In a summary, these are the critical features that allow us to ensure that every customer maximizes the return on their turbine investment. To better describe the bearing alignment features of this engine over my first post, I have added a few simple diagrams and modified the text to suit.

As mentioned before, the majority of this material copied directly from the user manual supplied with every TJT-3000 engine, to be downloadable for www.TJT.bz. The manual also provides an extensive safety outline and a short educational section for those interested in learning the basics of turbine operation. It does not serve as a prerequisite for a high-level thermodynamics course, but with the use of flow diagrams it accurately describes the theory of operation behind turbo-jets, turbo-shafts, turbo-props and turbo-fans. Although most modelers are already up to date on this technology, there is always a few that are not. We strongly believe that turbine education is critical to both TJT and its customers by allowing all of us to communicate more efficiently.

(1) Self-aligning Bearing Configuration

Correct bearing alignment design principles are the single largest benefit of the TJT-3000.

Many current turbine designs operate on simple variations of the original KJ-66 design. In this design the front bearing is simply press fit in to the backside of the aluminum diffuser and the rear bearing supported in the shaft tunnel near the rear of the engine. The rear bearing is assumed to be concentric to the front bearing resulting in an over constrained bearing configuration. Press fitting either of the shaft bearings into any fixed component automatically induces a combination of run out and non-concentricity between bearings. Both conditions are exaggerated in the attached diagram. In reality the variations are on the order of several microns, invisible to the human eye but not to a ceramic ball bearing.

The diagram to the upper-left is a visualization of non-concentric bearings (cut in half). The axes of the bearings are parallel yet offset, causing each ball to bind at the top and bottom of its travel and running loosed elsewhere. The binding action causes enormous stresses in the inner and outer bearing race grooves, destroying the surface finish. Likewise, the loose portion of the ceramic ball’s travel can cause them to slap together on a microscopic level, also affecting the bearings lifetime by wearing the ball’s surface. The upper-right diagram represents a bearing orientation due to run out in 1 or more of the bearing support components. Run out occurs when a planar machined surface is not accurately flat, skewing the position of any component constrained to that surface. In this case the effect on the bearings is very similar to non-concentricity, making the combination of the two very detrimental to bearing lifetime.

Press fitting bearings into aluminum components has also been eliminated in the TJT-3000’s design because of the extreme tolerances in achieving the correct bearing fit at all operating temperatures. For example, as an aluminum diffuser heats up it will expand in every direction, which loosens the fit on the front bearing. To accommodate for this the bore can be slightly reduced, achieving a solid press fit at high temperature and unfortunately an over press fit at room temperature. An over press fit will easily destroy a bearing far before its specified lifetime. While it is possible to achieve the correct press fit tolerance for all temperatures, the increased manufacturing cost and component scrap rate would not be sustainable at a reasonable cost to TJT or its customers. If a bearing’s outer race ever becomes loose it will begin to spin immediately, causing extensive damage to the engine. Many variations of the original bearing support scheme have been attempted with new preload directions and the incorporation of o-rings or wave washers to limit bearing spin, none of which stand up to the quality requirements set by TJT.

TJT has developed a radical new solution to these design hurdles in their ball bearing supported carrier system, shown in the lower portion of the attached diagram. The TJT-3000’s bearings are both press fit into stainless steel bearing carriers, preloaded with off the shelf wave washers (bearing preload shown by arrows) and self-aligning as soon as the shaft nuts are tightened. The multiple degrees of freedom allow each bearing to operate with absolutely no effects from run out or non-concentricity what so ever. Each ball remains is contact with the inner and outer bearing race al all times. Testing results and customer feedback indicate a dramatic increase in bearing life with this system.

The stainless steel carriers also expand and contract much less than aluminum components under temperature variations. The press-fit tolerance issues of other bearing support methods are now totally insignificant in this design.

“Woketman”, your point is valid and thanks for raising that theory… If we assume that the bearing outer race’s material is similar in composition to the stainless steel in the carrier, the expansion rates may be nearly identical. It is probably safe to expect a near constant press fit throughout the temperature spectrum.

(2) Forced Air Induction Through the Shaft Assembly

The TJT-3000’s unique bearing cooling design is second to none. The TJT design team has incorporated a small suction impellor directly behind the rear shaft bearing, rotating at full shaft speed. This causes a suction effect in the shaft tunnel, drawing case pressurized air through the bearings. The air is then ejected from the shaft tunnel directly onto the turbine hub for additional cooling.

The net result is a rear bearing temperature reduction by 80° – 120° C (depending on throttle position) when compared to an identical turbine without forced air induction. This feature is critical when the turbine is operating at idle speed and the case pressure is lowest, adversely affecting the natural flow of cooling air for the bearings.

(3) Turbine Wheel Hub Cooling

Several small air jets have been incorporated into the TJT-3000’s exhaust turbine stage to spray pre-combustion air at the turbine wheel hub where the stress is at its maximum. The result is a significantly higher safety factor due to inconel’s increased tensile strength properties at lower temperatures. Two side effects to the hub cooling are reduced heat conduction to the shaft and consequently lower rear bearing temperatures.

For those of you interested, I plan to eventually make some sample FEA stress diagrams available for download. Although I didn’t design the TJT-3000’s wheel or provide any engineering insight into the design, the graphics will roughly identify the highest stress positions in an average 66mm wheel casting.

(4) Advanced Aerodynamics Applied to Diffuser Designs

Several hundred hours have been spent in design and optimization of the TJT-3000’s diffuser. This is not just another redesigned KJ-66 diffuser opened up for a larger compressor wheel. As the airflow exits from the radial compressor wheel’s outlet, the diffuser vanes are designed to efficiently harness the radial flow and convert it to high-pressure axial flow. The primary and secondary vanes have been optimized in shape and position to avoid the negative effects of airflow approaching the speed of sound, Mach 1. Side effects from poor efficiency diffusers can cause elevated temperatures at the front of the engine, reducing the life of fittings and hoses. Optimizing the diffuser section also greatly effects the overall performance specification, due to the inefficiencies commonly associated with this part of the turbine’s thermodynamic cycle.

Other significant features include a clot-resistant fuel injection system, compressor inlet cone profile optimization, ECU features, turbine mounting strap grooves in the pressure casing, etc. For now I will leave it at that and direct interested parties to our website, www.tjt.bz.

As you can probably see by now, TJT has spent an enormous amount of resources on the bearing system in the TJT-3000/JMP-3000 engines. Our reason is simple, its the “Achilles heal” of the engine...

That’s it, Thanks to everyone for your interest.

Kelly W, TJT Canada