FIRST OF TWO PARTS

The Design of Pulse Jet Engines By Dick Klockner, EAA 3201 Drawings by Stan Dzik of pulse jet engines, Q thebut history very little about actual con-

the charge. The cycle is thus repeated at some frequency determined by the length and shape of the tube. (See Fig. 2). If the shape of the tube is nearly

uite a bit has been written about

struction. This is written in the hope that it may create a better understanding of the factors which influence the operation of the pulse jet. Many experimenters have either contemplated or have actually tried practical experimentation with these jets in the hope that they could be used for helicopters or sailplanes. Many of those who have tried actually building a pulse jet have given up in disgust after a few weeks or months of effort. The ideas presented in this article are based on experimentation carried on for about two years by myself and my good friend Dale Wolford. At least 90 percent of this time was filled with disappointment and failures. Ultimately, however, two types of pulse jets were made to run — the valve type and the valveless. Of the two, the valveless units are much more desirable.

FUEU

straight, th.- wave length can be assumed to be four times the length

of the tube, and the frequency can be determined by the simple formula: f = Vs f=frequency, cycles per sec. V s =velocity of sound in the hot gases

in the tube, ft. per sec. FIG.2.

VALVE

PULSE. JET

Drawings by Stan Dzik

A. The valves open sucking in a charge of air and fuel is injected. B. The valves are closed by the explosion and the gases are pushed out the tail pipe. C. The high velocity gases pass out the tail pipe, a low pressure area is produced in the combustion chamber and a new change of air is sucked in. The whole cycle is then repeated.

faction (or rarified portion of air) starts down the tube. This is reflected at the bottom and returns to the top, where the condition at "A"

is restored. This process is repeated

A

over and over again. If the air in the column is set in

motion by an outside force which has TUBES CLOS.EPKT OKIE E. KID

FIG. 1

The operation of a successful pulse jet tube is based on the resonance of a vibrating column of air. When a disturbance is produced in air, it is propagated by a series of compression and rarefaction waves in all directions. When the air is confined within a tube, any disturbance produced at one end travels to the other

end and is reflected back to the initial end, is again reflected, and so on. An example of this phenomenon is air blowing across the end of a tube. The condensation (or compaction) of the air produced at "A" in Fig. 1 travels down the tube, is reflected at the bottom and returns to the top, where it pushes the air aside as shown at "B". As a consequence, a rareSPORT AVIATION

periodic impulses such as a tuning fork, resonance will be produced if the natural frequency of the tube matches that of the tuning fork. In the case of the pulse jet this outside force is caused by the intermittent explosions of the engine. Briefly, we can say that a pulse jet engine will continue its cycles and resonate if, after the initial explosion, the rarefaction and condensation waves arrive in sufficient strength and in the proper sequence. The rarefaction wave causes a Jowor

than atmospheric pressure in the tube and the valves open, sucking in a charge of fresh air equal to ahov.t 1/7 the length of the tube. There is an influx of air at the exhaust end of the tube at the same time as it is sucked into the front of the jet. The condensation shock wave then compresses the gas-air mixture and the remaining burning particles from the previous explosion ignite

= wave length in ft. This formula is sufficient for most

experimenting. For example, we shall assume a pulse jet tube which is 4 ft. in length. The wave length of this tube is then 4 x 4=16 ft.

Since the speed of sound in the hot gas column is about 2100 fps we can quickly find the frequency of the tube with the above formula. 2100 ft./sec.

f=V s =————————=131 cycles/sec. 16 ft./sec. The tube shapes which will resonate seem almost infinite (although it is hard to believe at times) if all other factors concerned are proper.

There are a few conditions, however, which will not support resonance. One of the tube shapes which will

not function either in theory or in practice is a cigar or aerodynamic FIG.5

AERODVHMIC SHAPE VJi-HCM WILL. NOT

SHROUD-

PULSE. JET TUBE W I T H N -Hf ^-MEMS R

A similar condition occurs if the exhaust end of the tube is choked by

stronger explosion. Consequently it would be wise for a beginning experimenter to make his tubes long and

then gradually shorten them as experimenting progresses. If an increased diameter cylinder

is added to the tail section of a pulse jet, theory predicts that during one working cycle intake and compression can occur twice. This is not a

good condition for valve type units because it affects valve life. However, we have found it excellent for valveless units because the incoming charge can be doubled over a simi-

lar straight tail section. It is possible to have an engine which is

DIAMETER- \VJCHES

a flat plate with a smaller opening. The condensation and rarefaction waves are weak and are reversed when they reach the valves, so the tube does not function. If the jet is first running, it can be choked a small amount before operation ceases. Another condition which will not support resonance is a tube which is too short. The longer the tube, the less critical are the other parameters. The longer tube sucks in a larger charge of air and produces a

its

TV'PH PU LS.E-.

^ ^

JtST

^^

^^ 10 ^^

s*

^

.x

^^

^

^ ^^ *r*

Z

0

S 5

\Q

70

THR-UST

.50

IOO

ZOO

5OO

POUNDS

formance seems to be produced when the tail pipe has about twice the volume as the combustion chamber. As the maximum diameter of the

jet is increased, the ratio of the

total length of the jet to the maximum diameter can be decreased. In comparing a valve type jet of 6 in. with a valveless one of the same

size, the above ratio is about 9.5 for the valved tube and about 11 for the

valveless. For a valve unit of 22 in. maximum diameter, this ratio drops to about 6. For very small jet tubes

of 1 in. maximum diameter this ra-

much shorter using this technique. Even a simple straight pipe will

tio goes to about 20.

as a surprise to some, but it is perfectly logical when one considers a simple organ pipe. This configuration is very poor on performance, however. The best all around per-

shown in Fig. 4 will show the diameter of the combustion chamber for any particular thrust. This chart applies only to the valve type jets. Since the cross-sectional area of the

run as a pulse jet, which may come

DI Ami TE 2

1———b———i———•——•———

VS. STAKTK . THteUST FOR V/AL VE

In order to determine how large to make a pulse jet tube, the chart

combustion chamber should be about twice that of the tail pipe, you can determine this dimension once a combustion chamber diameter is picked.

The length can also be found by multiplying the maximum diameter by

9.5 for tubes of about 6 in. The construction materials for pulse jet tubes should be 347 stainless steel because of the corrosion and high temperatures. In our experiments

we have measured temperatures as

high as 1790° F. However, most experimenters cannot afford such expensive materials. We found that 16 gauge cold rolled steel works quite well for static experiments. At this temperature the cold rolled steel has very little strength, but it is ample

to withstand the combustion chamber gauge pressure of 20 psi.

In a valve type unit the valves are of the utmost importance, for their presence dictates to a great extent the life of the jet and also the efficiency. If a valve area is too small for a tube, it will have poor efficiency, and if it is too large the

tube will not operate. We have found

that the best method of expressing valve size is the ratio of the inlet area to the combustion chamber area. We have run valve type units with

THE "GOON"

ratios as low as 0.1 and as high as

SEQUEL

.378. After a jet is started, the ratio can be increased with a subsequent increase in thrust. The unit will not

necessarily start, however, at this

John Calcr Photo

John Caler, 1562 Colfax Ave., North Hollywood, Calif., writes us that he has purchased Art Chester's "Goon" from Harvey Mace who recently restored it. (See June SPORT AVIATION). John was a close neighbor of Chester and spent much time with him while he was working on his various raceplane projects. He hopes to restore the "Goon" completely to its original configuration and is seeking a C-6-S Menasco engine for it. Anyone able to be of help is

invited to contact him at the above address. 14

new ratio. If the tube is moved through the air at a high speed, it is the same as increasing the ratio. Consequently at some flight speed the jet will cease to operate. Next month we will describe and illustrate two of the most popular valve types used, and conclude with a short description of the valveless type pulse jet. A JULY 1958

Second of Two Parrs

THE DESIGN OF PULSE JET ENGINES By Dick Klockner, EAA 3201

T

he most popular valve for small units (under six inches maximum diameter) is a spring steel petal valve developed by William L. Tenney. This valve as shown in Fig. 5, and in Fig. 6 is shown mounted on a tube for

Fig. 8 (Right) Navy type valve grid and fuel injection system mounted on a tube ready for a trial. A large starting blower is seen on the left.

test run. The most popular valve for larger units (six inches maximum diameter

or larger) is one developed by the U. S. Navy. A valve of this type is shown disassembled in Fig. 7. Each leaf consists of a sandwich of blue spring steel, rubberized cloth and another leaf of blue spring steel. The

length of the valves thusly: Fig. 7 (Top) Flapper valve head assembly for a six in. pulse jet. Note the rudder bumpers for valve seating.

30% of 131 = 39.4 Design frequency=39.4—131 = 170.4

cycles/sec. f=33,300 t

tube, the size and thickness of the

valves will be determined by each particular tube design. The frequency of the valves can be determined by the following formula: f= 33,300 t 12

f=the frequency Fig. 5 (Top) A SVz in. valve head with

ports, the .010 in. blue spring steel petal valve and the back-up plate for the valves. This type of valve head is used in several small pulse jets, such as the Dyna-Jet.

where t=the thickness of the

valve in inches l = the vibrating length of the valve in inches If the frequency of the valves is purposely made about 30% higher than that of the tube, they will respond more quickly to pressure variations. To design a set of valves so they are 30% higher than the fre-

quency of the four foot tube mentioned previously, we use the above formula to obtain the dimensions. Assuming a valve material which is 0.010 of an inch thick, we obtain the

12 12 = 33,300x0.010=1.95 170.4 1 = 1.40 in.

Blue spring steel is the best material for valves, but since the modu-

lus of elasticity of all steels is approximately the same, you can use substitute materials if you compensate for loss of strength. One of the best running valves we had was made from tin can stock. As far as fuel is concerned, almost any hydrocarbon will work. Gasoline is the easiest to experiment with because of its volatility. I would say that one of the major causes of a pulse jet failing to start is due to poor atomization and/or a poor fuelair ratio. We have found that the best method of fuel injection is to either completely vaporize the fuel

Fig. 6 (Top) Petal type valve head mounted on a tube ready for a test run.

Navy type grid incorporates rubber bumpers for the valves to seat on in order to prolong their life, for they are destroyed usually by impact, not fatigue. The high velocity air through the throat section keeps the rubber from overheating. Fig. 8 shows this valve mounted on a tube for trial. Since the natural frequency of the valves should approximate that of the SPORT AVIATION

Fig. 9 Pulse jet using a tractor carburetor for mixing the gas and air. The fuel-air mixture in this case is passed through the valves into the combustion chamber. Note sweeper motor in foreground for starting.

25

FUEL

A. A charge of air is sucked in and mixed with

the fuel.

B. The gas-air mixture is exploded and the gases

are forced out both ends.

EOtL. —^

C. The reduced pressure in the combustion chamber sucks in a new charge of air and the cycle repeats.

Fig.

first by preheating, or by mixing gasoline and compressed air just before it is injected into the tube. Once the engine starts, it will preheat its own fuel.

A good needle valve for

fuel adjustment, preferably one with

about a 10° taper on it, is a very necessary item for successful operation. The fuel can be injected either into the combustion chamber or added externally allowing the fuel-air mixture to go through the valves. In the latter case there is always danger of a flash-back and burning outside of the valve head due to leaky valves. The gasoline can be added with an ordinary automobile carburetor if it is added outside the valves. Vitally necessary in testing any pulse jet engine is an understanding wife, patient neighbors and a good set of head phones to protect your ears. The noise from a six inch valve type pulse jet is about 136 decibels. This noise level is actually painful to the ears if you are close. The noise from a valveless pulse jet is somewhat lower. We have also found that the addition of water to the interior of the tailpipe cuts down on the noise. Besides this advantage the water also adds mass to the system and causes an increase in thrust. The theory of a valveless pulse jet is essentially the same as the valve type, but the physical appearance is somewhat different. (Fig. 10). First of all there are no moving parts in this type of pulse jet, consequently no maintenance is required. Along with this decided advantage valveless jets possess a lower thrust than a valve type unit of a comparable size. This is due to a lower combustion chamber pressure. The length to diameter ratio is also much higher than valve type units. Even with these two drawbacks the overall advantages of the valveless units make 26

them

10 VALVELESS TYPE

more

desirable

than the valve types.

powerplants

The tube shape of a valveless unit

consists of an inlet throat, a combustion chamber and a tail pipe. The inlet throat is critical as to both

length and diameter if optimum performance is to be obtained from the jet. The length of this inlet throat is about one-fourth the length of the tube (combustion chamber and tail pipe). This length, however, varies more or less, depending upon the volume of the combustion chamber and tail pipe. Therefore the best operating conditions are arrived at by experimentation.

We have run valveless units with an inlet area/combustion chamber area ratio as low as 10% and as high

as 24%. Here again the best results must be arrived at by trial and error, because changing any one of the

dimensions of the pulse jet tube produces an entirely different relationship between the other fundamentals of the tube. For example, changing

the angle of a diverging cone on the exhaust tube may allow a reduction

in the tube length, which in turn requires a shorter inlet throat. Cnang-

ing the angle of the exhaust cone even changes the noise level of the jet. A smaller angle will decrease the noise and also the thrust. Another characteristic peculiar to valveless jets is the fact that part of the thrust comes out the throat section. The amount of thrust exerted at the inlet is proportional to the area of the opening. A simple method of reversing this thrust is to simply bend the tube into a "U" section or to add a "U" section to the throat. The construction of the combustion chamber and tail pipe is es-

sentially the same as the valve type pulse jet, but longer.

We have found that starting is much easier for the valveless jets than it

is for the others. The equipment required for starting is much less de-

manding.

Fig. 11 A unique valveless pulse jet which has an airfoil cross-section and is designed to be mounted radially as a helicopter rotor !i!ade, the object being to serve as an engine as well as a rotor blade. The engine is mounted on a whirling test «>ta:id. Note the rotary fuel seal at the bottom of the photo, the sweeper for starting and the pressurized water injection equipment for internal surface cooling.

For valve type units a

high pressure air line or a high volume high pressure blower is required. The valveless jet is easily started with the wife's vacuum cleaner used as a

blower.

We feel that such a simple and maintenance-free engine possesses a terrific potential as a powerplant. It

is ideal for the man who likes to

experiment, since it is cheap and can be built to his own power requirements. ' '

• AUGUST 1958