed…was not to touch the brakes until the tail wheel was on the ground when making a wheel landing. Moreover, one should not try to force the tail down once on the ground in the wheel landing attitude but rather let the tail come down on its own, maintaining directional control with the rudder only; no brakes during this phase of the landing roll out.

The point had been previously stressed and understood by the student that forcing the tail down (pulling it down with the stick) prematurely was a good way to induce a ground loop because this action would dramatically increase the angle of attack on the wing when it still had enough speed remaining to generate some lift and enough lift, if helped along by any crosswind, to cause the aircraft to yaw and thereby cause the downwind wing to hit the ground and begin a ground loop event. Application of brakes while the tail was still flying could also cause enough adverse yaw to induce a ground loop or even worse, flip the aircraft over.

Conditions at the time were ideal. Wind was less than five knots. When the airplane touched down on the main wheels, directional control was good, and it was tracking straight. Then it began to yaw to the right as speed decreased and the tail began to lower. This is a critical time where the pilot flying needs to immediately arrest the yaw with opposite rudder even if aggressive opposite rudder is necessary but no brakes. Instead, the student hit the left brake fairly hard.

Now the right yaw, which was only about 10 degrees, suddenly became a sharp yaw to the left at about 45 degrees. At this point, the airplane was headed off the runway onto the grass, and it struck a runway light where it departed the runway. The critical error was that the student stomped on the left brake when the aircraft began to yaw to the right while the tail was still flying.

This is an antique aircraft. The instructor pilot sits in the front cockpit. The instructor cannot see what the flying pilot is doing with his feet or how he has them positioned on the rudder pedals.


Safe Landings: January 2015

“The Airplane was still in a Descent with Full Power”

Faced with little IFR experience, poor CRM, and airframe icing, the pilots of a Rockwell 112 were lucky to break out into conditions that would allow the ice to dissipate. Among the lessons this incident highlights are the need for an adverse weather “escape plan,” and the value of building actual instrument time with a qualified instructor until proficiency is attained.

• Sunset was imminent, this area of the country was new to me, and the more things changed for the worse, the more interest I had in parking the airplane and just spending the night in a hotel.

Always leave an out. The area over the airport…was in IMC. Ordinarily this would not have been an issue. The AWOS indicated a 1,500-foot ceiling. Things were going smoothly then at 6,000 feet, with no control input to cause a descent rate of more than 500 feet per minute, my VFR rated passenger told me that we were descending (I could see that and was trying to process why we were descending). He further stated that I needed to “fly the airplane.” Then he took the controls and pulled back on the yoke. The attitude indicator shifted to a very sharp indication of a left turn. The descent rate increased to about 1,500 feet per minute. I could not over power this person. I told him, “The airplane was flying a minute ago; let the airplane continue to fly.” He let go of the controls. I reiterated that announcing, “Your airplane/my airplane” prior to manipulating any controls was a requirement when flying with me.

The airplane was still in a descent with full power after he released the controls. It took a while to discover that we had ice on the wings. We broke out into VMC and ATC asked what my intentions were. I explained that I needed to stay VFR to dissipate the ice and would like the approach into [a nearby airport].

I do not have much experience as an IFR pilot, less than 20 hours in actual IMC. I thought my passenger, with more than 50 years of aviation experience, would be an asset in the cockpit. In VMC, he is a continuing source of information and a person I respect. But there is a difference between being IFR rated and VFR rated…. Being diverted 59 miles south due to the iced over runways was already putting me outside of my comfort range. I usually fly in [warmer states].

The majority of my flight instructors had minimal or no experience in actual IMC. I will be signing up for a course on “icing” in the near future.


“The Airspeed Was Decreasing Rapidly and I Began to Worry”

ATC helped to get a trio of pilots in a PA32 out of trouble as they dealt with zero IAS, no GPS, and ice on the leading edge of the wings. FAA and NTSB statistics show that accidents often result from similar scenarios, especially when “get-there-itis” is added to the mix.

• We decided to depart knowing that most of the flight would be VFR, but the last 100 miles would be in marginal conditions. We planned on stopping prior to encountering the marginal conditions…. Once at [the interim stop], we refueled and obtained a telephone weather briefing. For the briefing, I had one of the other pilots call the briefer. We spoke about the information the briefer gave him. We were told that the freezing level was at the surface and above. Our understanding was that other aircraft were not having problems with icing, but were experiencing moderate turbulence.

At this point, my mindset was that icing could happen but would not be an extreme hazard to us. I spoke with the owner of the airplane, (our passenger) and told him that as soon as we got any accumulation of icing, we would divert to an airport that was along our route. We chose our route with that plan in mind.

Another pilot performed the pre-flight. Thirty minutes after departure, the weather became marginal, and I avoided IMC while we got an IFR clearance from TRACON. We were cleared to climb to 7,000 feet to see if we could find VFR conditions. During the climb, we started to notice icing. We then climbed to 8,000 feet, but we were still IMC so we asked for 6,000 feet.

At 6,000 feet, I noticed that the airspeed was decreasing rapidly, and I began to worry. For a while, I thought we were losing the capability to generate lift due to the icing, and by instinct, I reacted by reducing the pitch of the aircraft to avoid a stall. At this point, the other pilot suggested it was the pitot tube getting clogged by the ice, which was the cause of the IAS decrease. I then noticed that our altitude was now 5,300 feet MSL….

While all of this was happening, the pilot beside me was attempting to coordinate with ATC to help us get back down to an airport…. The Controller gave us an Initial Approach Fix and told us to fly direct to that fix. As we were loading the information into the GPS system, we realized that the GPS had lost its signal. With IAS at zero, no GPS, and ice on the leading edge of the wing, we thought it would be best to get vectors to the nearest airport…. Flying at a lower altitude in VMC helped us regain the airspeed indicator and land in visual conditions.

The one factor that hurt us the most was “get-there-itis.” I had been asked to help out the owner with the flight since I had more experience than he did. The owner was spending a lot of money for each night at a hotel and wanted to get back soon.

I should have been more involved with the pre-flight and weather briefing and not just taken the other pilot’s word since I am the PIC. I also just assumed that the pitot heat was working since I assumed it was the responsibility of the owner to ensure that the plane was up to date on maintenance.

Low Level Ingestion

After experiencing a flameout and “vigorous” relight on one of the engines, a BE100 pilot was able to regain control after breaking out of the clouds. There was no mention of how low the airplane descended, but since a “climb to 3,000 feet” ensued, this was undoubtedly a chilling lesson on the dangers of engine inlet icing.

• Before the final turn for the ILS approach, with all deicing equipment on, the right engine seemed to stutter. I assumed it was the right engine from the direction that the plane was suddenly going. I corrected with left aileron and rudder. I did not see a decrease in torque with either engine when I looked at the gauges. The engine restarted vigorously and pushed the plane hard left and into a fast descent. I was able to control this just as I broke out of the clouds. I climbed to 3,000 feet, stabilized the plane, and made the approach without further incident.

After landing, I observed that the inlet to the right engine was clear of ice. The inlet to the left engine was significantly blocked by ice buildup. I suspect that the right engine flamed out, caused by ice breaking loose and entering the engine. The igniters were armed, which restarted the engine.

A possible cause is that I did not have the engine inlet heat system on soon enough to avoid ice buildup. I was watching the wings during flight and turned on the engine inlet heat system only after I observed ice on the wings. The icing was encountered at 5,000 feet enroute. There was no icing in the immediate vicinity of the destination airport.

Copyright © 2009, In Flight Media. All rights reserved.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 Unported License.
Creative Commons License

Designed by jbNadler Creative Labs