|

F-35B Ready to Begin Stovl Testing

Mar 3, 2009

Tied down over the hover pit, doors open and nozzles deployed, the Joint Strike Fighter is entering its most challenging phase of testing. Over the month ahead, Lockheed Martin will fully exercise the propulsion system in powered-lift mode to clear the F-35B to begin short-takeoff-and-vertical-landing (Stovl) flight testing.

Built around a shaft-driven lift fan, the Stovl propulsion system was a key discriminator in the JSF competition. On a single mission in July 2001, Lockheed Martin’s X-35B concept demonstrator completed a short takeoff, supersonic dash and vertical landing—an unprecedented accomplishment that Boeing’s direct-lift X-32B could not match.

Seven years into development of the F-35, the lift system has been tested only on the ground, and in the aircraft only at low power. In the coming weeks on the hover pit, tests will build up to full power and include manual and automatic conversions between conventional and powered-lift modes.

The F-35B’s lift system includes the aft three-bearing swivel nozzle, which vectors engine thrust downward; roll posts in the wing fed with bypass air diverted from the engine; and the liftfan housed behind the cockpit and driven via shaft and gearbox from the main powerplant. Together they produce a maximum of at least 40,550 lb. of thrust in the hover.

Conversion from CTOL to Stovl mode takes about 15 sec. In the first 6 sec., doors open for the liftfan inlet and exhaust, dorsal auxiliary intake, underwing roll posts and aft nozzle. Over the next 7 sec., the clutch engages to drive the liftfan gearbox. When spun up, solid bolts lock into place to take the torque and unload the clutch. Thrust split between engine and liftfan and roll split between wing posts is then used to control the aircraft in jetborne flight.

The ability to combine supersonic speed and vertical flight in a single airframe is a crucial attribute of the JSF, and while the U.S. Marine Corps, the U.K. Royal Air Force and Royal Navy are in the minority among JSF customers, their need for Stovl capability has been a driver of the program from the outset. The Marines’ urgent need to replace the Harrier has put the F-35B in pole position for flight testing, ahead of the numerically more important conventional takeoff F-35A for the U.S. Air Force.

Aircraft BF-1, the first F-35B, has been on the hover pit before, in May 2008, but at testing was limited to 30% thrust to avoid exciting vibration that caused turbine blade failures in two F135 engines during powered-lift ground tests at Pratt & Whitney. Those tests demonstrated manual and automatic conversion between CTOL and Stovl modes, albeit at low power, says J.D. McFarlan, air vehicle integrated product team lead.

Now the engine has been redesigned, reinstalled in BF-1 and the propulsion and flight control systems upgraded to the latest software release, which will be used from hover pit testing through to the first vertical landing around midyear “unless we learn something and need to change it,” he says.

Designed to collect the exhaust and duct it away from the aircraft to simulate operation in free air, the hover pit at Lockheed Martin’s Fort Worth, Tex., plant is 150 X 75 ft. and 80 ft. of its length is covered with steel grating. Lift-system exhaust flows through the grating, is deflected aft and ejected upward at the rear of the pit.

The aircraft is tied down to struts that connect to the nose and main gear axles and are anchored to the floor of the pit 14 ft. below. Load cells measure forces on the struts in all three axes and allow the overall forces and moments on the aircraft to be calculated and correlated with computer models of the propulsion and control systems.

The majority of testing will be conducted over the open pit, but for the final series of runs steel panels will be fitted over the grating to allow the ground environment during vertical takeoff and landing to be measured. “Our area of interest is under the three-bearing nozzle, which should be similar to a Harrier exhaust,” says McFarlan.

The outwash environment will also be measured to determine the impact on maintainers. In addition, tests over the plated pit will indicate the extent to which hot exhaust gas is recirculated and ingested by the engine, potentially reducing vertical thrust. An inlet rake will measure pressures and temperatures going into the engine.

During the concept demonstration phase it was shown that the lift fan reduces recirculation by creating a “dam” of cooler air that blocks hot exhaust flowing forward from the aft nozzle. “We don’t expect to see any hot gas ingestion on the pit,” says McFarlan, who adds that the inlet rake will remain in place for Stovl flight tests.

Pit tests will begin by stepping through the conversion process—opening the doors; prepositioning, engaging and locking the liftfan clutch; blocking the bypass duct and diverting air to the roll posts; and swinging the three-bearing nozzle down. “We’ll run though these steps manually, then do automatic conversions,” he says.

Propulsion system checkout will follow and include tests to measure forces and moments and time delays between the pilot making a control request and lift system responding; to verify that a particular nozzle angle produces the thrust angle expected; throttle transients and control sweeps to dynamically change thrust and roll split; and demonstrating the control mode to abort conversion, “which we can do at any time,” says McFarlan.

Pit tests are expected to take about a month, and generate the data needed to obtain clearance to fly BF-1 with the lift-system engaged. “We should begin initial Stovl flight work in the April/May timeframe,” says McFarlan. Initial Stovl transition work will be conducted from Fort Worth. After repeating an earlier doors-open flight test to check the updated control software, the lift fan will be engaged.

Flights from Fort Worth will focus on handling qualities and propulsion testing during Stovl transition at speeds from 250 kt. down to 100 kt., all above 20,000 ft. altitude. After a few flights with a KC-135 tanker for aerial refueling qualification, BF-1 will ferry to NAS Patuxent River, Md., in the “early summer” to continue Stovl testing at lower and lower speeds and altitudes.

Tests there will begin with Stovl-mode rolling landings at speeds from 130 kt. down to 70 kt. The F-35B will then transition from semi-jetborne to jetborne flight at speeds from 60 kt. down to 40 kt. where it will rely totally on the propulsion system for lift. The first hovers will be conducted in free air at several hundred feet for station-keeping work before proceeding to the first vertical landing on Patuxent River’s hover pit.

McFarlan says BF-1 will complete “probably a dozen” flights between arriving at Patuxent River and performing the first vertical landing, which is expected to be accomplished in June-July. This will be followed by about a year of envelope-expansion flying at the Navy’s flight test center, which will include Stovl operation in crosswinds and tailwinds, and with and without stores.

Loaded with flight-test instrumentation, BF-1 weighs in close to an operational F-35B and will have a similar thrust margin in vertical flight, says McFarlan. With a hover weight of 39,000 lb. and thrust of 40,550 lb., the F-35B is expected to operate at a nominal maximum hover weight ratio of 97%, providing a thrust margin of 3%. “BF-1 will probably have a little more margin,” he says.

Weight is not the issue it once was for the F-35B. “We are right on track to the weight plan set forth on exit from the SWAT phase,” says McFarlan. SWAT—the 18-month Stovl Weight Attack Team redesign completed in 2004—reduced the F-35’s weight by more than 3,000 lb. “We are no longer approving design changes for weight reduction—we don’t need to,” he says. The emphasis is on design improvements for affordability, with weight reduction seen as a bonus.

Lockheed Martin is waiting to measure the F135’s installed thrust on the hover pit before deciding whether any further weight reduction is needed, but with 40,550 lb. as the minimum vertical thrust expected from any engine in the fleet, McFarlan is confident the F-35B will meet its Stovl performance requirements.

Pratt & Whitney shipped the first redesigned Stovl flight-test engine to Lockheed Martin at the end of January, meeting the schedule agreed last spring following the turbine blade failures. “They did a good job getting the propulsion system to us on time,” he says.

While the turbine-blade problem delayed the start of powered-lift testing by four to six months, it has not extended the overall flight-test program, says Bill Gostic, vice president F135 program. The engine, meanwhile, is meeting its requirements for thrust, temperature and stall margin. The CTOL powerplant is on its weight target and the Stovl version is 6 lb. over, he says.

The F135 has its origins in the F119 engine powering Lockheed Martin’s F-22, with a larger fan and additional turbine stage to provide higher thrust. The increased-diameter two-stage fan has redesigned aerodynamics and the first stage is an integrally bladed rotor with hollow blades inertia-welded to the hub.

Compared with the F119, there are aerodynamic tweaks, material changes and manufacturing improvements in the compressor, which uses blisks throughout. The combustor has the same aerodynamics, with more advanced cooling allowing higher temperatures.

The single-stage high-pressure turbine has improved cooling, but the aerodynamics of the first and second turbine stages are unchanged from the F119. The biggest change is the addition of a third turbine stage to drive the liftfan. Pratt opted for an unshrouded third stage, Gostic says, because it was impractical to produce a cooled and shrouded single-crystal blade of the size required.

That design decision came under scrutiny after third-stage turbine blades failed in two Stovl engines during powered-lift ground testing, one in August 2007 and one in February 2008. The cause was a fatigue crack that initiated at the same internal cooling hole in both blades.

“The root cause was high-cycle vibratory stress,” says Gostic. “The pressure wakes from the third-stage vanes upstream of the turbine were exciting the blades.” The result was a stress concentration at a cooling hole near the base of a rib inside the blade close to its leading edge. The wakes are stronger in Stovl mode because the turbine is working harder to drive liftfan.

“The pressure ratio in CTOL is very different,” says Gostic. “We extract much more energy in Stovl and the pressure drop is much greater. This results in a higher vibratory response.” Increasing engine speed by 50-100 rpm. to get more Stovl thrust as part of the SWAT redesign effort also exacerbated the vibration excitation problem.

Pratt had not seen the problem before because the F119 has much smaller turbine blades and does not have the F135’s variable cycle. “There has never been a turbine blade like this that can instantly go from up-and-away flight to producing 28,000 shp.,” says Gostic. “Engage the liftfan and you effectively double the size of the fan. There is a sudden increase in load on the turbine.”

The fix has been to introduce asymmetry to disrupt the excitation causing the vibration. Asymmetry is already used in the F135 fan, alternate blades having different shapes. In the redesigned third turbine stage, instead of 54 evenly spaced vanes the lower half now has 28, the upper half 24. “This breaks up the timing of the excitation,” says Gostic.

The second part to the fix is to redesign the turbine blade itself to get rid of the stress concentration. This is achieved by eliminating the holes where the crack started, so this high-stress area is now solid and cooling air is redirected through slots above and below.

The fix was evaluated first in a ground-test engine with redesigned vanes and unchanged third-stage rotor. This showed a reduction in vibratory stress of up to 48% with just the vane redesign alone. “It gave us sufficient margin to survive with the old blade,” Gostic says. Another engine test over 500 tactical cycles showed no change to cooling effectiveness.

A redesigned flight-test engine was then subjected to a proof test intended to uncover any susceptibility to high-cycle fatigue. Previous tests involved stepping up engine speed in 50-rpm. increments and holding each test point for 10 X 6 vibration cycles. In the proof test devised for the F135, speed is increased in 25-rpm. steps from idle to the 9,000-rpm. redline and each is held for 10 X 7 cycles—the test takes about 30 hr.

“It’s an unprecedented proof test, 2-10 times more difficult than any previous test,” says Gostic, explaining that “high-cycle fatigue modes are peaky and otherwise we could miss them.” Pratt plans to proof test “a couple more” Stovl engines to be sure the fix is working.

Rolls-Royce provides the lift system components, meanwhile, and is testing elements of a lighter system for the initial service release (ISR) powerplant for production F-35Bs. The company plans to conduct system-level tests at Indianapolis in third-quarter 2009 before handing over to Pratt & Whitney for propulsion system tests. The Block 3.5 final production standard will incorporate a hollow second-stage ‘blisked’ liftfan rotor; hollowed-out first-stage rotor; titanium instead of aluminum liftfan case; and inlet guide-vane ice protection system.

These improvements cut an additional 39 lb. from the Block 3 development configuration, itself 150 lb. lighter than the original lift-system design. The complete system includes the three-bearing swivel nozzle, roll posts, gearbox, clutch, driveshaft and variable-area vane-box nozzle (VAVBN). Deliveries of the production-standard VAVBN have already begun to Northrop Grumman, which builds the F-35B forward fuselage.

The VAVBN vectors the lift-fan exit flow as well as back-pressuring the fan to control stall margin and is an integral part of the Stovl aircraft’s structure and one of the earliest modules to be delivered. By contrast the liftfan is almost the last to be delivered. “We won’t be delivering the first [production] liftfan until March 2010,” says Rolls-Royce Defense North America President Dennis Jarvi.

Six lift systems will be produced for the low-rate initial production Lot 2 (LRIP-2) batch of F-35Bs, with deliveries continuing through 2010. Rolls expects to finalize details of the 12-shipset LRIP-3 contract imminently. This calls for delivery of the first VAVBN in June 2009, and first lift-fan module in September 2010. Discussions have begun on long-lead items for LRIP-4, which is to include the first U.K. buy. “The rate is starting to move up,” he says.

While cost continues to be a concern across the F-35 program, Jarvi says Rolls-Royce is “making a lot of progress” on its part of the propulsion system. Further cost-reduction initiatives are planned, with potential changes to some materials and manufacturing tolerances. “We have identified some funding to do additional cost-reduction work and make it more affordable,” he says.

Rolls-Royce has also studied several advanced materials for the liftfan clutch aimed at ensuring it meets the required life of 1,500 engagements. While ultimately only flight tests will prove the durability of the present design, which uses aerospace-standard carbon-carbon brake material, “we still believe we will want to go after improved clutch life,” says Rolls-Royce liftfan program director Gregg Pyers.

Pratt is projecting a 1,350-engagement lift for a “minimum” clutch, says McFarlan. This is based on a usage spectrum that includes engagements in ground idle, in flight at various engine speeds, and while decelerating, maneuvering and in other stressing conditions. While research is underway on advanced materials, Gostic says these could be more specialized and expensive than the material now used. Another option, he says, is to monitor how the clutch is actually used in flight tests and see whether it could get to 1,500 engagements without change.

In 2006, Rolls surveyed more than 30 materials and narrowed the list to three finalists. “We’d like to do further experiments and downselect to one,” says Pyers, who adds the finalists include hybrids of existing brake materials.

Photo: Lockheed Martin



◄ Share this news!

Bookmark and Share

Advertisement







The Manhattan Reporter

Recently Added

Recently Commented