In 1949, Cornel Aeronautical Laboratory was working on a project for the US
Navy, called Project Squid. This work focused on the operation of valve less
pulse jets. In another work Bertin (1951) studied about
Escopette pulse jet. This type of pulse jets uses air diode instead of valves
like Marconnet pulse jet.
According to their research of the 1950s, the inlet had no preferred frequency; beside its high reliability and ease of using constant fuel pressure for feeding made it very attractive. The weakness of air diode is the flow back out which is the reason of reduction of the thrust.
Emmerich (1953) used pulse jets to produce tip-propulsion
for rotors. Tests were performed at various altitudes and noise tests were conducted
as well. They found that when the altitude was increased and density decreased,
the jet became more and more difficult to start. All data was converted to standard
data for comparison purposes. As suspected, a decrease in thrust was noted
as air density decreased.
Lockwood (1963) studied about U-tube pulse jets. This
work studied several methods to achieve thrust for lightweight engines. They
tested several exit geometries and combustion chamber designs and found that
changing the combustion chamber shape has a dramatic effect on thrust and efficiency
with TSFC levels less than 2.0 pph lb-1. Also, Logan
(1951) and Reynst (1961) studied about valve-less
pulse jets and pulsating firing systems, respectively.
Defense Advanced Research Projects Agency (DARPA) is commenced the investigation
about pulse jets. The main objective of these studies is research about scalability
of small Unmanned Aerial Vehicle (UAV) propulsion. Review of literatures show
that design parameters of pulse jets had not been fully investigated; therefore
any equations about pulse jets have not been developed. So, the objective of
present study is finding and scaling laws and design characteristics of the
pulse jets in order to optimization of the engines.
This valve less pulse jet engine comprised an inlet tube, combustion chamber and exhaust nozzle. Two fuel ports were located at the combustor chamber and immediately downstream of the transition section, a stainless steel sheet with 6 mm thickness was chosen because it was available and easy to machine (Fig. 1).
Fuel was first injected upstream of the reed valves through 6MM stainless steel tubing inserted through the constant area inlet, as shown in Fig. 2. The fuel was added via two small holes drilled circumferentially about the tube. The fuel used in this experiment is gaseous propane.
|| Geometry (all dimensions are in cm)
|| Experimental model of pulse jet
STARTING OF THE JET
Starting the pulse jet began with two things, the need to supply the first
intake of air and a way to ignite mixture of fuel and air (Foa,
1960). In the valve less configuration, the fuel used was gaseous propane
and directly injected in to the combustion chamber behind the inlet.
After the mixture entered the chamber, it was ignited by a spark. After the
initial combustion events occur, the engine continued to run on its own. A warm
jet was found to be easier to start than a cold jet; this is most likely due
to the effect of heat transfer at the walls (Foa, 1960).
After the jet is running, the ignition can be turned off and the forced air
can be stopped.
PRESSURE, TEMPERATURE AND VELOCITY
Pressure data is used to determine the operational frequency of the jet, also to find the peak pressure and the amplitude of pressure waves. A pressure distribution is shown in Fig. 3. Pressure inside the combustion chamber immediately before ignition is generally equivalent to atmospheric free-stream conditions, after ignition pressure arise in combustion chamber, then pressure drops due to an over expansion and convert to velocity in nozzle exit.
Figure 4 shows the temperature trends for the intake, combustion chamber and exhaust for the propane fuel. High heat transfer coefficient because of high speed flow of air in intake tube (Fig. 5) result that we observe the intake temperature is lower than the combustion chamber temperature and exhaust temperature.
|| Pressure distribution along the pulse jet
|| Temperature distribution along the pulse jet
|| Intake tube
|| Velocity distribution along the pulse jet
|| Influence of mesh count on thrust
In intake, we see maximum value of velocity because of effect of vacuum after ignition (Fig. 6). Also, grid study is done and its results are reported in Table 1. This table shows that grid sizes have less influence on results, therefore grid independent solution is achieved and second grid with interval count 20 is used as a main grid.
Obtaining time-resolved combustion chamber pressure was a goal from the start
of the project. It was known from work with the 450 cm jet that monitoring pressure
throughout the jets operation was the best way to obtain operating frequency
and validate CFD models.
|| Mercury manometer
In addition, measuring the peak pressure rises in the combustion chamber proved
useful in determining the pulse jets efficiency.
Mercury manometer: To measure the average combustion chamber pressure, a Fisher Scientific mercury U-tube manometer was used, shown in Fig. 7. Its scale reads to 50 cm of mercury. It was connected to the combustion chamber pressure port by several meters of 1/4 plastic tubing.
Thermocouples: Type B high temperature thermocouples were used to attain
temperature measurements of the intake, exhaust and combustion chamber temperatures.
The intake and exhaust temperatures were measured one diameter upstream and
one diameter downstream respectively shown in Fig. 8 and 9.
The combustion chamber temperature was measured at the instrumentation port
location 1/8 of a diameter away from the combustion chamber walls as seen in
From measurements maximum pressure of various locations and thrust force are obtained. These results are presented in Table 2. According to this table pressure of intake tube and exhaust have not any considerable difference, but pressure of combustion chamber is high. These effects are produced thrust force equal to 12 N approximately as indicated in Table 2.
|| Thermocouple locations at intake
|| Thermocouple locations at exhaust
|| Thermocouple location inside combustion chamber
|| Pressure and thrust force values
The effects of lengthening are similar on both valved and valve less jets. The valve less jet is less sensitive to changes in length than the valved jet. It is also noted that the valved jet operates closer to ¼ wave tube than the valve less. By increasing the exhaust length, the throttle ability increases as well. A maximum can be reached for certain inlets in the valve less configuration. Upper throttle ability limits plateau at specific exhaust lengths while the lower throttle ability limit continues to decrease.
The effect of changing the diameter of the inlet is that the frequency increases with increasing diameter. The change is linear and it can be modeled as a Helmholtz Resonator. The temperature effect of increasing inlet diameter is that the exhaust temperature rises (due to it being fuel rich) and the inlet temperature decreases (due to increased air intake).
The current configuration produces little thrust due to the experimental setup that includes opposing inlet/exits. The exhaust exit geometry is sensitive to shape. A flared tip is preferred and sometimes required for operation. The 50 cm class pulse jet can be modeled as the average of the inlets Helmholtz frequency and the exhausts 1/6 wave tube frequency.
Highest average combustion pressure occurs with 5/8 inlet, this is due to the flow not fully expanding at high fuel flow rates and combustion occurring in exhaust tube.
Combustion chamber peak pressures are significantly higher for valved jets.