History of the Flight
The aircraft, a Jetstream 4100 registered G-MAJA, was on a scheduled passenger flight from Southampton International Airport to Aberdeen International Airport. The co-pilot was the pilot flying. The first hour was uneventful, flown clear of cloud. The aircraft then entered cloud and an area of airframe icing with light chop and increasing icing. About 20 minutes later, while cruising at FL220 approximately 80 nm north of Newcastle, the commander's Attitude Direction Indicator on the Electronic Flight Instrument System (EFIS) went blank and the autopilot disconnected. The commander re-engaged the autopilot and performed the 'Symbol Generator Failure' checklist, selecting one of the two symbol generators with no effect. Switching the symbol generator reversion switch to the other generator also had no effect. About two minutes after the initial failure, the remaining three EFIS screens went blank. The commander flew the aircraft using the main altimeter and standby instruments. The co-pilot declared a PAN to ATC. After learning that weather to the north was unfavorable, the crew diverted to Newcastle International Airport.
During the diversion, the standby compass appeared stuck, so the crew requested ATC to initiate and stop turns. They descended to 3,000 ft amsl, became visual with the sea and coastline, and about 39 nm from Newcastle, after resetting the avionic master switches, the EFIS screens turned blue and faded into the compass rose, with the speed tape appearing on the ADI. The crew remained clear of cloud over the sea and conducted an uneventful visual approach and landing on Runway 25 at Newcastle.
Weather Information
A Met Office aftercast for the area indicated low pressure over Scotland and a very active frontal system with persistent, heavy rain. Satellite and radar imagery showed embedded convection and medium-level instability, which could lead to high-based cumulonimbus. Lightning was not recorded over the area, but the detection system is optimized for cloud-to-ground lightning and may miss cloud-to-cloud discharges. The instability was likely based at about 10,000 ft, increasing the chance of cloud-to-cloud lightning.
Recorded Data
The aircraft was fitted with a solid-state Flight Data Recorder (FDR) and a 30-minute Cockpit Voice Recorder (CVR). The FDR had recorded over 51 hours of operation; the CVR captured the approach and landing at Newcastle. Although EFIS parameters were not recorded, the FDR data matched the crew's account.
Aircraft Information
G-MAJA was equipped with electronic ADI and HSI, plus standby attitude indicator, altimeter, and ASI. In 1993, Jetstream Aircraft Ltd published Service Bulletin J41-30-007 calling for installation of electrostatic transient absorbers ('transzorbs') in the windscreen heat circuits to prevent EFIS screen blanking from static charge. G-MAJA had been modified accordingly, and the operator applied a 6,000 flying hour life to the transzorbs. The units on G-MAJA had been installed for 3,921 flying hours.
Technical Examination
The aircraft was inspected by the AAIB at Newcastle the day after the incident; no evidence of a lightning strike was found. The flight compartment maintenance panel showed no activated failure indicators. An EFIS system test confirmed normal function. The aircraft manufacturer provided information to test the windscreen anti-ice system transzorbs for failure with residual resistance. Due to lack of test equipment, the transzorbs were removed for further testing. After replacement, a function test of the windscreen anti-ice system was normal. The aircraft returned to service with no further EFIS defects reported.
Laboratory testing confirmed that the left inboard windscreen anti-ice transzorb had failed but retained residual resistance. This failure mode would not be detected by the windscreen anti-ice controller.
Analysis
Failure of any transzorb in the windscreen anti-ice system would increase the possibility of EFIS screen blanking in areas of high electrostatic activity. The system normally detects an open-circuit failure mode, but in G-MAJA the left inboard windscreen transzorbs failed with residual resistance, making the failure dormant and undetectable. The transzorbs had achieved about 65% of their operational life, but the undetectable failure mode could have existed long before the incident. Currently, there is no method for detecting such failures during routine maintenance, hence the operator's life restriction.
During the incident, the standby instrumentation continued to function normally, and the EFIS screens returned to normal after leaving the electrostatic activity area. The aircraft's safety was not compromised.
Conclusion
When flying through an area of electrostatic activity, both pilots' EFIS screens failed, probably due to a failure of the left inboard windscreen transzorbs. The failure mode rendered the transzorb failure undetectable to the windscreen anti-ice controller.