Douglas DC-8 accident at New Orleans, Louisiana, 25 Feb 1964

Accident narrative

On February 25, 1964, Eastern Air Lines Flight 304, a Douglas DC-8, crashed into Lake Pontchartrain approximately 19 miles northeast of New Orleans International Airport. The accident occurred at approximately 0205 central standard time. All 51 passengers and seven crew members were fatally injured, and the aircraft was destroyed upon impact.

### The flight

Flight 304 originated in Mexico City, with scheduled stops in New Orleans, Atlanta, and Washington, en route to New York City. The flight crew consisted of Captain William B. Zeng, First Officer Grant R. Newby, and Pilot/Engineer Harry Idol, along with four flight attendants. Captain Zeng had 19,160 hours of pilot time, including 916 hours in the DC-8. First Officer Newby had 10,734 hours, with 2,404 in the DC-8. Newby had previously been the flying first officer on another DC-8 that experienced an in-flight upset and steep dive in November 1963, from which he successfully recovered using reverse thrust.

The inbound crew arriving in Mexico City reported that the aircraft's Pitch Trim Compensator (PTC) was inoperative. The aircraft was dispatched with a flight plan for a reduced airspeed, in accordance with company procedures for this condition. The flight to New Orleans was routine, landing at 0051.

Following a crew change and passenger boarding, the aircraft's computed takeoff gross weight was 213,871 pounds, below the maximum allowable for the airport, with the center of gravity within limits.

### Sequence of events

At 0159:46, the local tower controller observed Flight 304 commence its takeoff. The lift-off appeared normal. At approximately 0201, the tower advised the flight to contact Departure Control, which the crew acknowledged. The controller estimated the aircraft was two or three miles north of the airport when its lights disappeared into the 1,000-foot overcast.

Departure Control established radar and voice contact, advising the flight to turn to a right heading of 030 degrees. At 0202:38, a radar handoff was effected to the New Orleans Air Route Traffic Control Center (ARTCC). Departure Control instructed the flight to contact the Center on frequency 123.6. At 0203:15, the crew replied, "OK." This was the last transmission received from the flight.

At 0205:40, having received no transmissions, the Center controller contacted the departure controller. Both confirmed that the radar target associated with Flight 304 had disappeared from their scopes. The last position noted was approximately eight miles from the New Orleans VORTAC.

Witnesses on the north shore of Lake Pontchartrain reported hearing an explosive rumble, and three described a tornado-like sound or terrible scream. Three witnesses also stated they saw a fire-like glow in the vicinity of the lake.

### What the investigation found

**The wreckage** The wreckage was located in Lake Pontchartrain. Recovery operations eventually retrieved approximately 60 percent of the aircraft by weight. The general pattern of breakup showed extreme fragmentation, but detailed examination revealed no evidence of in-flight fire, explosion, or structural failure. The flaps and landing gear were up.

The four engines evidenced severe disintegration at impact but no pre-impact operating distress. The recovered reverser assemblies indicated that reverse thrust was being used at impact. The Board noted that the use of reverse thrust in flight acts as a speed brake and produces a noseup pitching moment, indicating an attempt by the crew to recover from a diving attitude. The symmetry of the powerplant damage and the confined wreckage area led the Board to assume the aircraft struck the water structurally intact, essentially level laterally, and at a dive angle in excess of 20 degrees.

**Weather** A surface weather observation at 0210 reported a measured 1,000-foot overcast ceiling and seven miles visibility. The forecast called for moderate to severe turbulence in thunderstorms and heavier showers. Another large jet departing New Orleans immediately after Flight 304 reported light to moderate turbulence entering the overcast at 1,200 feet. The flight recorder from this second aircraft showed accelerations between +0.2 and +1.9-g from 2,000 to 6,000 feet, which the Board noted indicated severe turbulence. The Board believed that Flight 304 encountered moderate and probably severe wind shear turbulence while in the clouds below 6,000 feet.

**Flight controls and stabilizer** The DC-8 is controlled longitudinally by elevators and a variable incidence horizontal stabilizer. The recovered left and right stabilizer jackscrews were found within one turn of the full aircraft nosedown (AND) trim setting.

Examination of the stabilizer drive unit revealed abnormal wear patterns on the splines, indicating the unit had operated with a 1/4-inch displacement over an extended period of time. A lower support bushing was not recovered. Tests by the manufacturer demonstrated that if this bushing were installed inverted, it could fall free, allowing the drive shaft to drop 1/4 inch and reproduce the wear pattern. The Board concluded that Eastern Air Lines maintenance personnel had installed the drive unit with the bushing inverted in September 1963.

The rivets attaching the lower sprocket to the assembly had sheared circumferentially from loads applied in the aircraft noseup (ANU) direction. The Board concluded the drive unit failed while being operated in an ANU direction from the full, or near full, AND position.

**Pitch Trim Compensator (PTC)** The PTC system applies noseup control through the elevator system to offset nosedown pitching moments at high speeds. The PTC computer on the accident aircraft had been changed eight times recently, and the system was known to be inoperative upon departure.

The Board discovered that another DC-8 leased by Eastern Air Lines had been misrigged by company maintenance when a different model PTC actuator was installed, causing it to be extended 1/2 inch when fully retracted. A similar actuator model change had been performed on the accident aircraft. The Board noted that the aircraft's maintenance records showed 11 autopilot malfunctions in the last 30 days, many involving longitudinal control and automatic disconnects. The Board stated these were symptomatic of a partially extended PTC actuator, and accepted the possibility that the aircraft was operating with an unprogrammed PTC extension.

**Aircraft stability and instruments** Flight tests conducted after the accident revealed that trimming the DC-8 to a 2-degree AND position to counteract an extended PTC shifts the control column neutral point. At 220 knots, the stick force gradient becomes extremely light, dropping to about 13 pounds for a 2.5-g maneuver. A test pilot reported that attempting to maneuver the aircraft in this configuration resulted in sharp reversals in maneuvering stability.

The Board also noted that the aircraft's attitude indicator, a Collins 105 Approach Horizon, featured a small, all-black face. Its display ratio changed at higher pitch angles, meaning it indicated a reduced rate of pitch when attitude changes exceeded 20 degrees. The Board stated this could cause a pilot to view the aircraft in a less severe attitude than it actually was, potentially leading to over-control.

**Analysis of the sequence** The Board believed that the crew, unconcerned about turbulence below 14,000 feet, likely chose a normal climb speed of 310 knots rather than a lower rough-air speed. The Board hypothesized that the crew may have used full AND stabilizer shortly after takeoff to counteract an extended PTC actuator. When they attempted to trim noseup as airspeed increased, the misaligned stabilizer drive sprocket failed.

The Board concluded that the trim failure either contributed to the introduction of a pilot-induced oscillation in turbulence or hindered recovery from one. If the aircraft reached a steep nosedown attitude, the failed stabilizer drive system would establish a lower limit to the pilot's ability to recover. The Board noted that high stick forces can stall the trim motor, which the crew might have easily concluded was the reason for its failure to operate.

### Probable cause

The Board determined the probable cause of this accident was "the degradation of aircraft stability characteristics in turbulence, because of abnormal longitudinal trim component positions."