Our Jet Project
In September of 2017, we decided to start working on a turbojet engine project. We decided to create the project to experiment with and to assist us in learning concepts taught while at school. John is an aerospace engineering student who is currently a research assistant and working towards receiving his BS in AE as well as his MBA. Jake is a mechanical engineering student pursuing a BS in ME, a mathematics minor, and hoping to specialize in combustion engines. We started out with a 44mm turbocharger off a 1984 Mercedes-Benz 300 SD and built the entire engine around that platform. The turbocharger is a Triple K (KKK) turbocharger that had been recently replaced according to records surrounding the vehicle. After pulling the turbocharger, we began designing the combustion chamber and flame tube.
Reading through all the research we could find on basic radial compressor engines, we concluded that the dimensions were going to be completely dependent on the compressor inlet diameter. The length of the flame tube and combustion chamber were found to be optimal at 6x the inlet diameter. The width of the flame tube was found to be 2x the inlet diameter and the width of the combustion chamber was an extra inch added to the flame tube to allow cooling and proper atomizing. While these dimensions are the ideal dimensions, due to time constraints, cost, and mobility issues, we decided to cut the length of the flame tube in half to two feet. We calculated the dimensions for the holes in the flame tube based on formulas we acquired and chose to drill them in a spiral pattern to experiment with a "swirl" type atomization of the fuel.
The fuel chosen for the turbojet was kerosene as it is extremely similar to jet fuel. This allowed for lower combustion temperatures than gasoline or diesel, made it more readily available than propane, and most importantly, it was the safest fuel because it has to be highly compressed to really ignite. To atomize the fuel, we used a furnace nozzle with a max flow rate of 9 gph. Because kerosene is a liquid rather than a gas, it must be pumped at high pressure for it to be atomized. As a fuel pump, a Walbro GSL392 was used. When we spoke to Walbro about the pump that was needed, they mentioned that they did not endorse the use of kerosene through one of these pumps but had seen it done before. We used this fuel pump to pump the kerosene with 3/8-inch lines, a ball valve as a fuel control, and an inline pressure gauge placed after the ball valve to monitor pressure from the fuel pump. The electrical for the fuel pump is temporarily run directly to a 12v but will be implemented into the “black box” in the future.
As an ignition system, we originally had planned to use a spark plug and coil. After testing, we were unable to get the plug to properly ignite so instead, we switched to a basic electric grill ignitor. The two electrodes were sealed and welded into the engine next to the fuel injector. All the electrical from the motor is run to a “black box” and put on a switch panel. From there, the switch panel is run to an invertor and plugged into a wall. We decided to go with an ac to dc invertor, so we did not have to worry about carrying a car battery around when testing. This also minimized the chances of sudden electrical failure due to a dead battery. The next part of the electrical system is the oil pump.
The oil pump we are using is a 3.7 gpm oil scavenge pump made for remote mount turbochargers. The oil lines used are the same 3/8-inch lines used for the fuel system. The system starts in an oil tank, is gravity fed to the pump, is pumped through the lines past an oil pressure gauge into the top of the turbocharger and then drips out of the bottom back into the oil tank. The oil chosen for the jet is SAE 5W-30 full synthetic motor oil. We went with synthetic oil because at higher temperatures it does not break down nearly as much as conventional oil.
When testing the jet for the first time, the gap between the flame tube and combustion chamber was to large and not sealed well enough. This led to no compression and major fuel leakage. After a redesign, we ended up making the bolts holding the flame tube and combustion larger. On the second test fire, it was very close to firing but still was not sealed well enough. A metal gasket was added as a seal between the two and this solved the leakage issue. After this, the jet ran well and idled smoothly.
Issues we ran into when building the turbojet.
Initial combustion chamber/flame tube connection
Clocking the turbocharger
Faulty fuel pump
Burning oil
Spark plug
Initial furnace nozzles
Initial fuel pump