As I write this near the end of August 2020, the “ADS Crash” thread on the BeechTalk forum is approaching 100 pages. As most of you know, in June 2019 a tragic accident occurred at Addison airport (KADS) in Dallas, Texas. A nearly new King Air 350i departing on a flight to Florida, crashed into a hangar on the airport just seconds after liftoff from Runway 15. All 10 people aboard – eight passengers and a crew of two – died. The NTSB preliminary docket was just released. This document presents their findings thus far concerning the facts of the case, but no conclusion of the cause has yet been found nor presented.
As you may imagine, this tragedy impacted most of us in the King Air community very strongly and it is easy to understand why the thread on BeechTalk is as large as it is. One of the many discussion points that has been raised is the position of the pilot’s right hand at rotation. Let’s discuss that.
Not until I received my Learjet type rating back in 1978 was I ever exposed to the procedure of “Right hand moves from throttles to control wheel at V1.” The Lear was the first airplane that I ever flew that was certificated under the rules of FAR Part 25 … the “big airplane” rules. Here, an engine failure during takeoff had to be considered and charts/procedures had to exist to allow either a successful abort or a successful continued one-engine takeoff, depending on the speed at which the engine failure was recognized. That is, Accelerate-Stop and Accelerate-Go distances, both had to be possible in the distances available on the runway.
With the relatively stellar single-engine performance of the Lear, the “Going” procedure lacked the drama that we experience in a light, piston-engine twin trainer. It is difficult to make a cogent argument for aborting the takeoff after V1 when the airplane can continue so well even with an engine failed. In fact, high speed aborts were riddled with unsuccessful outcomes – going off the end or side of the runway, incurring major damage and, often, resulting in death to the occupants.
The procedure I was taught by the FlightSafety team in Wichita was to reposition my right hand from the thrust levers (“throttles” for us piston-trained types) onto the control wheel when the V1 call was made by the PNF (Pilot-Not-Flying). That helped to ensure that the thrust levers would never be erroneously retarded to initiate a stopping procedure after Decision Speed (the name for which V1 is the abbreviation) was reached … when the going procedure was the correct and expected reaction to the recognition of a power loss.
I can definitely see the benefit of repositioning the right hand as airspeed passes Decision Speed for airplanes that have guaranteed capability to continue on the remaining engine. All the airliner jets, certified under FAR 25 rules, do it this way and rightly so. How about King Airs?
What King Air are you operating? Just counting standard, civilian models there are over 25 different types, each with its own flight manual. The certification rules decide what performance charts must be in that manual. As years pass, as rules change, as aircraft categories change, as maximum takeoff weight changes … all of these variables have an impact on what performance charts are presented in the flight manual. Heck, most King Air models don’t even list a V1 speed!
From a legal standpoint, must the operator comply with all of the charts in the POM (Pilot’s Operating Manual) or POH (Pilot’s Operating Handbook)? Surprisingly, the answer is no. With the exception of the 300-series, all King Airs are “light twins” since their maximum takeoff weight does not exceed 12,500 pounds. Unlike the heavier airplanes that fall in the Transport Category of rules, there is no requirement for a light twin to have the ability to continue a takeoff with an engine failure. Yes, Beech does the flight testing that allows performance charts to be created that give both Accelerate-Stop Distance and Accelerate-Go Distance under differing conditions of weight, OAT and wind. But they are for information only with no requirement to operate off runways that are long enough to meet these distances. That’s a darn good thing, too! Taking your B90 into that 3,000-foot-long strip located near the boss’s lakeside cabin would not be possible if “big boy” rules applied.
As we all should know, there is a calculated risk factor associated with many of the runways we use: There will be seconds of time during the takeoff when we will be going fast enough that we will go off the end of the runway if we abort and yet we are still slow enough that we won’t be able to climb enough on one engine to clear the obstacles beyond the runway. With the reliability of turbine engines, the risk that an engine will fail during these critical few seconds of time is a risk most of us are willing to take.
However, what if the unlikely but possible event does occur? What if one of our PT6s does indeed have a catastrophic failure at, say 95 knots? If we abort under these assumed conditions, use the brakes to their maximum and maintain directional control, we will perhaps still be going 30 knots when we hit the ditch. But if we try to fly at 95 knots?! We are still more than 10 knots below single engine best-rate-of-climb speed (VYSE, Blue Line), we have not yet started retracting the gear, and unless we are equipped with autofeather we still have tons of windmilling propeller drag. If we try to continue it will be very likely we will lose airspeed as we try to climb in this impossible situation and quickly encounter the deadly VMC loss of control. At best, if we keep the nose down to maintain airspeed and control, we hit the trees still going 90 knots. Abort, hit the ditch at 30. Try to continue, hit the trees at 90. Let’s see: Three times the speed is nine times the kinetic energy to dissipate. There is a lot more chance we will live through the 30-knot hit into the ditch, eh?
So, where I am going with this is to state that my right hand is staying on the power levers at least until I am above nearby obstacles and the airspeed is at least at VYSE. Why? So, I can readily pull the power levers to Idle to initiate the abort. Isn’t this exactly what we should do in most other light twins? Due to the King Air’s increased single engine climb capability compared to a Baron, our seconds of exposure to bad risk during takeoff from short runways is less than in a Baron but it is still there.
So when does my right hand leave the power levers? If I am flying single-pilot, it does so when I reach for the landing gear handle. And when is that? When continuing looks more favorable than stopping. But as soon as the gear handle is up, I return my hand to the power levers. With the exception of the brand-new, just announced King Air 360 model, no other King Airs have any power limiters except for the human controlling those power levers. It is the pilot’s job to adjust the levers to get the torque or ITT value that is desired. As the airplane climbs, adjustments will have to be made. Hence, the right hand spends a lot of time on the power levers. It is not a “select the correct detent and forget” as in the FADEC-equipped turbines. (FADEC? Full Authority Digital Engine Control)
I hope that almost all who are reading this article are familiar with the phenomenon of “Power Lever Migration,” (PLM). That action may have played a role in the crash at KADS. For sure, it has contributed to other fatal King Air takeoff accidents. A spring at the engine-end of the power lever cable is always trying to retard the Fuel Control Unit (FCU) to its idle setting. Why? To ensure that the engine is not damaged when subjected to torque and temp limits that would be exceeded if/when the power lever cable malfunctions/disconnects and tells the FCU to go to maximum Ng speed. Because (1) the Power Lever (PL) connects to the FCU on the right side of the PT6 and (2) the cockpit end of the PL sits slightly left of fuselage centerline, the left PL cable is significantly shorter than the right. Hence, it tends to move more freely, to have less “stiction”… a combination of friction and stickiness. If the individual PL friction knobs are not both sufficiently “snugged up,” it is probable that both PLs will move, migrate back toward idle when unattended. Since the force a spring exerts increases as it is lengthened from its resting position, the pullback force increases as PLs are advanced.
Try it yourself in your hangar on your actual King Air. Turn both friction knobs fully counterclockwise (the looser direction) and push the PLs fully forward then take your hand away. Some of you will see no result, some will see one or both PLs move aft a little, and some will likely see the left lever move most of the way to idle and the right lever come maybe halfway back. It all depends on the “stiction” in your actual airplane. Before you leave the cockpit, remember to turn the friction knobs clockwise enough to ensure they are “snugged up” to where you want them!
If the checklist step of setting proper PL friction before takeoff is overlooked and if the friction is too loose, the stage is set for a huge surprise when the pilot releases the power levers to reach for the gear handle! In most cases (thank you, Lord!) the outcome is more humorous than scary. Of course the pilot sees the levers move and he immediately puts his hand back on them and advances them as desired … probably with the gear still down. Now it’s a matter of taking his left hand off the control wheel long enough to (A) get the gear up if the handle is on the left side of the cockpit as it is in all newer
KA models, (B) tighten the friction knobs and (C) go back to flying the airplane with his left hand.
If the PL migration aft is not noticed, however? It sets the stage for tragedy.
Realize the following facts. First, power lever position controls the engine’s compressor speed (Ng) directly, not torque. Of course, torque and ITT are what we watch as we adjust power but the reason they change is because we selected a new Ng speed. It’s said that about the last 50% of available power is all achieved in about the last 12% of Ng. Thus, it takes relatively little PLM to cause a major loss of power. In the case of the left PL moving nearly to idle while the right PL comes back only one-fourth of the way, we would have close to no power on the left and maybe only 50% on the right.
Second, the marvelous autofeather system must be able to know that a loss of power is unintentional not intentional. It achieves this knowledge by looking at PL position. If the PL is moved back, the autofeather “thinks” that the resultant power reduction was requested by the pilot and feathering the propeller would not be appropriate. But if the PL is still well-advanced yet little power is being produced … something’s wrong and autofeather will ride to the rescue and feather the propeller to reduce the drag. Therefore, when PLM occurs, autofeather is rendered inoperative. In fact, if either PL is retarded both propellers are incapable of automatically feathering.
Third, the +10-degree pitch attitude we have been taught to hold after takeoff (the 350’s POH is the only one that actually states that value but it works rather well for all KA models) will not be correct with the windmilling prop and, probably, reduced power on not only the left but also, to a typically lesser extent, on the right as well. Airspeed will rapidly decay if that attitude is maintained.
Keeping the right hand on the power levers obviously eliminates the possibility of PLM. I am so used to doing this in piston-twins as well as King Airs, personally I do not reposition my hand when Decision Speed is reached even in the 350. Don’t get me wrong: I am not saying that I’d pull power back and abort above V1 in a 350, but the hand stays on the PLs. I have no argument whatsoever with those 350 pilots who do move their hand to the wheel at V1.
But folks, I beg you … always prevent Power Lever Migration! How? First, by snugging up those friction knobs! How do you know they’re set properly in a King Air that is new to you? One way is to lift your hand just slightly above the PLs after you have set takeoff power. Did anything move? No? Then put your hand back down and continue. Did a lever start to retard? Depending on the length of the runway, either pull both back to idle and taxi back for another try, now with tighter friction, or turn the friction knob clockwise a little more before you reset the desired power as you continue rolling. And please realize that takeoff power should be set well before 60 KIAS. I’m not implying that you do this quick release of the PLs near rotation!
How about when flying as a crew of two? The earlier King Air models have the landing gear handle on the co-pilot’s side so there is never a need for the left-seater to release the PLs right after liftoff. Even when the handle is on the pilot’s side, some crews have the co-pilot reach across to activate it. The only thing I, personally, have against this procedure is that the reaching across looks rather weird and unprofessional. To guarantee no PLM if the pilot will raise his own gear, merely ensure that the co-pilot’s hand is positioned such that the PLs cannot migrate aft. I strongly advocate that this be standard operating practice (SOP). It is so simple for the right-seater to rest his hand on the power quadrant with the tips of his first and middle fingers touching the base of the right and left PLs. As the left-seater adds power, the right-seater’s hand slides forward to maintain light contact with the PLs. In the event of an abort, the co-pilot’s hand can be instantly withdrawn. After liftoff, after the gear is raised and the pilot’s hand returns to the PLs, then the co-pilot can move his left hand as desired.
As stated before in this article, no conclusion has yet been reached by the NTSB on the cause of the Addison crash. The cockpit voice recorder, sadly, shows that proper crew-coordination procedures were non-existent. No checklist was ever verbally requested. No briefing was ever given before takeoff. What happened, in my view, jibes perfectly with a Power Lever Migration scenario that was not noticed and not corrected. In addition, airplane control was not maintained. The airplane was allowed to yaw badly to the left and the airspeed was allowed to get too slow, leading to the loss of control. My condolences to the families and friends of those killed.