AlliedSignal's new version of TFE731 updates an engine which has been around since 1969

Guy Norris/PHOENIX

In 1969, Garrett-AirResearch began development of a purpose-designed business-jet engine, the TFE731. Twenty-eight years later, with almost 30 million operating hours accumulated and nearly 8,000 engines produced, Garrett's successor, AlliedSignal Propulsion Engines, is introducing the third of a family of new-generation successors to that popular series, the TFE731-60.

Like its new-generation -20 and -40 siblings, the -60 shares the basic two-spool geared turbofan architecture of the original TFE731-2/3/4 and -5 series. This is where the similarity ends, however. Practically everything else about the engine is completely new, radically enhanced, altered or redesigned - a point driven home by Business Aviation engineering manager Karl Johnson, who says that "-this is definitely not a 'warmed over' -731. So much has changed that it is more heritage than derivative".

AlliedSignal goes to great lengths to underscore the "new" nature of the 731-20/40/60 family for tactical and strategic reasons. The decision to base the design on the well-known -731 series rather than adopt an all-new centreline was not easy. AlliedSignal knew that the competition would seize on it eagerly to denounce the family as "money for old rope", despite the fact that 80% of the original -731 parts have been redesigned and improved. Because the derivative approach was driven mainly by the airframe manufacturers and users, however, the company knew that it was on safe ground. As Dominique Hedon, vice-president, commercial propulsion, says: "It fits in with the evolution of the marketplace. In former days, performance was top. Now, business aviation is a tool, not a toy." The sound business case for adopting a derivative approach, and the subsequent introduction of three new versions of the engine "-proves that AlliedSignal is fully dedicated to business aviation. It really is the core of our business," he adds.

 

Back to basics

The design roots of the new series go back to 1991 when AlliedSignal began conducting "voice of the customer [VOC]" surveys with operators, airframers and "all aspects of the business-aviation community", says TFE731-20/40/60 programmes manager, Brian Sill. The time was ripe to begin planning for the future, as the uprated 731-5B had just been certificated for the Dassault Falcon 900B and Falcon 20 retrofit, and certification of the 731-4 for the Cessna Citation VII was almost completed. Manufacturers were also making rumblings about new business jets and Learjet, a major user of the TFE731 on the Learjet 31, 35 and 36, was in the early stages of designing what was soon to become the Learjet 45.

"We tried looking at a new standard of engine and used the VOC process to see if we could come up with upgrades of the existing powerplants," says business aviation programme director Richard Horvath. "Then we formed a cross-functional team made up of marketing, engineering and pilot-advisor groups to look at a new type of engine. We asked everybody: 'What would you like to see in terms of performance and cost of ownership?' We also looked at the competition and asked: 'what is the differential'?". The answer - in every case - was cost of ownership, a theme which would come to dominate the engine's development.

AlliedSignal's VOC dovetailed with the onset of an era of austerity in the business-aviation community, so the emphasis on cost, maintainability and reliability, rather than outright performance, was perhaps not unexpected. Business aviation customer support manager Ralph Alberto recalls: "We were hearing the same message from various groups, and there was a deliberate attempt to capture that and drive it into the design from the beginning."

It quickly dawned on the team that the goals derived from the VOC would be met by developing a family of engines around the original common core. To meet the extra performance needs, as well as the vital low-cost targets, the team also planned a vast array of enhancements. "We said: 'Let's apply technology in a low-risk way'. It allowed us to get these new engines certificated with extremely high inspection intervals, and high performance," says Horvath.

From the start, therefore, the design team attacked two key parameters of the engine cost equation: the major periodic inspection interval (MPI) and compressor-zone inspection interval (CZI). By combining tried and tested designs with advanced materials and design technology, AlliedSignal aimed for a 2,500h MPI and a 5,000h CZI "out of the box". This was a radical advance over the previous generation, as Sill explains. "After a few years in service, the -5B was finally at 2,100h MPI, and here we were right at entry-into-service going in with 2,500h," he says. Based on field experience, the fan-gearbox inspection interval was extended to the CZI. This helped minimise the work scope of the MPI, making it a "real hot-section" inspection, adds Sill.

The first member of the new generation became the TFE731-20 and was adopted in 1992 by Bombardier for its Learjet 45. "That really started it rolling, but we were in concert talking to Dassault and IAI [Israel Aircraft Industries]," says Sill. "The core was therefore designed to provide as much commonality as possible from the start. Because of the way we'd structured the core, we could quickly adapt to new requirements," he adds. This approach ideally suited the significant retrofit/upgrade market which had been identified by the marketeers. "In effect, this led to the [IAI] Astra SPX moving from the -3 to the -40, the Falcon moving from the -3 to the -40 and the Falcon EX from the -5B to the -60," says Horvath.

As a natural evolution from the VOC process, AlliedSignal invoked an integrated production, delivery and support method to move the engine in six phases from drawing board to customer. The steps involved were concept definition, design, development, certification and testing, entry into service and, finally, support. Like everything else about the new generation -731s, the focus on reducing cost of ownership resulted in a wide variety of support measures, ranging from a five-year warranty with no pro rata charges to the setting up of an international support network supplied with spare or "bank" engines.

"Our preparation for these engines is unprecedented," says Alberto. "We have over 20 authorised centres for the new family and rental, or bank engines, in place". By the end of September, the firm hoped to have eight -60s, seven -40s and four -20s in place worldwide.

Another initiative with a striking similarity to tests being conducted on new auxiliary power-units (APUs) by the APU side of the house was the accelerated mission test (AMT). Jim Kidwell, business-aviation engineering director, explains: "We're taking a pro-active approach. We're going to test things before we go out to the field. It was one of the results of the VOC. Customers said they didn't want to be the testbed for any modifications. Also, it means our confidence level in resolving any issues that crop up is higher, and it helps us set up a precedent with these aggressive warranty levels."

Each AMT consists of a fully instrumented engine running for 2,000 cycles, with each cycle lasting around 20min, to simulate the full spectrum of conditions from "light-off", through taxiing and take-off to cruise, flight idle and thrust reversing to shut down. "When we ran the first AMT in 1993, we discovered a durability problem with the high-pressure [HP] turbine nozzle cooling well ahead of time," says Johnson, who adds that "-the part was redesigned without affecting the schedule". The AMTs helped test 27 design enhancements made to reinforce durability and a further 42 improvements to enhance maintainability.

The pilot community was not forgotten in the creation of the new family. Operability was made easier with the introduction of N1 (low-pressure spool) control channelled through a digital electronic engine-control (DEEC) unit. The DEEC is equipped with a hydromechanical back-up and ARINC 429 glass-cockpit interface.

"The operational improvements are significant," says the Pilot Advisor Office's Burnell Rundall. "If the pilot were to step from one to another, the differences would seem invisible. It would still feel like a -731, but the additional features would make control much easier. For take-off, for example, you now push the throttles to the stops but, in the older engine, you'd get some 1-2% of N1 overshoot before it would roll back. The Mach number input [via the DEEC] provides very precise N1 control."

The -60 DEEC also hosts an engine-condition trend-monitoring (ECTM) system which automatically records engine parameters and flags up exceedences and other anomalies. The ECTM system was introduced as a direct result of the VOC and is playing a vital role in reducing cost of ownership. Data from this system not only help mechanics in troubleshooting problems and in planning periodic maintenance, but it will be used to validate a proposed move to even longer inspection intervals in future. It also helps increase confidence levels and therefore forms a key part of AlliedSignal's aggressive power by the hour, or maintenance-service plan. Base rates for the -60 within North America have been set at less than $120/h compared to around $125/h for both the lower-thrust -5A and -5B engines.

 

Inside the -60

The most obvious external feature of the -60 is the 770mm-diameter wide-chord fan. According to Johnson, the distinctly swept and snubberless (or clapperless) solid-titanium fan blades are "one step beyond" the fan design used on the TFE1042 engine developed for Taiwan's Ching-Kuo Indigenous Defensive Fighter.

The -60 fan diameter is almost 70mm larger than those of the -20 and -40 engines which use a fan design based on that of the -5B. As a result, the larger -60 has a bypass ratio of 3.9, compared with 3.2 for the -5B, and a pressure ratio of 1.7. The fan produces a 16% greater specific airflow than the fan used in the earlier engine, and therefore plays a key part in achieving a 17% increase in cruise-altitude thrust against the older -5A. Against the advanced fan of the -5B, the -60 scores a 6.6% improvement in thrust at cruise altitude. Thermodynamic thrust capability at sea level is 24.5kN (5,500lb) with standard take-off thrust (sea level, static) rated at 22.25kN to 32 degreesC - a 5.3% improvement over the -5B.

The energy of the large fan proved too great for the original all-titanium containment-collar design during blade-out tests in the spin pit. "I'm really glad we did those tests in the pit," adds Johnson, who says that the revised design is made of a solid-titanium ring wrapped with Kevlar. The smaller -20/40 containment systems are constructed of titanium only.

A key design change from earlier engines in the series is found in the epicyclic fan gearbox, where the spur gears have been replaced with helical gears, and there is now a single-piece sun gear. "This allows us to carry a higher torque load in the same size package. It also reduces part count and eliminates a bearing," says Johnson. The redesign also makes the gearbox quieter and less prone to vibration.

Some of the most important material changes made in the -60 are found in the gearbox. The ring and planet gears are made from CBS600 alloy, a stronger material than the 9310 used in the problematic gear assembly of the -5, as well as in the fan shaft. The -60 fan shaft is made from C250 alloy. The gearbox casting has also been changed from magnesium, which is prone to corrosion, to aluminium. An oil tube, which, on the -2/3 and -4 engines could only be reached by removal of the entire gearbox, has been made more accessible by being built into the cavity casting.

Aft of the fan and abeam the gearbox are the sharply raked, composite, fan stators. The leading edges of the stators are protected from abrasion and foreign-object damage by foil wrappings. The Kevlar core of each stator is split near the attachment point on the bypass duct to ensure that, in the event of a birdstrike, it will break off cleanly and pass through the duct without shedding debris into the core. For ease of maintenance, the stators can be individually removed.

The outer lining of the core section just aft of the fan frame is made up of a four-section, segmented, oil cooler (three sections on the -20/40). The cooling fins on the -60 are smaller than those on the earlier engines, and the gaps between the sections are sealed to prevent leakage. Johnson says that the smaller fins help clean up the bypass flow path without any significant degradation of cooling performance.

The four-stage low-pressure compressor (LPC) is relatively unchanged, with the exception of a move to "beartooth" fittings at the base of the blades. The cleaner design gives "significant" performance improvement, says Johnson, who adds that the enhancement is being fed back into the -2/3 and -4 engines. All of the LPC discs, blades and stators are made from titanium, with the exception of the fourth-stage stator, which is made from Inconel 718. The first-stage LPC rotor shroud is made from aluminium, while the rest are of titanium.

A centrifugal HP compressor, running on a separate Inconel 718 shaft, is located aft of the LPC. The HP compressor was the focus for much of the improvement and is primarily responsible for a 33% increase in cycle pressure ratio generated by the enhanced core. This, in turn, contributes most to a 12% reduction in cruise thrust-specific fuel consumption (TSFC) relative to the -5. "The compressor was really low-risk technology as it was a spin-off from the T800 [turboshaft] and other research-and-development work," says Horvath.

Inconel 718 is used throughout the flow path and makes up the impeller shroud and HP diffuser. The "de-swirl" is also made of 718, like its predecessor, but the design is "significantly different", says Johnson, as is the impeller, which has a higher pressure ratio largely because of a change in the shape of the bore. "We've also changed the de-swirl vanes to trap completely the energy in the flow path. This is the most valuable air in the engine because it has done all the work it's going to do at this point."

The change to the de-swirl vanes design also forced AlliedSignal to alter the configuration of the plenum. The more rigid design of the vane attachment meant that the plenum could no longer be "spring loaded" into position. The whole section is therefore cast in one piece, helping to remove another area where locking wire was previously needed. "We were on a quest to get rid of lockwire on this engine," says Johnson, who adds that all of the maintainability enhancements result in a 20% reduction in overall line-maintenance time per flight hour.

The compressed air is forced into the combustor, which is redesigned, with a changed orifice pattern and fuel nozzles located in "tailored" positions to suit the thermal flow patterns. The combustor is otherwise conventional and constructed of stacked rings made from Haynes 188, a cobalt-based high-temperature-resistant material. The rings are welded together to allow the highest possible operating temperatures. "We want to be able to get a 175íF [80íC] margin and well beyond," says Johnson.

The combustor outer transition liner is effusion cooled, representing AlliedSignal's first use of this technology on a commercial turbofan. The fuel nozzles are individually replaceable and feature solid-state ignition controlled by the DEEC.

Hot gases pass from the combustor and HP diffuser through a tangential on-board injector (TOBI). This spins the gases to match the swirling of the seal rotor, producing better efÌciencies and helping reduce TSFC. The TOBI is made from Inconel 718, while much of the downstream turbine flow path, including the centre coupling and seal rotor disc, is made from nickel-based Waspalloy. The HP turbine (HPT) stator vane is made from another nickel-based alloy, Mar-M 247.

A further important material change was the switch from Mar-M 247 to single-crystal SC180 material for the single-stage HPT blades. The higher-temperature-resistant material was also introduced into the first stage of the three-stage low-pressure turbine (LPT). Mar-M 247 is still used in the first-stage LPT stators and second-stage nozzle, however. The third-stage LPT nozzle is made from Airesist 319, a modified version of an alloy developed originally by the old Garrett AirResearch company. Stage two and three LPT blades are made from Inconel 100 and Airesist 713, respectively.

An important design feature of the LPT is the case-tightened shroud. This wraps around LPT stages one and two and is made of a honeycomb material. "With a cooler case, as the engine gets hot we get a tighter Ìt. We're not using active clearance control because we don't really need to," adds Johnson.

Hot exhaust gases are ejected towards the mixer nozzle through the rear supports of the turbine frame, which are angled to "-take some of the swirl out of the air", he adds. Located in this section is the N1 monopole, or fan-speed sensor. The revised monopole can be quickly and easily removed by detaching two screws, rather than the previous method on the -5 which involved detaching the entire aft body to get at the unit.

 

Into service

The 100th-engine barrier was broken in August 1997 and, by the end of the year, around 20 -60-powered Falcon 900EXs will be in service. By early September, more than 7,500 fleet hours had been amassed, more than with previous new types at this stage because of the longer-range routes flown with almost airline regularity.

Some issues have cropped up, and AlliedSignal's scramble to solve them seems to have convinced Falcon owners that the VOC pledges were sincere. One of the most serious concerned an auto-ignition/lean-blow-out issue, which resulted in hung starts in cold weather and one in-flight flame-out.

AlliedSignal strapped a -60 to the side of its Boeing 720 testbed and flew it to Alaska, where revised DEEC software was evaluated. The new software was certificated by the US Federal Aviation Administration in August and the whole fleet was due to be retrofitted by the start of this year's National Business Aviation Association meeting in Dallas, Texas.

Other problems relating to the breather pressurisation valve and turbine temperature sensor harness are now in the process of being "resolved", says the company.

Source: Flight International