Build a cheap turbofan: Difference between revisions

From NPrize
Jump to navigationJump to search
Line 83: Line 83:
* [[Turbofan:Compressor|Compressor]]: A three stage compressor, with a design allowing easy manufacturing.
* [[Turbofan:Compressor|Compressor]]: A three stage compressor, with a design allowing easy manufacturing.
* [[Turbofan:Blades|Blades]]: How to design an cheaply manufacture compressor, turbine and fan blades.
* [[Turbofan:Blades|Blades]]: How to design an cheaply manufacture compressor, turbine and fan blades.
* [[Turbofan:Combustors|Combustors]]: Combustor are the power input of the engine, and need not to melt while maintaining the combustion.
* [[Turbofan:Combustors|Combustors]]: Combustors are the power input of the engine, and need not to melt while maintaining the combustion.


[[Category:Turbofan]]
[[Category:Turbofan]]

Revision as of 22:01, 2 May 2011

This page gathers general information on turbofans. Our proposed design is scattered in several pages, with an index at the bottom of this page.

How to build a cheap (~ $150) turbofan?

Turbofans are the most efficient engine design for subsonic speeds cruising. They are more powerful and way lighter than reciprocating engines, fly at higher speeds than turbopropellers, and are less fuel-greedy than supersonic-enabled turbojets. They are however very difficult to manufacture as well as very expensive. On this page, we will explore how costs can be reduced while still having a reasonable efficiency, which is our primary concern here.

General principles

Lots of information are available on Wikipedia's page. General principle is that there is a combustion that feeds a turbine, which drives the fan and the compression stage feeding the combustion. The fan provides thrust from creating a massive air flow, and the turbine creates thrust by evacuating a hotter but less important air flow. As air is compressed from the intake, more air becomes available for combustion, and thus create more work on the turbine, and more intake.

Some design properties and configurations have to be properly calculated depending on the use of the engine, mainly for the intended aircraft speed:

  • The Bypass ratio (BPR) is a ratio between the mass flow rate of air drawn in by the fan but bypassing the engine core to the mass flow rate passing through the engine core. A BPR = 0 would be a turbojet engine. The higher BPR, the more efficient the engine, but also the slower exhaust speed.
  • The number of spools: modern engines embed a second and sometimes a third concentric shaft for high pressure operations. The low pressure shaft, the innermost has the fan mounted on. One stage engines exist and are less complicated and expensive to build, but are also less efficient. Indeed, higher rotation speeds in the internal spools allow to provide a more efficient compression. A gearbox may be needed to drive the fan if the shaft has a too important rotation speed in the case of a single-spooled turbofan. Multi-spooled engines prevent this issue, by keeping the low-pressure stages at relatively low speeds, suited for the fan.
  • The compression ratio is the ratio of the pressure of intake air on compressor discharge air. It is closely determined by the number of stages in the compressor and their efficiency. More compression means more air to blend with fuel and to cool the engine, and even more pressure at output, increasing the speed and mass of output gas, and thus the work that can be extracted by the turbines and overall engine efficiency.

Turbojet/turbofan engine simulation software from NASA: EngineSim

A must-read book by Klaus Hünecke: Jet engines: fundamentals of theory, design, and operation.

Video documentaries from a turbine renovator in Canada, probably the best resource on the Web for seing what's inside real engines: on youtube. Thanks AgentJayZ!

Design versus manufacturing

Design configurations and properties taken into concern on real engines tend to increase efficiency, i.e. higher thrusts for lower fuel consumption, but also try to reduce the exhaust noise. Cost is of course a concern, and an efficiency by itself, but maybe not a hard-constraint as it is for us. Safety of operation is their primary concern, whereas cost and ease of maintenance are our primary concerns -- and maintenance will be an important part of the job if the quality goes down because of the price.

Shaped core or shaped shaft?

An important optimization to reduce cost and complexity of manufacturing could be to have a simpler design of the parts creating the gas volume of the engine's core, i.e. the rotor(s) and the stator. In the above schema, we see that the shaft is straight and that the core envelope is curved suit required volume on each stage, although in real life, both are curved. If we take the required volumes on each stage and that we fix the core's envelope shape to a cylinder, the shaft will have a bumped profile (small-large-small diameter). This is much less expensive to produce, with a simple lathe (turning). Earlier engines, like the J79, have a cylindrical envelope. A curved envelope is complicated to build, requiring welding, pressing, stage bolting, the same techniques used in stator-construction in modern engines.

Real-world engines don't have a massive turned shaft because of the weight. They consist of plates for each compressor and turbine stage, that are linked together to the next stage using a cylindrical bolted joint. So basically, the shaft has no core, it's hollow, except for the plates on each stage. Our small engine design allows us to have a more simple design, since having a massively-turned shaft won't change much on its final mass. Moreover, we may think about a turbine-stage mechanism embedded in the stator to try to cool it, which would make it hollow. The main issue is now how to properly fix the blades to it and how to balance it/them?

Compressor and turbine blades

The most complicated parts to build in a turbofan or turbojet engine are the turbine and compression blades. The high-pressure turbine specially have to face very high temperature and pressure. On real engines, they are made of nickel-based superalloys. It's the inability of blades to withstand heat and work that limit the power of the engine, because the gas generator (combustion) and the compressor can provide more power to the turbine.

The compressor is not only made of blades on the rotor, but also blades on the stator. They prevent a rotating air flow to form inside the engine, which would decrease the enthalpy of the gas (its internal energy), driven by the action of rotor blades. Stator blades redirect the airflow on the next compression stage in the more appropriate and efficient direction.

Highest efficiency is reached in turbofans when gaps are reduced between rotor blades and the stator, as well as between the stator blades and the rotor. As always, good efficiency means good high precision and higher cost. Anyway, the precision of blades will have to be very good if we don't want it to dislocate when it reaches the high rotations-per-minute achieved by such engines.

Blade geometric design by itself can reveal complicated. The first engine(s) had flat blades. At the time, the efficiency of the engine was so terrible that it was believed that turbojets would never beat reciprocating engines. Then, in 1922, XXX proved that it blades were designed as airfoils, the engine would behave way better, and would even be efficient enough to be built. Airfoils for blade design allow the compressor stages to better increase the velocity, since they provide a reducing area for the air to pass through (= a compressor), converter to pressure by stator blades. For turbine blades, it's the opposite, they provide a gas expander by increasing the area through which hot gases flow.

Design considerations

Temperature control

Cooling is always needed in turbines, even if recent advances in materials and coatings increased the ability of blades to withstand heat. Since we will use low cost metals, cooling will be the main issue once we figured out how to build the engine. Expected combustion chamber temperature is around 2000°C for hydrocarbon or alcohol fuels. Iron melting point is around 1500°C, but will deform before melting.

Several cooling ways are used in a turbofan/turbojet engine: in the combustion chambers, only a small amount of the actual air flow is used for the combustion, around 20%. The rest is injected on the walls of the chamber and in the end of the combustion to dilute the hot gas, and to prevent the walls from melting (film cooling). Then, the first object struck by this hot gas is the stator of the turbine, which is, on actual engines, made of a ceramic coated high temperature alloy, but more importantly, hollow. Blades are welded on the stator ring, around which air from the compressor discharge circulates, enters the blades, and evacuates through drilled holes in the blades (convective cooling and film cooling again). For the rotor blades, the same principle is used, but with air coming from inside the rotor.

Startup and ignition

Startup can be done at ground manually, with compressed air for example, which will allow to reduce the weight and complexity of the engine. On the other side, a turbine engine is a nice way of having power on-board, using reducing gears and an alternator. That would also reduce the weight required for batteries, and the alternator would be used reversely as a startup DC motor. Also, the accessories attached to the reduced shaft would allow hydraulic or pneumatic power to be considered.

Igniter mechanisms must be integrated to the engine, possibly a self-maintaining igniter like a thread of tungsten, as used in miniature R/C engines. The combustion should be self-maintaining, but if pump or throttling malfunction, or more generally if a turbulence in the intake happen, leading to a discontinuous flow of fuel or air, ignition would have to be made during the flight.

Sensors

Engine must be designed with sensors, at least to determine if the engine is running properly or if it's under failure, and to control its rotation speed to ensure it's running at an efficient enough value, with regard to altitude (pressure and temperature). That can be done with a rotation sensor, measuring the magnetic field disturbances created by the blades or the rotor. Engine temperature should be controlled and recorded too. Pressure at different stages would be very useful for engine development, then for behavior indications when running at high altitude, but may be too heavy or expensive to put on the real engine. The rotor speed information would be redundant with some of the pressure information.

Fixing blades to rotor

In real engines, blades are fixed like this, with a dovetail or fir-tree shape that allow them to be mounted and remove axially but not orthogonally. The main problem appearing with this kind of mount is related to the size of the engines we need. As the diameter of the fan shaft gets smaller, the available space for the blade roots gets smaller, and require a higher precision for their manufacturing. The strength applying to the fixation is luckily reduced due to the small weight of the blades. A simpler design in blade root would be nice for manufacturing ease, maybe a simple square-section root is enough.

Fixing blades to stator

This is a major issue. On real-size engines, the stator is thick enough to have a rail into which the perpendicular-to-the-blades-roots are inserted and fixed. Creating a perpendicular root is already a challenge. Rotor's root would be able to compensate this problem by having longer roots with a locking mechanism on their side, but for the stator, the limited thickness of the stator's wall forbids it. Maybe bolting is to be considered. In that case, the screw heads would likely surpass the core's envelope and lightly disturb the fan flow.

External hardware

Fuel tanks in the wings, fuel pumps, fuel lines, and engine mounting will have to be designed too. Electrical wires for pumps, sensors, ignition and possibly the startup motor/alternator will also be required. Sensors will require input ports on the computer, and pump driving (= engine control loop) will require at least one output port for each engine on the computer.

Bearings

Two kinds of bearings are used in turbines.

  • Ball bearing: stator and rotor are joint using a ball bearing constantly bathed in oil to survive to high speeds/temperature.
  • Fluid bearing: pressurized oil prevents parts from touching, due to hydrostatic. Longer life and no maintenance, but harder to build and to operate.

Carbon or composite lip or blade seals prevent the oil from escaping to other parts of the engine. They may be arranged as labyrinth seals to increase their effect. Compressed air (a.k.a. bleed air) from the compressor discharge is often used to counteract the oil pressure on the seals.

Our Design propositions

From the different concerns expressed above, we propose a design for a low-cost turbofan. We also consider and propose innovative alternative turbofan designs. Several pages have been created in the Turbofan category to explain each subsystem and parts manufacturability:

  • Compressor: A three stage compressor, with a design allowing easy manufacturing.
  • Blades: How to design an cheaply manufacture compressor, turbine and fan blades.
  • Combustors: Combustors are the power input of the engine, and need not to melt while maintaining the combustion.