Efficiency of a pop-pop engine.

Our first estimate was done in early 2005 with a small Indian boat. The first following
lines are a copy of what was written at that time.
Small Indian boat.
Because of the propulsion principle, the boat being sailing at the speed , the output of
the water!et needs to be at least once par cycle faster than .. "
ma#
$
%et&s suppose the flow perfectly sinusoidal. '"ery point of the li(uid sna)e mo"es by
Ft aSIN t aSIN d 2 = = * being the fre(uency of the cycle. +eri"ation of this e(uation
gi"es the speed of the water inside the pipe. Ft FaCOS v 2 2 =
"
ma#
,2*a$. -ith *,./0 and ,0.15m2s we get a$3,2.10
43
m a$3,2mm
a being the half amplitude of the displacement, the total stro)e is more than 5.6mm.
These last three data allow to compute the global efficiency which is the ratio between
the released mechanical power 7drag force multiplied by speed8 and the supplied heating
power which is the one of the candle9
F
V T
r

=
*or this application
: 0011 . 0
2.
15 . 0 002 . 0
=

= r
. It&s pathetic; <s the measures were not performed as laboratory
ones, the relati"e uncertainty on some of them is big= but they are only three. <ssuming we
were "ery bad or "ery unluc)y so that on the three of them we made an error in the same way,
and from single to double 7it is ne"ertheless enormous to do that8, the efficiency would
become 0.00..:. -e can accept that during the >endurance? test the flame of the candle was
not as big as it could ha"e been, and if we accept a worse condition 10 times less, the best
result could only reach 0.0..:. It is to be compared to the 35: of a classic propulsion 750:
for the engine and @0: for the propeller8. This is still "ery "ery bad and !ustifies that there has
been no industrial application of the pop4pop engine.
Aomparison with a mechanical propulsion 7springBpropeller8 on toys of the same si0e.
1C8 Spring. To wind up the spring re(uires about 10 turns of the )ey with a tor(ue of 200mDm
70.2Dm8. Aorresponding energy9 0.2E2EpE10,12.5Foules. It is minute. This engine propels the
boat for appro# 12.5 seconds 7to simplify8. Therefore, the power is 1-. Ta)ing into account
the mechanical efficiency and the one of the propeller the deli"ered power is smaller. %et&s
say 0.5-.
2C8 Aandle. Though it is not ob"ious, the power and the energy deli"ered by a candle are
relati"ely big. < small birthday candle 7mass9 1 gram8 deli"ers appro# 35- as heat, and it
ta)es 10 minutes to burn. Aorresponding energy9 21)F. -ith 2 grams burnt in 5 minutes 7data
from Grofessor %e Bot8 it means 62)F and 160-.
3C8 'fficiency. The efficiency of such a small toy is li)ely about 10 times less than the one of
a big ship= i.e. appro# 3.5:. Grofessor %e Bot measured similar thrusts with mechanical
propulsion 71- 3.10
42
D8 and pop4pop propulsion 7160- 1...10
42
D8. It means a ratio of 233 in
fa"or of the mechanical propulsion. +i"iding 3.5: by 233 gi"es 0.015: and it can be
chec)ed that it is lower than 0.0..: calculated before by e#cess. This consolidates our
measures and computations.
%ater e#periments.
Specific e#periments ran in 2005 with an electric heating source allowed to impro"e
the )nowledge of the efficiency and to impro"e the efficiency in some circumstances. < pop4
pop engine can sustain any type of heater. The electric heater is reser"ed to the test bench. It
allows to )now and to control the power consumption.
There are so many different pop4pop engines 7with or without diaphragm, with one or
two pipes, sometimes three, and with "arious pipe diameters and lengths, with or without
no00leH8 that we are not going to e#amine all of them. -e will loo) at the efficiency of two
of our more powerful engines, a coil engine and a diaphragm one.
The purpose of this document is an efficiency estimate. Therefore, we must say that it
is not necessarily the more powerful engine which is the best for efficiency. /owe"er,
according to the e#amples we ha"e loo)ed at, for a gi"en engine the efficiency increases with
the power. -e will gi"e an e#ample.
The figures used hereafter correspond to the mean "alues during rather stable
conditions for se"eral tens of seconds. *or "ery short durations we got sometimes (uite bigger
figures 7almost double8 but due to the thermal inertia of "arious components 7engine and
heating de"ice8 using these figures would be cheating.
+iaphragm engine.
Aonstitution9 drum 7e"aporator8 of which the lower wall made of copper has an internal area
of 12.5 cm2 and a "olume of 1 cm3. Gipe made of brass with internal diameter 5mm and
thic)ness 1mm. %ength9 330mm. Gipe outlet slightly groo"ed on the outside.
This engine e#erted a thrust of 30mD for an electrical consumption of 100-. The same thrust
was got with a groo"ed no00le of the same diameter when using a permanent flow of 26.5
cm
3
2s 7cf. document IThrust measuring test benchI8.
By di"iding the flow by the no00le area we get the speed 9 0..@m2s. <nd the power is got by
applying the formulae
2
2
1
V Q P
v
=
which in this case gi"es J.2m-. The efficiency is the
ratio of this power by the heating one= i.e. the efficiency is roughly 0.1: 7ta)ing into account
some dissipation in the air8.
On this picture the soldering iron "isible
on the left has been modified. It is fi#ed
to the steam drum.
<t the bottom, towards the right side is
the single pipe.
+uring the tests the engine was pac)ed
inside an isolating material to limit the
heat losses in air.
Aoil engine.
Aonstitution9 engine made of copper 5#1 7i.e. internal diameter 6mm8 with 6 turns around a
diameter 12mm. %ength of each pipe out of the coil9 appro#. 1J0mm. Strait cut.
This engine deli"ered a thrust of 2#1@mD for an electrical consumption of 50-.
The same thrust was got with a no00le of the same diameter when using a permanent flow of
16cm
3
2s= which corresponds to a "elocity of 1.11m2s and a power of 5.@m-= i.e. 7when
multiplying by two because there are two no00les on this engine8 an efficiency of 0.023:
rounded to 0.03: to ta)e into account the losses in air.
*or information, this engine deli"ered 2#12mD for 25-. The same no00le e#erts 12mD for a
permanent flow of Jcm32s= which corresponds to a "elocity of 0.@2m2s and to a power of
2,3m-= i.e. an efficiency of roughly 0.02:.
This efficiency could loo) bad compared to the pre"ious one because a the denominator the
consumed power is half the pre"ious one, but a the numerator the flow and the "elocity
7which inter"enes as s(uare8 are each one reduced by almost one third.
Conclusion :
It is confirmed that the efficiency of pop4pop engines is pathetic. That is the reason why there
were no industrial applications, and "ery li)ely there will ne"er be one.
This conclusion doesn&t alter the fascinating aspect of this ama0ing engine without any
mo"ing part.
<s it can be seen on the picture, the
pipes are on both sides of the hot
section. This is a special engine for test
bench use only.