NTSB Factual Report

On March 12, 2010, about 1804 central standard time, an Embraer EMB-145LR, N16561, operated by ExpressJet Airlines, Inc. as flight 5685, received minor damage when it sustained a lightning strike while on approach to runway 4R (8,075 feet by 150 feet, grooved asphalt) at the O'Hare International Airport (ORD), near Chicago, Illinois. The flightcrew encountered elevator control binding and the flight landed without further incident. There were no injuries to the 2 pilots, 1 flight attendant, and 42 passengers on board. The 14 CFR Part 121 flight was being conducted in visual meteorological conditions at the time of the incident and was on an activated instrument flight rules flight plan. Flight 5685 departed from the Greenville Spartanburg International Airport, near Greer, South Carolina, about 1650, and was destined for ORD.

The captain reported that the flight was on the ROYKO 3 arrival into ORD. He, in part, stated: "As we approached one of the arrival fixes (MONKZ), we were struck by lightning. The altitude we were flying at was 7,000ft MSL. We notified Chicago Approach control and informed them of the strike and that we needed to land as soon as practical. We were given an assigned heading of 270 and the controller asked if everything was ok. When the strike occurred, the autopilot was engaged and after the strike it was disconnected by the aircraft. The aircraft was now being hand flown by the first officer. In addition, the colors on the PFD [primary flight display] and MFD [multifunction display] changed from their standard colors to variations of red, purple, green, blue and white. … The flight was uneventful until about 100ft AGL [above ground level] when the first officer went to lower the nose a little bit to soften the landing and pointed out that the control column would not move forward from the current position. The yoke was about 60 [percent] pulled back at this point. I attempted to push forward without any success. The aircraft landed in the touchdown zone as the landing was uneventful."

The co-pilot, in part, reported: "After the strike, the autopilot disengaged. I began hand flying the aircraft and silenced the autopilot aural warning. We noticed the screens in the aircraft had changed to different colors, but all information was still present and correct. ... I reengaged the autopilot several minutes later and no abnormalities occurred. We discussed the possible location of the strike and the possibility of the strike blowing a tire. ... I disengaged the autopilot on final approach and the approach was uneventful. During the flare, at approximately 10-20 [feet AGL] I noticed an increase in pressure be required to lower the nose during the flare. I pushed hard and the aircraft didn’t respond. I called attention to the abnormality and the captain attempted to push on his yoke. Approximately 1-2 seconds later the aircraft touched down. The yoke seemed to be able be pulled aft but would not move forward past about 60 [percent]."

The flightcrew reported that as the captain shut down the engines the aircraft went into essential power. Emergency light were turned on and the crew assisted the passengers off the aircraft with the use of the emergency flashlights. The crew found the damage to the tailcone of the aircraft during a post flight inspection. On-scene images revealed that an internal bulkhead frame came to rest on the elevator bellcrank.

The tailcone was a monocoque assembly manufactured with fiberglass, reinforced with two frames internally. All its outside skin was further covered with an expanded copper foil. The assembly was installed at rear horizontal stabilizer in order to provide means to install the position and anti-collision light assembly and to improve aircraft flight performance.

The damaged components were removed and sent to an operator’s repair facility. A National Transportation Safety Board aerospace engineer examined the damaged empennage components as well as their typical installation on exemplar airplanes at the repair facility and compiled a factual report. The report indicated that surface paint was missing in the aft portion of the t-tail fairing near the navigation light assembly. Expanded metal foil was visible on the surface of the fairing where the paint was missing. The paint was not present at the forward end of fairing, where the fairing attached to the upper skin surface of the horizontal stabilizer. Embedded and deformed copper foil was visible where the fastener holes were located and embedded strips of copper sheet were visible in the area where the fastener holes were located. A screw that attached the navigation light lens to the fairing showed significant melting on its head. The upper right navigation light receptacle was very close to the tip of the screw shank inside the fairing. A section of this upper right light receptacle was melted. This receptacle’s housing along with other three navigation light receptacles and the strobe light assembly appeared to be electrically bonded to their back plate. The back plate’s wire harness was burned through at the strain relief shield termination connector backshell. Remains of its melted wire shield could be seen where the rubber environmental boot was attached to the connector backshell. Two bonding jumpers were attached to the navigation light fixture back plate. One of these bonding jumpers showed arc pitting. The fairing’s internal aluminum bulkhead was separated from its attached location and was loose inside the fairing. The wire bundle from the navigation light fixture remained routed through its hole in the loose bulkhead. Some rivets that attached the bulkhead to the fairing had pulled through their holes in the flange of the bulkhead. There was a crack in the bottom pointed end of the bulkhead with scratch marks near the crack.

The flight’s digital flight data recorder was downloaded. The downloaded data showed the location about where the autopilot disengaged. The airplane was about 7,000 feet above mean sea level and was about three miles north-northeast of Chicago Midway International Airport (MDW), Chicago, Illinois.

A NTSB Meteorologist compiled a Meteorological Factual Report for the investigation. The report indicated that the Surface Analysis Chart for 1800 on March 12, 2010 depicted a low-pressure center in eastern Kentucky to the southeast of the accident site with a central pressure of 988-hectopascals (hPa) with an occluded front extending southward from the low. The chart depicted a trough of low pressure extending from Iowa through central Minnesota. The region surrounding the southern tip of Lake Michigan was under overcast skies, with surface winds generally from the north at magnitudes between 5-15 knots. Station models depicted light rain throughout the region, with numerous reports of mist to the north and east of the Chicago area.

A radar mosaic at 1802:30 indicated the majority of Illinois and a large portion of Indiana were experiencing light to moderate rainfall, with several smaller areas of heavy precipitation. Radar reflectivities values in the Chicago area were mainly 10-35 dBZ, with a few localized areas of +35 dBZ.

MDW was located about three miles to the south-southwest of the incident site at an elevation of 620 feet, and was equipped with an Automated Surface Observing Systems (ASOS).

At 1802, MDW reported wind from 030 degrees at 13 knots, visibility of 4 miles, light rain and mist, a broken ceiling at 1,100 AGL and an overcast cloud base at 1,600 feet AGL, temperature 8 degrees Celsius (C), dew point temperature 6 degrees C, altimeter setting 29.63 inches of mercury. Remarks: automated station with precipitation discriminator, hourly precipitation one-hundredth of an inch.

At 1806, MDW reported wind from 010 degrees at 11 knots, visibility of 4 miles, thunderstorms and light rain with mist, a broken ceiling at 1,100 AGL of cumulonimbus type and an overcast cloud base at 1,600 feet AGL, temperature 8 degrees C, dew point temperature 6 degrees C, altimeter setting 29.62 inches of mercury. Remarks: automated station with precipitation discriminator, thunderstorm began at 1805 CST, occasional cloud-to-ground and cloud-to-cloud lightning overhead, thunderstorm overhead moving northwest, hourly precipitation one-hundredth of an inch.

The ASOS at MDW was not equipped with a single site lightning sensor and received its lightning data via the Automated Lightning Detection and Ranging System (ALDARS), which provided lightning information for cloud-to-ground strikes only. In addition, lightning and thunderstorm information may be reported by the human observer with or without regard to data provided by ALDARS. In a fully-automated surface observation, the report will begin to carry “TS” when two cloud-to-ground (CG) strikes occur within 10 miles of MDW within a period of 15 minutes, with at least one of those strikes within a range of 5 miles. In the context of the 1806 MDW SPECI, this would indicate that the second of the necessary two lightning strikes occurred at 1805 CST. Because an observer can add information to the body of the report that is not discernible from the automated reporting information, the aforementioned strategy for deducing lightning time and range information is not applicable. In addition, the report indicated cloud-to-cloud lightning, which is not a product provided to the ASOS by ALDARS, indicating the 1806 SPECI was actively augmented.

The Meteorological Terminal Aviation Routine Weather Report [METAR] at 1851 indicated the thunderstorm ended at 1845, suggesting lightning activity near the airfield well beyond the initial thunderstorm beginning time of 1805. No SPECI was taken when the thunderstorm ended.

ORD was located about 12 miles to the northwest of the incident site at an elevation of 672 feet, and is located approximately 13 miles north-northwest of MDW. ORD was equipped with an ASOS and these reports were issued while an official weather observer was logged into the ASOS system.

At 1751, ORD reported wind from 030 degrees at 10 knots, visibility of 8 miles, a broken ceiling at 1,600 AGL and an overcast cloud base at 2,900 feet AGL, temperature 8 degrees C, dew point temperature 6 degrees C, altimeter setting 29.64 inches of mercury. Remarks: automated station with precipitation discriminator, rain began at 1731 CST and ended at 1740 CST, sea-level pressure of 1003.9 hectopascals (hPa), trace of hourly precipitation, trace of precipitation in the last 6 hours, temperature of 8.3 degrees C and a dew point temperature of 6.1 degrees C, 6-hour maximum temperature 12.8 degrees C, 6-hour minimum temperature 8.3 degrees C, 3-hour pressure rise of 0.1 hPa.

At 1840, ORD reported wind from 010 degrees at 11 knots, visibility of 7 miles, light rain, a broken ceiling at 1,000 AGL and an overcast cloud base at 1,600 feet AGL, temperature 8 degrees C, dew point temperature 7 degrees C, altimeter setting 29.62 inches of mercury. Remarks: automated station with precipitation discriminator, rain began at 1807.

With the proximity of ORD to MDW, no “TS” or “VCTS” reported in the body of the ORD METAR is indicative that no more than one CG lightning strike occurred within ten miles of ORD as reported via ALDARS. It could also indicate that the ALDARS system at ORD was not operating during this time and/or no lightning was observed by the human observer. In addition, lightning within 10 miles of MDW would fall within 30 miles of ORD, making a remark of distant lightning appropriate for a timely report. No such report was remarked.

Atmospheric data retrieved from a rawinsonde launch at 1800 from Davenport, Iowa (DVN), located approximately 128 miles west of the incident site, indicated the lowest 3,000 feet of the atmosphere was nearly neutrally stable, with most remaining portions of the atmosphere through 15,000 feet exhibiting stable conditions. A sharp temperature inversion was identified at 5,000 feet, with a relatively dry layer indicated by the sounding between 5,000 and 7,000 feet. Convective available potential energy (CAPE) values for this atmosphere were zero.

The Lifting Condensation Level was approximately 2,800 feet, and the freezing level was at 4,000 feet. The DVN sounding indicated a 5-knot northeast wind at the surface, veering 8 to 35 knots above 15,000 feet. Several small areas of light/moderate speed and/or directional windshear are noted in the wind profile. Calculations made by the RAwinsonde OBservation program (RAOB) indicated intensities of severe turbulence near 9,000 feet. Layered clouds from 3,000 to 12,000 feet were also identified. RAOB also indicates lightning potential in a layer between 5,000 and 6,000 feet.

A report of cloud-to-ground lightning strokes was obtained from the National Lightning Detection Network (NLDN). The results indicated that within a 15-mile distance from the incident site between 1730 and 1830, the NLDN detected two CG lightning strokes. The NLDN does not detect intracloud (IC) strokes. Neither CG stroke was coincident with the aircraft position at the time of the incident. The NLDN has a median accuracy of 500 meters or better.

A report of lightning strokes was also retrieved from the WeatherBug Total Lightning Network (WTLN). The WTLN reports both CG and IC stroke detection capabilities. The results indicated that within a 15-mile distance from the incident site between 1730 and 1830, the WTLN detected four CG strokes and two IC strokes, with one of the CG strokes occurring within one second and approximately one mile of the incident time and location. The WTLN claims a spatial accuracy of 250 meters or better for CG strokes in this region. Information concerning the spatial accuracy of IC strokes was not available.

There were no Convective SIGMETs issued surrounding the period over Illinois. The closest advisory for thunderstorms was over Indiana, which included a warning for a line of embedded thunderstorms.

The National Weather Service convective outlook indicated that there was a slight risk of severe thunderstorms over southern Indiana and Ohio, eastern Kentucky, eastern Tennessee, and Georgia. There was a general risk of airmass type thunderstorms in the Chicago region.

A meteorological impact statement issued for Chicago airspace at 1535 advised of embedded thunderstorms south of a line that stretched from northeastern Illinois through northern Indiana. The thunderstorms were moving from 150 degrees at 30 knots with maximum storm tops detected at 28,000 feet.

The flight’s dispatch was reviewed and contained weather products that were included in the Meteorological Factual Report. The dispatch did not forecast lightning in the area of the lightning strike about the time of the strike.

Subsequent to the incident, Embraer redesigned the tailcone also known as a Tail Boom Faring. The new design, part number 145-54608-407, incorporates three reinforcement frames internally. The frame installed in the center of the fairing is made of aluminum and designed without lightening holes. It divides and closes the fairing internally. The other frames are made of fiberglass and have an open design. A light assembly is secured to the tip of the tailcone by an aluminum retaining strap. Four diverters are installed on the exterior tailcone skin. The rear tip of each diverter is electrically bonded to the metallic retaining strap of the light assembly and the front tip of each diverter is electrically bonded to the metallic bonding plate installed in each side of the fairing. According to an Embraer technical advisor, the parts and method to comply with the tailcone's redesign are to be listed in Service Bulletin 145-55-0030 titled, "Stabilizers - Improvement in lightning protection of tail boom," with a release date of June 30, 2011.

NTSB Probable Cause

An in-flight encounter with lightning. Contributing to the incident was the damaged and displaced tailcone bulkhead frame that inhibited the full travel of the elevator during approach.