The NTSB’s narrative report on the Cessna 421B Golden Eagle crash in Hammond, Louisiana, on October 14, 2015, in which the pilot and passenger were killed, repeats a deficiency we occasionally see in other reports from the agency. It lays out facts and provides a carefully worded probable cause for the accident, while failing to alert us to things investigators discovered that might help others recognize the potential for impending doom and do something about it before it’s too late. One can appreciate that the Safety Board prefers facts to hearsay. However, in this accident, there was some hearsay I think we all should hear. It just might help us recognize when an engine is trying to tell us that something catastrophic is about to happen.
The airplane was going back to Atlanta, Georgia (KATL), after having been flown to Hammond to pick up an employee of a business. The weather at Hammond was solid VFR, with no ceiling and visibility at least 10 miles. A witness, who was an ATP-rated pilot, told investigators that he was at an FBO’s hangar and watched the airplane take off from Runway 31 at Hammond Northshore Regional Airport (KHDC). Runway 31 is 6,502 feet long by 100 feet wide. The witness said that the airplane had climbed to about 100 feet above the pavement by the runway’s midpoint. He said, “I heard a loud ‘pop’ occur, followed by slowing of the aircraft’s right engine and right propeller; the aircraft yawed to the right and the pilot appeared to apply left rudder to return the aircraft to runway heading.”
The witness continued, “...the aircraft began a right turn into the failing engine, which still had a windmilling prop at the time, towards Runway 18. After he cleared the tree line by approximately 150 feet, he then rolled over and went straight down into the field north of the approach end of Runway 18. The aircraft impacted the ground in a near vertical attitude followed by a massive explosion and fireball.”
Another witness was standing on the ramp and observed that the airplane “...appeared to be in a VMC situation at which time I saw the right wing drop. The aircraft pitched down about 60 degrees and impacted the ground; upon impact, the aircraft burst into flames.”
Up in the tower, the local controller who had cleared the airplane for takeoff saw what was happening and told the ground controller to “call CFR (Crash, Fire and Rescue), this guy is going down.” He recalled that just after the pilot had radioed that he was ready for departure, the pilot changed his mind, radioing that he needed a minute. “I told the pilot to advise when ready for departure,” he told investigators.
About 30 seconds after 3:39 p.m., the pilot radioed, “Hammond tower, Golden Eagle three three foxtrot alpha holding short runway three one, uh, ready to go...actually, no, we are going to hold short for a moment.” Just under a minute later, the pilot radioed, “...we’re ready to go, three one.”
The controller instructed the pilot to “...fly runway heading, climb and maintain two thousand, runway three one, wind calm, cleared for takeoff.”
The pilot replied that he was “...clear to go, three one, runway heading, Golden Eagle three foxtrot alpha.” The controller responded, “...climb and maintain two thousand,” and the pilot radioed, “Roger, sorry about that, climb and maintain two thousand, here we go, rolling three one.” The controller confirmed, “roger.”
At 3:41:46, the pilot radioed, “Mayday, mayday, mayday, we gotta come back right away, Golden Eagle three foxtrot alpha.” The controller replied, “November three foxtrot alpha, say again.” There was another transmission from the airplane, but it was unintelligible. The controller radioed, “November three foxtrot alpha, Runway 18, wind calm, cleared to land.” At 3:42:00, a transmission consisting of the sound of screams was heard on the tower frequency. There were no further radio calls from the airplane.
“The airplane was fragmented, and had extensive fire damage. The landing gear was retracted. Physical damage was consistent with near-vertical impact.”
The crash site was about 1,600 feet from the approach end of Runway 18. The airplane was fragmented, and had extensive fire damage. The landing gear was retracted. Physical damage was consistent with near-vertical impact. Both propellers had separated from the engines. Investigators determined that the right propeller had not been feathered, allowing it to create excessive drag. They said it’s likely the pilot allowed the airspeed to drop below the minimum required to maintain aircraft control which, combined with his failure to feather the right prop and the right turn in the direction of the failed engine, resulted in a loss of airplane control.
Toxicology testing on the pilot was performed by the FAA’s Civil Aerospace Medical Institute in Oklahoma City. It was found that the pilot had used the drugs diphenhydramine and ibuprofen. Diphenhydramine is a sedating antihistamine, which carries a warning that it may impair mental and physical abilities to perform potentially hazardous tasks such as driving and operating heavy machinery. It’s used to treat allergy symptoms and as a sleep aid. Ibuprofen is used to treat pain and fever, and is available over-the-counter. The Safety Board didn’t say whether use of these drugs affected the ability of the pilot to respond to the emergency, implying that it feels they did not.
The seven-seat Cessna 421B was powered by two Continental GTSIO-520-F-K engines. The Safety Board identified them as each producing 435 horsepower. The six-cylinder engines were designed with oil sumps holding 12 quarts. Typical speeds at a cruise altitude of FL250 would be 210 to 230 knots, at 65 percent to 75 percent power. An approximately 357 nm trip between Hammond and Atlanta would have taken about one and three-quarter hours, without factoring in anything for wind.
The engines underwent teardown inspections. Nothing was found that would have prevented normal operation of the left engine. When disassembly of the right engine began, investigators noted that all of the cylinders appeared to be new. They found that some of the nuts that were supposed to be tightly holding cylinders to the engine case were, in fact, easily loosened.
Records indicated that all of the cylinders had been replaced at the airplane’s most recent annual inspection, which took place eight months before the accident. The manufacturer’s engine overhaul guidance provides step-by-step instructions when doing cylinder replacement, and specifies the sequence for properly tightening cylinder bolts on the studs that hold things together. The steps tell when, where and in what sequence to use calibrated torque wrenches for applying forces starting at 300 inch-pounds and going all the way up to 700 inch-pounds to be sure bolts will stay put.
Inside the engine, investigators found that the crankshaft was fractured adjacent to the No. 2 main bearing. The bearing was spun, and the surfaces where the halves of the crankcase come together in the area of that bearing showed signs of fretting damage caused by there not being enough force to hold the case halves tightly together.
That discovery was enough for the Safety Board to draft its probable cause for the accident, which was “...the loss of right engine power on takeoff due to maintenance personnel’s failure to properly tighten the crankcase through studs during cylinder replacement, which resulted in crankshaft failure. Also causal to the accident were the pilot’s failure to feather the propeller on the right engine and his failure to maintain control of the twin-engine airplane while maneuvering to return to the airport.”
What must remain in the realm of hearsay is a handwritten statement from four people working at the FBO in Hammond, which handled the 421B. It was written the day after the accident and submitted to investigators. The NTSB elected not to include its contents in the official narrative of what happened, so it doesn’t have the gravitas of NTSB-endorsed facts. Nevertheless, you may find it as interesting as I did. I’ve removed the real names of the people who were the witnesses and signed the statement.
The statement said, “At approx. 2:40 p.m., (the Golden Eagle) taxied onto our ramp. (FBO Person #3) parked him. Pilot got out and went to the right engine. (FBO Person #3) asked the pilot if he needed fuel and the pilot said, ‘If the oil is okay I will top it off; if not, no.’ He was on a cell phone while standing by the right engine for several minutes.
“(FBO Person #2) noticed white smoke coming from the right engine when passenger arrived. They both (were) walking around by the engines. They came inside and told us to put three quarts of 100W50 into right engine and then top the plane off.
“After (FBO Person #3) and (FBO Person #4) put oil in the right engine, (FBO Person #4) topped four tanks with a total of 50.9 gallons. (FBO Person #1) walked out to the airplane with pilot and passenger, pulled the chocks and was told they’d be back Sunday 10/18/15.”
The actions of the pilot as reported in the statement seem to indicate he was aware of some sort of problem with the right engine. The smoke seen coming from the engine could have been caused by leaking oil hitting hot engine parts. It’s reasonable to ask, “If an FBO person saw the smoke, didn’t the pilot see it, too?”
We don’t know to whom the pilot was talking on his cell phone. It could have been the airplane’s owner or maybe a mechanic. It also could have been a call totally unrelated to the airplane or flight. But, the pilot apparently being seen hovering around the right engine might reflect that he knew something wasn’t quite right, just as the fact that he delayed takeoff at Hammond for a bit under a minute also makes us wonder what he might have been concerned about.
“And, the sight of smoke coming out of an engine compartment should be enough to make any pilot wonder whether there’s an oil leak or something that demands attention before taking the airplane aloft again.”
We don’t know whether the pilot checked the oil level for the right engine before leaving Atlanta. He would have been foolish not to, and we all know that pilots always check the oil during preflight, right? During a flight like the one from Atlanta, we might expect a healthy engine to use not more than one-fifth of a quart of oil. I’ve been told by some mechanics that a high-time engine can burn up to a quart per hour before they’d pull it from service. But, the pilot asked for three quarts to be added to the right engine after a flight that should have taken under two hours. What would compel him to check the oil after landing, and how else would he know that the right engine needed three quarts? Did the engine become an oil-burner after receiving new cylinders? The NTSB report doesn’t tell us.
I’ve heard that some operators of TSIO-520 engines find they’re happy running with eight or nine quarts in the 12-quart sump, sometimes as low as six quarts. The NTSB doesn’t tell us what was normal for the accident airplane, so we don’t know what portion of the oil in the right engine that the three replacement quarts represents. For the comparatively short flight from Atlanta to Hammond, a loss of three quarts could have been terribly significant. And, the sight of smoke coming out of an engine compartment should be enough to make any pilot wonder whether there’s an oil leak of some sort or something else that demands attention before taking the airplane aloft again. In this Golden Eagle accident, if an FBO employee saw smoke, as reported in the statement that was downplayed by the NTSB, we can assume others did, too.
Had that hearsay information been pursued, and had confirmation and additional facts been obtained, the NTSB might well have added to its probable cause that a contributing factor to this accident was the pilot’s decision to conduct the flight despite obvious warnings of impending engine trouble.
The next time I land and find that my engine is down three quarts of oil, or I see smoke coming from under the cowling, you can bet I’m going to think of this accident. I’ll recall the hearsay information that surfaced after the accident. And, I’ll likely be staying at my destination longer than planned while I put a mechanic to work finding out what message the engine is trying to send.
Peter Katz is editor and publisher of NTSB Reporter, an independent monthly update on aircraft accident investigations and other news concerning the National Transportation Safety Board. To subscribe, visit www.ntsbreporter.us or write to: NTSB Reporter, Subscription Dept., P.O. Box 831, White Plains, NY 10602-0831.