Friday, October 1, 2004
Saving the Osprey
Bell Boeing's V-22 program is back on track with procedures in place to prevent catastrophic flight profiles.
The U.S. Marine Corps MV-22 Osprey program is fully back on track and received a positive review from the Pentagon's Defense Acquisition Board last July--despite three fatal accidents that at one point left it "hanging by its fingertips from a rain gutter 15 stories above the ground," according to V-22 Chief Test Pilot Tom Macdonald.
The aircraft has actually crashed four times, although the first crash on June 11, 1991, was caused by roll rate stabilizing sensors being wired backward. On the maiden flight of V-22 Number Five at Boeing's Wilmington, Del. facility, the aircraft went into uncontrollable oscillation and flipped over. Both pilots, however, walked away from the crash.
The three subsequent crashes proved to be fatal and finally led to the aircraft being grounded in December 2000. The aircraft remained grounded for 17 months while the accidents were studied from every conceivable angle to determine how to prevent future accidents. And in the "that which does not kill me makes me stronger" philosophy, those three accidents have proven to be a major impetus in the V-22 improvement program.
Of the three fatal accidents, the second, occurring in April 2000, is perhaps the most controversial. A contributing cause of that accident has been determined to have been vortex ring state, although the key word is "contributing," according to Macdonald.
While the instruments onboard the fatal aircraft were able to measure airspeed and descent rates, they couldn't determine specifically that the aircraft had entered vortex ring state, Macdonald said. However, the V-22 integrated test team has now conducted a high rate of descent program where they flew the aircraft into vortex ring state. As a result, they have been able to fully scribe the curve of the V-22s performance in regard to entering that condition.
Vortex ring state can occur when a helicopter develops an excessive sink rate with little or no forward speed. At the point where a sustained sink rate velocity (the air going by the helicopter as it descends) approximately equals the velocity of the air drawn downward through its rotors in the opposite direction, the vortices normally shed clear of the rotor blades become entrapped around the rotor in a turbulent, chaotic air mass that essentially prevents the production of rotor thrust. This differs from settling with power in that the latter occurs when the aircraft is at a high hover or slow forward air speed and simply runs out of sufficent power available to produce enough thrust to control or arrest sink rate.
Once recognized, recovery is possible from both situations, assuming the pilot has sufficient altitude to lower the nose and increase forward air speed.
In that accident, at Marana Airport near Phoenix, Ariz., the aircraft was the second in a formation of two involved in an operational test exercise. The two aircraft had arrived at a check point too high and too fast for the approach they were scheduled to execute. As a result, the lead aircraft initiated a rate of descent that exceeded the limitations cited in the flight manual. When the pilot realized his sink rate was excessive for his airspeed, he bumped the nacelles forward slightly to allow him to continue reducing airspeed but with a safer rate of descent.
To avoid overtaking the lead V-22, the pilot of the second aircraft further reduced airspeed and increased his descent rate. Still in the helicopter mode, he reached 2,400 fpm at 25 kt., exceeding the prescribed descent rate limit of 800 ft. below 40 kt. At about 300 ft. AGL, it is believed that one rotor entered vortex ring state while the other was still producing thrust, causing the aircraft to flip over and crash.
However, based on reenactments in the V-22 simulator, it is believed that even if the aircraft had not entered vortex ring state, its rate of descent would not have allowed the pilot to pull out in time to avoid a catastrophic crash.
In effect, the accident occurred when the pilot exceeded the envelope in order to maintain a tight formation with the other aircraft. As a direct result of that accident, significant modifications have been made both to the physical and operational aspects of the aircraft. The Vertical Speed Indicator (VSI) has been redesigned to give the pilot a more accurate reading. The earlier VSI markings went from 0 fpm to 10,000 fpm descent, with only the first 1,000 ft. ticked off at 200-ft. increments. Below that, the markings were too compressed for the pilot to have any kind of accurate reading of his vertical descent rate. The VSI scale has now been decompressed so that the first 2,000 ft. are marked with 200-fpm ticks.
Both the VSI needle and scale have also been programmed to change to red when the aircraft exceeds 800 fpm descent below 40 kt. Along with the warnings on the VSI, the aircraft has visual and audio warning features that alert the pilot when airspeed and sink rate reach the critical point.
The audio warning is a female voice--affecctionately known as "Bitching Barbara--stating "sink, sink, sink" while a red warning message light flashes on the pilots primary flight display. These establish a boundary level alerting the pilot that a critical sink rate is approaching, but does not limit the operation of the aircraft.
While the original airspeed criterion for settling off the warning system was to be 40 kt., the accuracy of the V-22's airspeed indicator allows that to be reduced to 30 kt., Macdonald said. He also noted that the original prohibition of entering an 800 fpm descent below 40 kt. was in earlier versions of the flight manual based purely on legacy helicopter parameters rather than from specific engineering data on the V-22.
As with all performance parameters for aircraft, the actual critical point is based on a "power curve" involving a combination of factors, primarily aircraft weight, airspeed and descent rate. The danger point for entering vortex ring state for the V-22 would therefore normally be considerably below the 800 fpt/40 kt. indicated in the manual, although well above the 2,400 fpm at 25 kt. induced by the pilot of the fated aircraft, Macdonald said. Basically, the current settings now give the pilot a warning well above "the point at which things will start going wrong," he said. "This is a minimum of about 1,000 fpm before (vortex ring state) will happen."
New emergency procedures have also been incorporated into the training manual as a result of the Marana accident. If a pilot feels he is entering vortex ring state, the procedures are to immediately initiate a two-second full-rate beep of the nacelle control thumb wheel switch to move the nacelles forward, followed by adding thrust to the engines to increase forward movement and initiate a climb. The two-second beeping of the nacelle control thumb wheel switch will move the nacelle forward about 15 deg., moving the aircraft more into the airplane mode.
The first fatal crash of a V-22 occurred on July 20, 1992 during a ferry flight from Eglin AFB, Fla. to Newcastle, Del., with a planned stop at Quantico Marine Base in Virginia. During the flight some flammable fluid, most proably proprotor gearbox oil, accumulated in the right nacelle forward of the engine. As the nacelle transitioned upward into helicopter mode, the fluid was dumped backward into the engine, causing an explosion and fire that resulted in a series of power surges and subsequent failure of the engine. The resulting heat melted the interconnecting drive shaft. At that point, the pilot lost control of the aircraft and it crashed into the Potomac River.
As a result of that crash, four major improvements were made.
- More effective fire containment was developed.
- Sensing devices were installed to report any failure of the interconnecting drive shaft to the pilots.
- The crew alerting system was improved to provide prioritized information to the pilots.
- The fluid and electrical lines for the triple redundant hydraulics/flight control systems were segregated.
Prior to the accident, all the flight controls and hydraulics were clustered together going through the trailing edge of the wing. Now two of each system goes through the trailing edge while one of each system goes through the leading edge.
Coincidentally with that first fatal crash, the full scale development phase involving six aircraft ended, showing a number of problems with the aircraft, Macdonald said. The Marines told Bell Boeing to build a better aircraft--stronger, lighter, safer.
That resulted in a totally redesigned V-22B model, with aircraft number seven having its maiden flight in February 1997.
The second fatal accident was the Marana crash in April 2000, followed by the third fatal crash seven months later. That crash occurred at MCAS New River, N.C. in V-22B ship number 18, with two pilots, plus two crew chiefs on board.
While practicing night instrument approaches using night vision goggles, they were turning base leg to final and starting the transition from airplane to helicopter when they experienced a dual hydraulic failure.
"This would have been pretty bad in the airplane mode, but not so bad in the helicopter mode," Macdonald said. They still had a third backup hydraulic system. However, the crew alerts began lighting up like a Christmas tree, advising them of all the problems caused by the hydraulic failure.
At that point one of the pilots punched in the switch that resets the flight control system to verify authenticity of the reported faults. Unknowingly, a flaw in the flight control software caused that switch depression to also neutralize the blade angle in the obscure case of a dual hydraulic system failure in the airplane flight mode.
This abruptly slowed and yawed the aircraft, throwing the pilots forward in their seats. One of the pilots is believed to have hit the switch multple times, trying to find the problem but unwittingly resetting the blade angle, causing violent movement in the cockpit. Eventually, the aircraft stalled and crashed, killing all four personnel on board.
"That accident grounded the program for 17 months," Macdonald said.
The cause of the dual hydraulic failure was determined to have been a rupture of the hydraulic line by an electrical wire rubbing against the tubing in an area of the nacelle that was virtually impossible for the flight engineers or maintenance personnel to inspect.
Because of the complexity of the rotating nacelles, a vast amount of wiring and tubing pass through the wings and nacelles. As a result of the accident, all of the hydraulic lines have been made more robust and the flight control system has been reconfigured to prevent any contact between the electrical wiring and the hydraulic lines. "And the nacelles are being improved even more now," Macdonald said.
Along with a more rigorous flight control system being developed, an improved software package has also been installed.
As a result of the New River crash, the cockpit crew alerting system has now been changed so that the "cascading failures" do not overcrowd displays with needless, redundant detail giving pilots more data than they can assimilate during an emergency. "If you lose a generator, you don't need 15 separate caution messages," Macdonald said. "Now only the key message is sent."
After acknowledging a crew alerting message on the cockpit's central display screen, the pilots can immediately refer to any of five discrete system schematic displays (flight controls/hydraulics, electrical, fuel, engine/drive train and anti-/de-icing) that use intutitive symbols and color coding to help them quickly gain complete situational awareness and "find all the things that are wrong with that system," he said.
With the major redesigns accomplished to prevent future accidents, the V-22 program appears on track for eventual deployment to troops in the field. "OPEVAL," or operational evaluation, is scheduled to take place from January to May of next year, allowing a decision next September on whether to shift from low-rate production to full-rate production.
A series of developmental tests are expected to be completed by the end of the this year
However, this is not to say that minor glitches are not rearing their ugly heads. Pentagon concerns raised during last year's August 2003 acquisition decision memorandum are still not yet fully resolved. These concerns focus mostly on affordability and reliability, and the manufacturers' ability to ramp up quickly to full-rate production. Since the Osprey was grounded four years ago, Bell Boeing has been manufacturing just the minimum number of aircraft necessary to sustain production; that's just 11 V-22s per year.
But if and when the Osprey enters full-rate production, that number will climb quickly to 48 aircraft per year. There remain serious concerns within the Pentagon and V-22 program office about Bell-Boeing's ability to meet this target, considering Bell's involvement in other programs.
However, Bell has stated that it has the assets, the manpower and the equipment to meet the challenge. Its new Amarillo facility is now up and running and it has the executive management team in place to produce the required aircraft once OPVAL is completed and the V-22 goes into full production.
As a company that once produced 200 Hueys a month at the heighth of the Vietnam War, it should be able to do it.