Power Plant - The UltimateL39 TFE731-3 Conversion

Email Questions or Comments

This is kind of an interesting part of the project and one that enables everything else.  The conversion was very straight forward as you will see from the simplicity of the elements needed.  Behind that simplicity there was a lot of thought, time, energy, documentation and tooling.  That effort is not required for doing a one off but I suspect this solution will prove itself to be the most elegant and cost effective approach to keeping L39s in the air.  A bold statement to be sure but initial results are promising.

The TFE731-3 is used in several business class jets.  A lot of these airframes are getting to the point where they are no longer economically viable and thus are coming out of service.  Each one retired frees up two engines which would ordinarily be out of the realm of possibility for this type of conversation on cost alone.  The availability of the -3 motors has exceeded the demand potentially making them a cost effective alternative to the L39's AI-25.  This assumes the L39 owner is accounting for the approximately $200 an hour set aside for overhauling an AI-25.  Most do not as this cost has been masked in the past by the availability of surplus motors from around the world.  Those motors are being consumed and calendar limited components within the fuel control and other items are becoming an issue thus costs to operate the L39 are going up.  This has grounded many aircraft in the US.  The TFE731-3 has an initial cost to install then an ongoing cost to own and operate.  Acquisition costs for good motors that have been on Honeywell's MSP program are less than half that of the cost per hour for the AI-25.  Operating costs, namely fuel, are also significantly less than the AI-25.  The one time installation/conversion costs are more than reasonable and returned in less than half an AI-25 refresh cycle.  I found the 731-3 conversion to be compelling even before considering the improvements in flight and safety.

The following is a bucket load of pictures.  The general areas covered are -

Front Motor Mounts.        These were challenging.  Business jets hang their motors on one side and have an articulating link on one of the mounting points.  In the L39, the front mounts take thrust and gyroscopic loads so the mounts must be hard fixed to the motor (or compressed rubber mounted like the L-139 and its TFE731-4 installation).  Bolting a solid mount across a span of over twenty inches drives you to carefully consider thermal expansion to make sure there are not monstrous stresses going from a hot ramp in Arizona to the bitter cold of FL28.  Using Titanium like the motor center housing you are bolting to is the easiest solution.  It is also the lightest but then less weight costs money.  The other option was 410 Stainless Steel as its expansion coefficient is close enough to that of Ti to make the stresses reasonable.  The Ti mounts below were wire EDM'd from a solid block then finish CnC machined to shape.  The 410 SS parts were finish machined from a casting.  Doubling the cost saves you 18 - 10 or 8 lbs plus another three pounds of ballast in the nose.

Rear Motor Mounts.       The L-139 moves the rear motor mounts from the stock L39 position.  I wanted to avoid modifying the engine mounting structure so that an AI-25 could be rolled back into the airframe with just an adapter harness for the Garmin G3X system.  The most straight forward way to do this was to use the original AI-25 rear mounts and make provisions to attach them to the fan duct rear flange area.  If you look closely at the pictures, you can see the uni-directional material that was included in the composite lay-up to transfer loads from the rear mount location to the front duct to motor mounting flange.  These strips leave at about a 35 degree angle.

Inlet Cone with "Velocity Stack" Locating Ring     Controlling airflow into the front of a high bypass turbofan engine is important.  Any turbulence around the periphery of the inlet changes the loading on the blade tips as they rotate in their circular path.  Changing the loading each revolution is a high cyclic rate fatigue issue and one none of us wants to face.  Composites offer a tremendous advantage in obtaining complex shapes with smooth transitions.  Composite structures also have a weight advantage along with avoiding the fatigue issues described below..

Fan Duct  This is the most complex part and it is where the project started.  The original approach was to hand form aluminum quarter panels then weld the pieces together to form a complete duct.  Manufacturers have the luxury of drop form tooling but small run production or one offs simply can not support the tooling costs.  The first quarter panel took two weeks of hard work.  It became clear that hand forming several fan ducts was not going to fly (literally).  In addition, we had concerns about cold work hardened aluminum panels being welded together.  The welding would anneal the parent material in the vicinity of the weld.  Unless you heat treat the entire assembly, fatigue cracking will likely occur at the interface in material hardness.  If you've ever tried to heat treat a large thin wall structure then you know what it is like to create potato chips.   The solution was a carbon composite structure but this too had its challenges.  First in my mind was temperature.  The duct does not carry much load so strength was not a concern.  I took inspiration from the Rolls Royce motors in the Gulfstream 550.  They use carbon composites for both the fan duct and the thrust reversers.  The secret is using a high temperature resin system.  Fan duct temps on shutdown with thermal soak were reasonable.  Fan duct temps due to fan compression at full power on a hot day were more but still reasonable. These temps were well under 200 degF for a part that was free standing post cured, in part, at 350 degF for two hours.  I think we got the resin system just right.  Pictures from inside the duct as installed shows the composite aero fairing used to exit the ACM bleed air.  We chose to go straight down and use the AI-25 bleed air valve to make things simple.  This plumbs directly into the ACM bleed air line prior to the air cleaner so all L39 flow control and over pressure features remain.

At the bottom of the pictures below I've included some of the engine in airframe.  Keep in mind that a lot of the elements are first article bits for development and proof of concept.  The throttle bell crank for example uses two full disks so that different motion ratios can be tested.  The best will be chosen and a normal two arm style bell crank will be fabricated.  I'm already working on the final molds for the throttle and bleed air mounts.  In short, some items need, and are getting, a little polish.

And polish they have received....  I've added some pictures towards the bottom of the latest throttle bellcrank and fan duct aero faring.  These are now of production quality.  In addition, there is a picture of the adapter for mounting the AI-25 bleed air valve to the Turbo Cooler's air cleaner so that the original bleed air switching components can be used.

The integrated HP/LP mixed bleed air plumbing solution is taking shape.  We anticipate this will be the working solution for conversions moving forward.  It uses an AI-25 bleed air valve for the Turbo Cooler shut off while using a TFE731 (fails open) valve for Anti-Ice control.

Currently chewing on the idea of using a TFE731-5BR.  Chewing (as of mid 12/18) has now become doing.


 

What came out to make room for the TFE731

Front Motor Mount

The Prototype

6al4v Titanium Motor Mount

10 lbs v. 18 lbs for 410 Stainless

The back side of that Ti Front Mount

410 Stainless Motor Mount

18 lbs v. 10 lbs for Ti

The Pin that holds the stock Roller

Rear Motor Mount

Ed's initial inlet modeling based on L-139 drawings

The Inlet Cone with "Velocity Stack" Guide

Cone on Y duct

Cone in Plane

What the air sees

What the air sees

My favorite

The Concept

First Mock Up of the Fan Duct

Part under vacuum

Fresh out still on the mold

Note the strips going up at an angle - These are the re-inforcing Uni-Directional plies

The guy who built the last part

Viewing the Fan Duct from the inside

Bleed Air aero fairing - downward exit

One of the Ti front motor mounts doing its thing

A look out the back from the motor through the tail

ACM bleed air through the bottom

That development throttle linkage I mentioned above

You can not forget the EPA can

Rolling in for first start

Exit cone.... My first real aircraft sheetmetal effort

Production fan duct aero faring

Adapter for mounting the AI-25's bleed air valve to the Trubo Cooler air cleaner

Carbon Mast throttle bellcrank

The following pictures show mixed HP/LP bleed air plumbing to the Turbo Cooler and Inlet Anti-Ice

How mixed HP/LP exits the mixing manifold