Infinite Energy 23: 23 (1999)
Eugene Frenette: US Patent # 4,143,639
Eugene Perkins: US Patent # 4,424,797
E. Perkins: US Patent # 4,483,277
E. Perkins: US Patent # 4,501,231
E. Perkins: US Patent # 4,651,681
E. Perkins: US Patent # 4,779,575
Ralph Pope: US Patent # 4,798,176
December 1998 Kinetic Furnace Test: Previously Reported
Results Retracted
By Jed Rothwell & Ed Wall
We first reported on the Kinetic Furnace, invented by Eugene Perkins and
Ralph Pope, in Issue # 19. The device had, at that time, been tested by
several independent engineering laboratories and services. The Kinetic Furnace
is, as the name implies, a device for heating and forcing airflow. Heat is
generated by means of a rotor that flings water from the hub to the rim of its
chamber, through some precisely dimensioned nozzles. This "stirring" action is
driven by a 6 HP electric motor. The heated water is driven out of the rotor
chamber into a radiator and out an output duct.
In April 1998, Eugene Mallove and Jed Rothwell assessed the furnace for
themselves at the inventors’ facility in Cumming GA, where they observed the
apparent production of excess heat.
Furhter testing was carried out by Mallove and Ed Wall, June through
September of 1998, in Bow NH at NERL (New Energy Research Laboratory), but no
significant excess heat was observed during that period. Another machine was
shipped from Georgia, but it too showed no excess. Finally, Pope loaded a
third unit into a van and drove it to New Hampshire himself. He helped install
and test it, but this third test also failed. Different sources of water were
tested, the operating temperature and the rotation speed of the motor were
varied slightly, but no significant excess energy was observed. In IE #22 we
reported on this briefly, expressing continued hope that the machine would
produce excess heat. We reported a COP (Coefficient of Performance) of 115%
(155% excess heat) This level of excess heat is difficult to establish with
certainty using airflow calorimetry. A 200 or 300% excess could be detected
with confidence, but 15 to 20% could be the result of subtle errors.
Pope returned to Georgia, discouraged. It was clear that we had hit a dead
end, and that if the machine does work, there must be something different
about the way it was being operated or the water or some other material in
Georgia. We decided that the only way we would ever get to the bottom of this
mystery would be to conduct extensive tests on site in Georgia, using our
instruments and Pope’s in parallel. The machine is large enough to allow
several temperature probes and ammeters to be attached simultaneously, unlike
the small hand-held cold fusion cells, which often only have room for one set
of instruments.
In November 198, Pope reported that he was now achieving a COP as high as
180% with the machine he had brought to Bow, which had been reconditioned and
reassembled with a new rotor and pipes. Rothwell conducted a half-day of
testing of this machine in the Cummings GA machine shop location, using the
same instruments and techniques Rothwell and Mallove used in April. Most of
this 180% turned out to be an artifact of Pope’s anemometer, which suffered
from a power supply probem caused by worn out rechargeable batteries. The
measured air speed was too low. The high excess heat results reported by Pope
in previous issues of this magazine were also probably caused by this error.
Ralph Pope does not agree with this assessment and believes that the air speed
was measured correctly. The blower power did not change and so it is highly
unlikely that the airspeed fell. While the large excess was clearly wrong,
apparent 46% excess heat was seen, which was in line with what we observed in
April. We were encouraged by this preliminary result, yet puzzled and wary by
our inability to replicate it in New Hampshire. We decided to press ahead with
full-scale tests in Georgia. Ed Wall went to Georgia bringing several tools
and precision instruments, listed on page 27.
In the series of tests from December 4-9, 1998, Wall, Rothwell, and Pope
tested the Kinetic Furnace extensively, using higher quality instruments and
more sophisticated techniques than Pope had ever used. Unfortunately, no
significant excess heat was observed. Based on the December results, we
believe our initial assessment in April was incorrect, and there was never any
significant excess heat in the tests we performed in Georgia or New Hampshire.
We believe we have discovered the source of the error which caused the
artificial heat in Georgia. The error was in technique rather than instruments
or formula. In the December tests we used an improved technique, a computer,
an HP 34970A Data Acquisition System, and an array of 11 K-Type, 20 gauge wire
thermocouples (four on the inlet and seven on the outlet side). The
thermocouples were calibrated carefully through the temperature range of
interest and compared to NIST traceable mercury thermometers. By performing
this calibration, we learned that the thermocouples read about 0.5° F less
than the calibration thermometer over the temperature range of interest. At
the same time we used the computerized instruments, we repeated the tests
using the same relatively crude, hand-held instruments --- ammeters and
thermometers employed in November. In this second test with hand-held
electronic, alcohol and mercury thermometers, we measured no excess heat, thus
confirming the computer thermocouple readings.
The biggest problem with the April and November tests in Georgia was the
lack of a calibration heater This was not an error or an oversight --- we did
not have the time to install one during these preliminary, one-day tests. The
tests in Bow NH were conducted over two months, and they employed a
calibration heater to avoid dependence on air speed measurements and
formula-based calculations. With a calibration heater, results from the
electric heater were compared with those measured with the Kinetic Furnace.
Even of the air speed, electric power, or duct cross-section measurement is
inaccurate, the comparative results should show an excess, if one exists.
The first day and a half of testing in Georgia were devoted to installation
and testing of the thermocouples and the calibration heater, which was run at
three power levels, up to 3.25 kW. At the end of the second day we turned on
the Kinetic Furnace, which also consumes about 3 kW electric power. All tests
were done with the heater in place, whether it was active or not, to maintain
consistent air flow patterns. He Kinetic Furnace testing protocol calls for
the machine to be run with the cooling fan turned off until the internal water
temperature rises to at least 160° F. Then the blower is turned on, and
internal temperature drops rapidly at first. Stored up heat in the rotor and
water are removed. In 20 to 30 minutes the rotor and outlet temperatures
stabilize. Twenty minutes into the first Kinetic Furnace test, the initial
burst of saved up heat was exhausted, and the temperature fell to about the
same level seen with the calibration heaters at 3 kW. It was obvious that the
furnace was producing no excess heat.
In our first test it was apparent that the Kinetic Furnace was producing no
excess heat. This left two possibilities, which we investigated over the next
5 days:
1. That the previous results were an artifact.
2. That the machine previously produced excess heat, but it was not producing
it on December 4.
To check for possibility #1, an artifact, we began by repeating the tests
with the thermometers, hand-held ammeters, and other instruments used in the
previous tests. We placed the thermometers in the same locations as the
computerized thermocouple arrays. The hand held instruments were used at the
same time as the computerized equipment, during both calibration heater runs
and live Kinetic Furnace runs. The hand held instruments showed the same 9 or
10° F delta T as the computerized thermocouples, which indicates no excess
heat. We then moved the thermometers to a location roughly as far away from
the Kinetic Furnace as Rothwell selected in November and we observed a 13 or
14° F delta T. To assess possibility #2, we tried changing the rotor, the
water, the air flow speed and other parameters, which we hypothesized might
have a controlling effect on an excess heat phenomenon.
A hypothesis discussed by Horace Heffner in the Vortex Internet forum came
to mind. Heffner thought that a warm stream of air might be moving from the
outlet duct 15 feet back to the inlet. Although that seemed unlikely, we
looked for a stream of air by placing the anemometer next to the outlet duct,
at a spot 50 cm back from the end of the duct, toward the Kinetic Furnace. We
moved the impeller around, searching for a stream of warm air, checking the
left side of the duct, the right side, the top and bottom. The anemometer is
quite sensitive to small streams of moving air. The impeller did not spin, so
we conclude there was no discrete stream of air going from the outlet duct
back toward the Kinetic Furnace. However, the hypothesis stuck in mind, so we
did a more careful examination on the air surrounding the Kinetic Furnace and
duct on all sides. We now believe there is an area of circulated air around
the machine that is warm in comparison with air in the greater volume of the
room. This was more apparent during tests on Sunday when the machine shop was
deserted and the air in the rest of the building was quiescent. The machine
shop is a 5000 sq. ft. steel frame building with the ceiling 14 ft high at the
eaves. Outside of this envelope of warm air around the machine, at locations
20 and 30 ft away, the ambient air temperature was roughly 13° cooler than
the Kinetic Furnace outlet, and roughly 3 degrees cooler than the air
surrounding the inlet. Thus, the actual delta T temperature between the inlet
and outlet was 9 or 10° , indicating no excess heat.
In April and November, we measured the inlet temperature at a spot too far
from the Kinetic Furnace, outside the cloud of warm air. This spot was picked
because Ralph Pope cautioned us not to place the sensors too close to the
furnace where they would pick up heat radiating from the rotor and other hot
machinery. However, this was incorrect. There was little significant radiant
heat; most of the heat near the machine was convective, and it went away
during the test. In the first round of tests in December, four inlet
thermocouples and three thermometers were placed in various locations around
the inlet. The closest ones were about 6 inches away from the rotor and
calibration heater. The farthest ones were 35 inches away from the inlet and
reasonably well-shielded from radiant heat, yet they were only 0.9° F cooler
after the fan was turned on. The difference would have to be 4° F if the
excess heat was as high as it appeared to be in November, so radiative effects
were not large enough to nullify the apparent excess heat. During the warm up
phase of the experiment, before the fan was turned on, the difference between
the inlet thermocouples and thermometers was 2 to 3° . Evidently this was
convective heat, because when the fan was turned on and the air pulled past
the thermocouples and rotor this temperature difference largely disappeared.
The confusion about the inlet temperature underscored a serious weakness in
our test setup that continued even after the first round of December tests. We
were still not doing the calorimetry the way a heating and air conditioning
(HVAC) engineer tests a furnace. The HVAC engineer places the inlet
temperature sensor in a single point source. In our tests we did not know
precisely where the inlet air originated because we did not have a
concentrated point source. When we realized this, we constructed an inlet
duct. The inlet was initially about 20" x 6" located 6" below the bottom of
the furnace, in a source of cool air. We believe there is no heat path from
the Kinetic Furnace rotor or the calibration heater back to the thermocouples.
In runs with the calibration heater, the heat balance computed according to
the formula came out close to unity, with a COP between 96 and 106%. This
inlet duct draws warm air from the cloud surrounding the Kinetic Furnace and
its environs, but that makes no difference.
After installing the inlet duct and making other improvements, we tested
intensively for three days. Pope altered the pump several times, changing out
the rotor and water, but these changes had no effect, just as they had had no
effect in Bow. Based on these tests and the exhaustive testing in Bow, we
conclude that the three machines we have tested never produced excess heat. It
is possible that a Kinetic Furnace produced excess heat in earlier tests at
Pope’s facilities with the Air Techniques engineer, or in tests at Dunn
Laboratories, Inc., and elsewhere. Pope reports that during these tests, the
outlet duct was always passed through a plywood barrier in a window and vented
outside, so the error we observed in December could not have occurred.
Rotor heat up rates were similar to those measured in Bow, and rotor steady
state temperatures were nowhere near those reported by Pope (140-150° F).
Such high temperatures would be difficult to explain, except as apparent and
strong excess heat, but they could not be confirmed. This steady-state rotor
chamber temperature remains a key unresolved issue. If there are conditions
during which this temperature is higher than what we have seen, then it is
possible that Pope and Perkins saw better results. Attempts were made to
increase the rotor temperature by restricting the intake plenum cross-section
area. The rotor temperature was raised ~ 10° F by this method, but this
introduced another factor. The air moved much faster in the intake than the
exhaust, so it was cooled by the Bernoulli effect. This was seen during
calibration when the blower alone was run for an extended period. The COP came
out slightly over-unity for the blower alone because we did not take into
account the Bernoulli equations. The actual COP must obviously be under unity
for the blower.
Why It Took So Long ~
He reader might wonder why it took 9 weeks to confirm the conservation of
energy. Our test results in New Hampshire showed no significant excess heat at
any time, and the first test of the Kinetic Furnace in Georgia conclusively
proved there was no excess heat. On the face of it this is a simple,
straightforward measurement very similar to those conducted by HVAC engineers
every day, so you would think that an experienced engineer would get it right
the first time with ease. Indeed, Mallove and Wall did get it right the first
time. They spent the next 9 weeks making sure. The installation in Bow
included an inlet duct, so the apparent excess cannot be caused by the same
problem we fixed in Cumming. However, it was clear that the calibration heater
was also producing noisy, nominally over-unity results, putting 15% within the
range of error.
The apparent correlation of rotor RPM to slightly over-unity COP turned out
to be unconfirmed by a large number of other tests.
Another reason it took so long to resolve this issue is that people think
slowly and research takes time. Consider the electric generator and motor.
Oersted discovered that electric currents produce magnetic effects in 1820.
This triggered intense research by Henry, Faraday, Fresnel and other leading
scientists. It took Faraday roughly 10 years to prove the converse: that
magnets induce electric fields. Faraday devised the first crude electric
generator in 1831, and it was a while after that before anyone realized that
generators can also be used as motors.
Difficulties ~
Like most experiments, this was a running battle with recalcitrant
equipment, fatigue and inadvertent carelessness. Here are some of the things
that went wrong.
At first we measured the power into the resistance heater incorrectly,
because of the complicated network of two transformers and the autotransformer
(variable voltage transformer).
The power meter interface to the computer failed to work, perhaps because
of software conflicts with the HP 34970A, so we were unable to download
instantaneous power graphs. We depended on the computed power average and
total energy. Power input was very steady so this was not a significant
problem. In the previous visit with Mallove, power graphs downloaded
successfully and showed steady-state operation.
The computer interface to the anemometer also failed to function correctly.
The air speed changed every time we altered the configuration, and twice we
deliberately slowed down the blower by rewiring to increase heat retention in
the rotor housing. We thought this might promote excess heat generation.
Because we could not record data automatically from the anemometer to the
computer, each time we changed the wind speed we had to go through a laborious
20-minute process to manually record the data. It was measured in FPM
(ft/min.) with the DTA4000 electronic anemometer. The anemometer was mounted
on a camera tripod. The impeller was positioned at 9 points on a 3x3 array,
with points equidistant 3 inches apart. The impeller was placed at a grid
point and left to stabilize for one minute. Eight readings were taken at
15-second intervals. The average value and standard deviation was computed.
After the inlet duct was completed, 4 thermocouples were installed in
various locations within it. Wide variations and fluctuations in temperature
were noted. Apparently, eddy currents produced warm spots within the box. All
thermocouples were moved to spots exposed to the incoming rush of air, and the
temperatures all registered the same. However, they were probably all
registering a fraction of a degree cooler than they would have in the same air
motionless because of the Bernoulli effect. This fraction of a degree
difference might be mistakenly interpreted as excess heat.
The volume of air flowing through the duct every minute is computed by
multiplying the speed of the air, in feet per minute (FPM), by the size of the
duct in square feet, to give cubic feet per minute (CFM). However, the
cross-section of this duct was irregular. One side was slightly longer than
the others and the corners were not right angles. We straightened out the
corners somewhat with steel angle brackets. We traced the exact inside
dimensions of the duct onto a piece of plexiglass, copied that onto graph
paper, and determined surface area, which was 130.7 inches (91% of one sq.
ft.). When this correction factor was applied to the formula, the calibration
runs and Kinetic Furnace runs agreed to an uncanny extent. The numbers were so
close at one point that we worried we were making a mistake.
A section of Rothwells’ November report describes a typical instrument
malfunction:
"At minute 75, I placed the DTA4000 near the stool to measure ambient
temperature with the built-in thermometer. At minute 105, I discovered that
the milling machine nearby interfered with electronics in the control box.
When I lifted the control box, the temperature display changed from 71.6° to
71.1° F. I put it down again and it changed back to 71.6° , repeatedly. I
moved it a meter away and it dropped to 71.1° and remained stable.
The red alcohol thermometer registered 71° , and the Acu-rite registered
68.9° and 68.5° . I moved the stool two meters further inside the building,
to a location where all the instruments indicated the air was slightly colder,
and all reached the same spread of values they showed before the run: 70.7 on
the DTA4000, and 70° , 68.7° , and 68° on the others. In the new location the
anemometer moved with a slight draft of 70 FPM. The air was moving toward the
Kinetic Furnace".
This illustrates the importance of using instruments based on different
physical principles. We ues mercury thermometers as well as electronic
thermometers because mercury thermometers cannot be affected by the electric
fields generated by a milling machine.
How Heat Was Measured ~
We measured heat from the Kinetic Furnace by two methods. First, we simply
compared the control run to the Kinetic Furnace run at the same power level.
When the control run temperature went up 9.5° , the Kinetic Furnace went up
9.5° . When the flow air was restricted then the control run went up 19° ; the
Kinetic Furnace also went up 19° . Second, we applied the HVAC formula to
compute the actual heat flow. The formula is:
Delta T x 1.08 x FPM (air speed measured in feet per minute by anemometer)
x Duct opening as a fraction of one square foot = BTU heat output.
Here are two typical Kinetic Furnace runs:
December 5, Run 3
Input power 3.40 kW = 11,604 BTU/hr. Output power: 10.9° F x 1.08 x 1171
FPM x 0.91 sq. ft. = 12,509 BTU/hr.; COP = 108%.
This indicates no excess within the margin of error. In other words, some
of the resistance heater control run are also over 100%, and the standard
deviation of the anemometer readings was 46 FPM, so this result was between
106 and 110% Overall heat recovery from the system was excellent, so you would
expect the COP to be in the range of 90 to 100%.
December 5, Run 4
Input power 3.39 kW = 11,570 BTU/hr Output power: 10.1° F x 1.08 x 1171
FPM x 0.91 sq. ft. = 11,581 BTU/hr; COP = 100%.
Here is a calibration run with the resistance heater and a different
airflow:
December 8, Run 4
Input power = 3.34 kW = 11,399 BTU/hr Output power: 19.0° F x 1.08 x 1022
x 0.55 sq. ft. = 11,534 BTU/hr.; COP = 101%.
Future work, if time and resources allow, will be with water flow
calorimetry, which is easier and more precise. Air as a calorimetric fluid is
difficult to work with because it is turbulent, compressible, does not mix
well, is difficult to meter, and requires a huge duct. The flow of air through
the duct varies from one spot to another, and it varies over time. The
anemometer impeller is not large enough to cover the entire duct, so t is used
to sample the flow at many points. Flowmeters and temperature sensors immersed
in a stream of water also test a small sample of the flow at one point.
However, a stream of water can be diverted into a graduated cylinder to test
flow, and the fluid in the cylinder can be stirred to be sure the probes
correctly register the average temperature. You cannot divert the entire
stream of air into a container.
Once factors like the size of the duct cross-section were determined with
reasonable accuracy, the results from the calibration and Kinetic Furnace runs
at different power levels began to line up with unexpected accuracy. For
example, in the first set of tests they all showed COP of 96% within 1%.
Later, at another air speed, they lined up between 97 and 99%.
Instruments & Equipment ~
Although the test procedure is simple in principle, we took great care to
be sure we were getting the correct answer. One method of doing this is to use
redundant instruments based on different physical principles. For example, to
measure temperature one can rely upon high precision thermocouples with
confidence. In this case we only need to measure temperature to within 2 to 4°
F. A cheap thermometer will work adequately for this purpose. We did in fact
use some discount store thermometers, and one grade-school science class
thermometer. We also used 16 K-type thermocouples, 6 mercury thermometers of
various ranges, two bimetallic dial thermometers, a hand-held, high-precision,
high-temperature dual thermocouple (HP-52), and a red alcohol thermometer.
The HP-34970A thermocouple differences, as received, were less than 0.1
degree. The other instruments did not agree so precisely, varying as much as 3
F. In one test of ambient temperature, which was most accurate, the
thermocouples settled at 73.9° , 72.3° , 72.7° , and 72.0° ; the mercury
thermometer, which was the most accurate, settled at 72.3° ; and the red
alcohol which is marked in 2-degree increments, indicated 74° F. In a test of
the outlet duct temperature, the thermocouples and thermometers registered
82.4° , 82.9° , and the red alcohol thermometer which had a consistent 2° bias
at all temperatures, registered 84° F. At that moment the HP-34970A
thermocouples registered: 82.6° , 82.7° , 82.8° , 83.0° , 83.1° , and 83.0°
degrees. This 0.5° spread of value was real: temperatures within the air
stream did vary. The thermocouples agreed more closely when calibrated in
stirred water or left in calm, ambient air.
Even though the cheaper thermometers did have pronounced biases, each
agreed with itself. That is to say, when we moved a mercury thermometer, a
thermistor, and the red alcohol thermometer from the inlet to the outlet, they
all rose 9.5°, even though they started at different values. The cheaper
thermometers were inaccurate, but precise. "Inaccurate" means the starting
point in the temperature scale --- the absolute temperature was correct.
Precise means the temperature rose the same extent as the NAST traceable
thermometers.
Equipment used in this test included:
HP 34970A Data Acquisition system
11 K-type, 20-gauge thermocouples
Toshiba laptop computer interfaced to the HP 34970A
A Compaq portable computer to take notes and compute preliminary results with
a spreadsheet.
Mercury thermometers to measure ambient air
Amprobe DM-II recording power meter
Pacer Ind., Inc., model DTA4000 impeller anemometer. The built-in thermometr
was used in November, and malfunctioned
Amprobe "Ultra" clamp-on inductive analog ammeter and voltmeter, and a
Micronta clamp-on inductive analog ammeter and voltmeter. These instruments do
not detect power factor and they tend to overestimate electric power. However,
in the second set of tests in December, the results they showed were close to
the power measured with the more sophisticated Amprobe DM-II
Acu-rite dial thermometer with two thermocouples
Red alchol thermometer from ABC School Supply, Inc.
Dial thermometer on rotor chamber to measure the water temperature.
Two ducts made from 6’x4’ sheets of building insulating material
Stopwatch
Electronic camera
To calibrate, a variable voltage autotransformer, two transformers, and a duct
heater with 3.2 kW maximum output.
Was It Worth It?
We wrote above, "the actual COP must obviously be under unity for the
blower". A cynic might say that the actual COP of a water mixer must also
obviously be under unity, our tests were in vain, and we made a gargantuan
effort to prove the conservation of energy and the fixed ratio of work and
heat. This ratio was established in the 1840s by J.P. Joule. He used a falling
weight to drive a paddle that stirred water and raise the water temperature.
It sounds similar to the Kinetic Furnace --- it sounds as if we were trying to
overturn an observation established 150 years ago and confirmed countless
times every day by scientists and HVAC engineers everywhere. But, there is an
important difference between Joule’s experiments, stirred water, and ours. The
stirrer in the Kinetic Furnace rotates much more quickly than Joule’s, so
quickly that it almost certainly creates cavitation. Similar cavitation on the
smaller scale have apparently produced excess heat and nuclear effects. The
nuclear claim is controversial, but widely accepted. Much of the investigation
into apparent nuclear effects caused by cavitation is being performed in the
mainstream by conventional scientists, and approved of by the New York
Times, Scientific American, and Popular Science (e.g.,
P.S., December 1999). The Kinetic Furnace and Griggs Hydrosonic Pump
probably perform cavitation on a scale thousands of times larger than any of
the experimental sonoluminescence devices. We must say "probably" because we
have no direct proof that cavitation is occurring, because we cannot see
inside the steel chamber. Perhaps the Kinetic Furnace was previously
cavitating and producing excess heat, but it later stopped.
It would be absurd to question the validity of Joule’s experiments.
Cavitation has been carefully studied since it began damaging marine
propellers about 150 years ago. But, as far as we know, cavitation and heat
together have not been carefully researched. People have felt no need to study
heat evolved from cavitation, because no one suspected the heat might be
unusual. Science works a little like a national park. Thousands of people
cluster around the main attraction and the visitors’ center. Hundreds hike
down the nearby well-worn paths, measuring heat and cavitation. But the moment
you step off the path into the woods, you leave the crowds behind. In a
national park it is unlikely that you will stumble into an unexpected rock
formation or a hill that has never been climbed, but in a quiet spot you might
find a fossil or a new species of insect. The unexplored avenues of science
are infinitely larger than the physical paths on earth. The lesson of cold
fusion, the Marinov motor, and other strange phenomena described in this
magazine is that you can reach the unexplored wilderness of science in a few
minutes with simple tools.
Previous Tests & Recent Work By Pope & Perkins ~
The Kinetic Furnace reportedly produced large excess heat in other tests
over the years at Dunn Laboratories, Inc. (1982, 1983), Pittsburgh Testing
Laboratory (1984, 1986), Automated Test Labs (1986), and elsewhere. What
happened during those tests? Were the professional laboratories incorrect? We
do not know. In the papers provided to us by Pope, the tests are not described
in enough detail to judge with finality. It seems unlikely that professionals
in these laboratories made the same kinds of mistakes we did initially, before
we installed the inlet duct. After all, their business is to determine the COP
of furnaces. However, they never pursued development of the Kinetic Furnace.
That is inexplicable behavior. Other companies in the US that tested
over-unity cold fusion devices have been quite enthusiastic. Heating and air
conditioning companies have often contacted out magazine and asked whether any
practical device is available. They seem anxious to proceed with development,
and totally unconcerned about the fact that the scientific establishment does
not believe these devices exist. This is speculation, but perhaps after Dunn
Labs and the others wrote the reports provided to us by Pope, they realized
that they might have made a mistake of some sort. The HVAC engineer in Atlanta
who performed the tests on the Kinetic Furnace many years ago stood by his
work, but he explained that it was a preliminary test.
It is possible that for the past few years, Pope and the late Eugene
Perkins were performing invalid tests, and their results may have been
meaningless. They never employed a resistance heater or an inlet duct, so they
never would have caught the errors we discovered. They did not keep adequate
records by our standards, they had no computerized data collection, and they
did not organize their tests in a methodical, step-by-step fashion. To their
credit, they did the best they could at the time under difficult
circumstances.
Their open and cooperative attitude, and their willingness to honestly face
the facts is extremely laudable. Many inventors of exotic technology will
refuse to allow their machines to be tested in the first place, and even when
you find an error with a machine, most will refuse to listen or believe it.
Ralph Pope debated the issue with us at first, and he demanded rigorous proof
that the inlet temperature was not affected by radiant heat. This forced us to
devise a good test to prove our point, with the inlet turned 90° downward and
the thermocouples shielded from the furnace above. Pope accepts our conclusion
that the present set of experiments show no excess heat, but he believes that
previous experiments were successful. He intends to continue testing if he
can, and we will do so as time allows.
These results must be seen in the light of the James Griggs HydroSonic Pump
excess heat claims in a superficially similar device. Mallove and Rothwell
made measurements on the Griggs machine in early 1994. The Griggs results may
support the idea that cavitation excess energy is real but highly variable,
for reasons not yet understood.
HydroSonic Pumps have not yet been replicated widely. However, Griggs used
much better instruments and techniques than Pope-Perkins, and he uses water
flow calorimetry, which is easier and more reliable. We have a HydroSonic
Pump, and we intend to press ahead with our plans to test it at NERL when we
have the time and resources in 1999.
Should anyone be devoting weeks and thousands of dollars testing this sort
of claim? Mainstream science says no. We think it is worth doing. We are
disappointed, and we have no immediate plans to continue testing the Kinetic
Furnace at this time, but we do not consider these last few months a waste of
time. The instruments have been quite useful with other projects and the
skills and techniques for air flow calorimetry may prove valuable.
US Patent # 4,143,639
( Cl. 126/247 ~ 13 March 1979 )
Friction Heat Space Heater
Eugene Frenette
Abstract ~
A furnace or space heater is operable at low cost by a small electric motor
which rotates an elongated cylindrical drum on a vertical axis, within an
elongated cylindrical casing at a clearance of about one eighth of an inch in
the annular chamber formed therebetween. A supply of light lubricant normally
occupies the lower portion of the annular chamber but rises to fill the
chamber during rotation of the drum. The casing is enclosed in a housing,
having a fan chamber containing an electric motor and fan or blower. The motor
shaft may rotate both the fan and the drum.
Description ~
BACKGROUND OF THE INVENTION
It has heretofore been proposed in U.S. Pat. # 1,650,612 to Deniston of
Nov. 29, 1927 to rotate a stack of discs relative to a coaxial stack of fixed
discs on a horizontal axis within a casing to generate frictional heat in hot
water flowing through the lower portion of the casing. In this heating device
a supply of oil is contained in the upper portion of the casing to lubricate
the discs and to float on the water at a predetermined level.
In U.S. Pat. # 3,333,771 to Graham of Aug. 1, 1967, a pair of vaned rotors
are each enclosed within a chamber of a casing, and mounted to rotate in a
vertical plane on a horizontal axis as depicted in FIG. 7 thereof. As in the
Deniston patent water flows through the device and is heated by friction.
In U.S. Pat. # 4,004,553 to Stenstrom of Jan. 25, 1977 a single disc like
rotor is revolved on a horizontal axis in a vertical plane, within a casing to
heat water passing through the device.
SUMMARY OF THE INVENTION
Unlike the above mentioned patents wherein thin discs or vanes, in single
or stack configuration, comprise the rotor, in this invention an elongated,
cylindrical smooth surfaced, inner drum is the rotor. The drum is rotated in a
horizontal plane on a vertical axis within an elongated cylindrical, smooth
surfaced casing, or outer drum, to form an annular sealed, liquid, chamber
therebetween having a clearance of about one eighth of an inch. A quart of
relatively light oil is captive in the annular chamber and at rest occupies
only the bottom thereof. However upon rotation of the drum, by an electric
motor of about one horse power, the oil rises to fill the chamber due to the
pumping action of the drum.
Thus friction heat is generated not by two metal, or other, surfaces
contacting each other, but by the contact of the opposing surfaces with the
oil which not only lubricates but generates heat.
A portable space heater is formed by enclosing the casing and drum in the
lower chamber of a housing and drawing ambient air inwardly and around the
heated outer surface of the casing for fan discharge back into the ambient
atmosphere by a large diameter, eight bladed fan driven by the drum motor, or
preferably by a separate motor. For use as a furnace an air blower and
separate electric motor blow ambient air around the casing for discharge into
a heating system.
BRIEF DESCRIPTION OF THE DRAWING
FIG. 1 is a front elevational view of the portable space heater of the
invention, in half section;
FIG. 2 is a top plan view in section on line 2--2 of FIG. 1; and
FIG. 3 is a view similar to FIG. 1 of the device of the invention in its
preferred form.
DESCRIPTION OF A PREFERRED EMBODIMENT
FIGS. 1 and 2 illustrate one embodiment of the friction heat heater 20 of
the invention which includes an upstanding, hollow, cylindrical housing 21
formed of imperforate sheet metal 22 and having legs 23 for supporting it on a
floor 24 of a building. The space heater 20 is portable and in the portable
embodiment illustrated in FIGS. 1 and 2 the housing 21 is of predetermined
diameter of about twelve inches and of predetermined height of about
thirty-two inches.
Fixed within housing 21 by suitable brackets 25 and 26 is a hollow
cylindrical casing, or outer drum, 27 which is of predetermined diameter less
than the diameter of the housing, such as ten inches, and is formed of
aluminum sheeting 28 for efficient transfer of heat. The cylindrical side wall
29, top wall 31 and bottom wall 32 of casing 27 are imperforate to form a
sealed enclosure except for the filler tube 33, which is closed by a removable
threaded cap 34.
The casing 27 divides housing 21 into the lower air heating chamber 35,
which it occupies and an upper fan chamber 36, there being an annular air
chamber 37 formed between the cylindrical side wall 29 of the casing and the
coaxial, concentric cylindrical side wall 38 of the housing 21.
Air inlet means 39 is provided in the lower portion of the housing 21 in
the form of spaced apertures 41 extending around the cylindrical side wall 38
and air outlet means 42 is provided in the top 43 of the housing in the form
of apertures 44. The annular air chamber 37 connects the air inlet means to
the air outlet means of the fan chamber 36.
A reversible electric motor 45 is mounted in the fan chamber 36 with an
eight bladed fan 46 fast on one end 47 of the motor shaft 48, each blade being
of about 25° pitch and the motor being about one horse power for rotating the
shaft 48 at between 1800-3600 R.P.M.
The other end 49 of motor shaft 48 extends into the air heating chamber 35
to rotate the hollow, cylindrical drum 51 which is supported in suitable
bearings 52 for rotating around the central, vertical axis of the casing 27
and housing 21.
The inner drum 51 is sealed and hollow and includes the top wall 53, bottom
wall 54 and cylindrical side wall 55, the walls being of stainless steel. The
exterior cylindrical surface 56 of the cylindrical side wall 55 is smooth as
is the interior, cylindrical surface 57 of the aluminum of the cylindrical
side wall 29 of casing 27 and the surfaces 56 and 57 are at about one eight
inch clearance from each other to form a narrow, annular liquid receptacle 58
therebetween.
It should be noted that the annular liquid receptacle 58 is not a passage
through which liquid to be heated is continually flowed, as in the above
mentioned prior art patents. Instead it is a sealed chamber and is provided
with a supply of liquid lubricant 59 such as a quart of No. 10 oil which
normally rests in the horizontal space, or shallow liquid receptacle 61
between the bottom wall 54 of the drum 51 and the bottom wall 32 of the casing
27.
It has been found that the best results are obtained when the lubricant 59
is Quaker State F-L-M-A-T Fluid, Ford Motor Company Qualifications No.
2P-670306 M 2633F. Unlike prior patents, no water is in contact with the oil.
The motor 45 is connected to a thermostat 62, of any well known type by
cord 63 and to a source of electricity by male plug 64 so that it is energized
under the control of ambient temperature by the signals of the thermostat.
In operation the motor 45 drives the drum 51 at a substantial speed, which
causes the oil 59 to rise up into the annular liqud receptacle 58 to
substantially fill the same. The heat of friction between the inner drum 51
and outer drum, or casing 27 is transferred by the oil while it prevents wear
on the surfaces 56 and 57 so that the exterior aluminum surface 65 of the
fixed outer drum 27 becomes heated. Meanwhile the large diameter, multibladed
fan 46 is drawing ambient air through the air inlet means 39, thence up
through the annular air chamber 37 and past the elongated heated surface 65
for discharge through the air outlet means 42 back into the room.
As shown in FIG. 3, it is preferable to provide a separate electric motor
70, usually about 1/8 H.P. and driving an air blower 71, these being mounted
in a lower air chamber 72 for driving ambient air upwardly in an annular flow
path in chamber 37 from the air inlet means 73 to the air outlet means 74. Air
outlet means is the intake duct 75 of a hot air heating system 76 so that the
heater 20 becomes a furnace rather than a space heater, the separate electric
motor 70 enables the thermostat 62 to initiate rotation of the drum until a
predetermined temperature is reached in the aluminum outer drum 27, whereupon
the thermostat automatically de-energizes the drum motor 45 while continuing
to rotate the separate fan, or flower motor such as 70, to furnish hot air to
the room or heating system 76 until the casing 27 cools to a predetermined
temperature.
US Patent # 4,424,797
( Cl. 126/247 ~ 10 January 1984 )
Heating Device
Eugene Perkins
Abstract ~
A heater for heating a liquid including a housing defining a closed
elongate heating chamber therein with a cylindrical chamber surface, a rotor
body rotatably journalled in the heating chamber with a cylindrical peripheral
surface thereon concentrically of the chamber surface so as to define an
annular space between the chamber surface and the peripheral surface on the
rotor body, drive means for effecting relative rotation between the rotor body
and the housing, and pump means for circulating the liquid through the annular
space so that the rotation of the rotor body heats the liquid passing through
the annular space.
Description ~
BACKGROUND OF THE INVENTION
This invention relates generally to liquid heaters and more particularly to
a liquid heater which heats liquid by shearing the liquid.
Various attempts have been made in the past to mechanically heat liquids.
One type of such mechanical heating device heats the liquid by shearing the
liquid between rotary and stationary blades in a chamber. A device of this
type is illustrated in U.S. Pat. No. 2,683,448. This type of heating device
creates a high degree of turbulence in the liquid passing through the device
to be heated and consumes a large amount of power in driving the rotary blades
in the chamber. As a result, the heating efficiency of this type of device is
relatively low.
In another type of these prior art devices, the heat to heat the liquid is
generated by the frictional contact between rotating and non-rotating members.
Examples of this type of heating device are illustrated in U.S. Pat. Nos.
2,625,929; 3,164,147; and 3,402,702. The problems with this type of heating
device are that a large amount of power is consumed in generating the
frictional heat, and excessive wear is encountered between the surfaces of
frictional contact with each other within the heating unit.
SUMMARY OF THE INVENTION
These and other problems and disadvantages associated with the prior art
are overcome by the invention disclosed herein by providing a heating unit
which uses a cylindrical rotor rotating in a cylindrical heating chamber so
that the flow of liquid in the chamber is laminar rather than turbulent and
with the rotor and chamber not being in contact with each other so that
frictional losses within the heating unit are minimized. It has been found
that sufficient liquid shear is generated by the rotating rotor in the heating
chamber so that the liquid is heated, yet the power consumption associated
therewith is minimized so that the heating efficiency of the unit is
maximized.
The apparatus of the invention includes a heating unit which may be
incorporated in a heating system adapted to heat air in a prescribed space
such as a building or residence. The heating unit includes a housing which
defines an elongate heating chamber therein with a cylindrical chamber
surface. A rotor body is rotatably mounted in the heating chamber and defines
a cylindrical peripheral surface thereon concentric with respect to the
cylindrical chamber surface. The peripheral surface on the rotor has an
outside diameter a prescribed amount smaller than the inside diameter of the
chamber so as to define an annular space between the rotor body and the
chamber through which the liquid to be heated is passed. Drive means is
provided for effecting relative rotation between the rotor and the housing and
pump means is provided for circulating the liquid through the annular space
between the rotor and the chamber as the rotor is rotated so that the liquid
is heated due to the shear of the liquid in the annular space between the
rotor body and the chamber. In the embodiment of the invention shown, the pump
impeller for circulating the liquid through the chamber is mounted on the
rotor so that the drive means simultaneously rotates the pump impeller and the
rotor.
When the heating unit is incorporated in a heating system, the liquid
heated by the heating unit is passed through an air-to-liquid heat exchanger
through which the air to be heated is also passed so that the air is heated as
it passes through the heat exchanger. The operation of the heating unit is
controlled so as to maintain the temperature of the air exiting the heat
exchanger within a prescribed temperature range while the operation of the fan
circulating the air through the heat exchanger is controlled in response to
the temperature of the air in the conditioned space so as to maintain the
temperature of the air in the conditioned space within a prescribed
temperature range.
These and other features and advantages of the invention will become more
apparent upon consideration of the following description and accompanying
drawings wherein like characters of reference designate corresponding parts
throughout the several views and in which:
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a view illustrating the invention incorporated in a heating
system;
FIG. 2 is a longitudinal cross-sectional view of the heating unit of the
invention;
FIG. 3 is a transverse cross-sectional view taken generally along line 3--3
in FIG. 2; and
FIG. 4 is a transverse cross-sectional view taken generally along line 4--4
in FIG. 2.
These figures and the following detailed description disclose specific
embodiments of the invention; however, it it to be understood that the
inventive concept is not limited thereto since it may be incorporated in other
forms.
DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS
Referring to FIG. 1, it will be seen that the invention is embodied in a
heating system 10 used to heat air in a space to be conditioned such as a
building or residence. The heating system 10 includes generally a heating unit
11 connected to a liquid-to-air heat exchanger 12. The liquid-to-air heat
exchanger 12 is housed in an appropriate duct system 14 adapted to supply air
from the space to be conditioned to the heat exchanger 12 and to deliver air
from the heat exchanger 12 back to the space to be conditioned. A fan 15 is
provided in the duct system 14 for forcing the air from the space to be
conditioned through the duct system 14 and the heat exchanger 12. The heating
unit 11 is also illustrated housed in the duct system 14 although it is
understood that it may be located remotely thereof.
The duct system 14 defines a heat exchanger chamber 16 therein in which the
liquid-to-air heat exchanger 12 is mounted with an intake plenum 18 connected
to the space to be conditioned by an appropriate return duct 19 so that the
air from the space to be conditioned is supplied to the heat exchanger chamber
16 through the intake plenum 18. The air passing from the intake plenum 18
through the heat exchanger 12 in the chamber 16 passes out through a supply
plenum 20 connected to the space to be conditioned by the supply duct 21 to
supply the heated air back to the space to be conditioned. The fan 15 is
located in the heat exchanger chamber 16 so that the fan 15 forces the air
from the intake plenum 18 through the heat exchanger 12 in the chamber 16 and
out through the supply plenum 20. It will be noted that the heat exchanger 12
extends completely across the chamber 16 so that all of the air passing from
the intake plenum 18 to the supply plenum 20 must pass through the heat
exchanger 12.
The operation of the fan 15 is controlled by thermostatic switch 22 which
is located in the space to be conditioned so that when the temperature of the
air in the space to be conditioned drops below a prescribed value, the switch
22 operates fan 15 to circulate air from the space to be conditioned through
the heat exchanger 12 until the air in the space to be conditioned has been
raised to a higher prescribed value. Such thermostatic switches 22 are
conventional and need not be described in detail. As will become more
apparent, the operation of the heating unit 11 is controlled by a thermostatic
switch 24 located at the air exit side of the heat exchanger 12 as will become
more apparent. The thermostatic switch 24 serves to activate the heating unit
11 when the air exiting the heat exchanger 12 drops to a prescribed lower
temperature to heat a liquid and supply the liquid to heat exchanger 12 until
the temperature of the air exiting the heat exchanger 12 has been raised to a
prescribed higher temperature.
The heating unit 11 is illustrated mounted in the heat exchanger chamber 16
under the heat exchanger 12 and includes a liquid heater 25 driven by drive
motor 26. In the particular embodiment shown, the drive motor 26 is connected
to the liquid heater 25 through a bell and pulley arrangement 28. It is to be
understood, however, that the drive motor 26 may be directly connected to the
liquid heater 25.
As best seen in FIGS. 2-4, the liquid heater 25 includes a housing 30 in
which is rotatably mounted a rotor assembly 31. The housing 30 is fixedly
mounted in the heat exchanger chamber 16 while the rotor assembly 31 is
rotated by the drive motor 26.
The housing 30 includes a cylindrical side wall 32 closed at opposite ends
by end plates 34. Each of the end plates 34 defines a cylindrical projection
35 thereon which fits within the cylindrical side wall 32 and is provided with
an annular groove 36 therearound which receives an O-ring 38 therein to seal
the end plate 34 to the inside of the side wall 32. The end plates 34 are held
in position by tie bolts 39 so that the closed chamber is defined by the side
wall 32 and end plates 34. This chamber is divided into a heating chamber 40
and a pumping chamber 41 by a divider assembly 42. The divider assembly 42
includes an annular spacer wall 44 having an outside diameter so that it will
snugly fit within the side walls 32 adjacent one of the end plates 34 so that
spacer wall 44 projects a prescribed distance away from the end plate 34. The
projecting end of the spacer wall 44 is closed by a circular end plate 45 so
that the pumping chamber 41 is defined between the end plate 45, spacer wall
44, and the end plate 34 against which the spacer wall 44 abuts. The heating
chamber 40 is thus defined between the end plate 45, the end plate 34 opposite
that against which the divider assembly 42 abuts and the housing side wall 32.
The heating chamber 40 has a diameter d.sub.1 defined by the inside surface 48
of the side wall 32 and a length L.sub.1 defined between the end plate 34 and
the end plate 45. The side wall 32 defines an inlet opening 49 therethrough to
the chamber 40 adjacent that end plate 34 opposite the divider assembly 42
while the spacer wall 44 and side wall 32 define a common outlet opening 50
therethrough which communicates with the pumping chamber 41. The circular end
plate 45 on the divider assembly 42 defines a transfer opening 51 therethrough
about the central axis A.sub.1 of the chambers 40 and 41 of diameter d.sub.2
so that the heating chamber 40 communicates with the pumping chamber 41 as
will become more apparent.
The rotor assembly 31 includes a support shaft 55 which mounts a rotor body
56 thereon at one position along the length of the shaft 55 and a pump
impeller 58 at another position along the support shaft 55. The rotor assembly
31 is mounted in the housing 30 so that the support shaft extends coaxially of
the axis A.sub.1 with the rotor body 56 located in the heating chamber 40
while the pump impeller 58 is located in the pumping chamber 41. The support
shaft 55 extends through the transfer opening 51 through the end plate 45 in
clearance therewith so that liquid can pass from the heating chamber 40 into
the pumping chamber 41 and extends out through the end plates 34 through
appropriate openings therein. The shaft 55 is rotatably journalled in bearings
59 mounted on each of the end plates 34 and held in position by retainers 60
on the outside of the end plates 34. A seal 61 is provided around shaft 55
immediately inboard of each of the bearings 59 to prevent liquid from passing
out of the housing 30 around the shaft 55 at the end plates 34. The shaft 55
is provided with a drive projection 62 which extends out of the housing 30
through one of the retainers 60 so that the belt and pulley arrangement 28 can
be connected thereto to rotate the support shaft 55.
The rotor body 56 is hollow and includes a pair of spaced apart
washer-shaped end plates 64 which are fixedly attached to that portion of the
support shaft 55 within the heating chamber 40 with one of the end plates 64
spaced inwardly of the end plate 34 and the other end plate 64 being spaced
inwardly of the end plate 45. The end plates 64 are connected by an annular
rotor side wall 65 which extends therebetween with the side wall 65 being
fixedly attached to the end plates 64 and the end plates 64 being fixedly
attached to the support shaft 55 so that the rotor body 56 rotates with the
support shaft 55. The rotor side wall 65 defines a peripheral surface 66
thereon which is cylindrical and located concentrically of the central axis
A.sub.1 of the heating chamber 40. The surface 66 has a diameter d.sub.3 which
is a prescribed amount less than the inside diameter of the surface 48 so that
surfaces 66 and 48 defines an annular space 68 therebetween of a radial
distance d.sub.4. The surface 66 has a length L.sub.2 shorter than the length
of the heating chamber 40.
The pump impeller 58 is fixedly attached to that portion of the support
shaft 55 within the pumping chamber 41 and includes a disk portion 70 oriented
perpendicular to the axis A.sub.1 with an outside diameter slightly smaller
than the inside diameter of the spacer wall 44 so that the pump impeller 58 is
freely rotatable with shaft 55 in the pumping chamber 41. The pump impeller 58
also includes an attachment portion 71 used to attach the pump impeller 58 to
the support shaft 55 through an appropriate key arrangement. The disk portion
70 defines a centrally located counterbore 72 therein which opens onto that
side of the disk portion 70 facing the circular end plate 45. The counterbore
72 has a diameter larger than that of the support shaft 55 to define an
annular cavity in the disk portion 70 around the shaft 55. The disk portion 70
further defines a plurality of radially extending passages 74 therein which
open at their inboard ends into the counterbore 72 and open at their outboard
ends into the outer periphery of the disk portion 70. The pump impeller 58 is
attached to the support shaft 55 so that the passages 70 are aligned with the
outlet opening 50 as they rotate within the pumping chamber 41. It will be
seen that the diameter of the transfer opening 51 and the diameter of the
counterbore 72 are such that liquid can freely pass from the heating chamber
40 through the transfer opening 51 and into the counterbore 72 so that the
liquid will be forced outwardly along the passages 74 as the pump impeller 58
is rotated with the support shaft 55. As will become more apparent, this
serves to force the liquid out of the housing 30 through the outlet opening
50. The outlet opening 50 is connected to one side of the heat exchanger
through a supply pipe 75 while the inlet opening 49 to the housing 30 is
connected to the other side of the heat exchanger through the return pipe 76.
In operation, it will be seen that the heating chamber 40 and the pumping
chamber 41 as well as the passage through the heat exchanger and the pipe 75
and 76 are filled with a liquid to be heated such as water. When the drive
motor 26 rotates the rotor assembly 31, this causes the rotor body 56 to be
rotated in the heating chamber 40 while the pump impeller 58 is rotated in the
pumping chamber 41. The pump impeller 58 pumps the liquid through the liquid
heater 25 to the heat exchanger 12 and then back to the liquid heater 25 so
that the heating chamber 40 and pumping chamber 41 remain filled with liquid
at all times. As the rotor body 56 is rotated via the drive motor 26, the
liquid at the cylindrical peripheral surface 66 on the rotor body 56 tries to
move with the rotor body 56 while the liquid at the inside surface 48 on side
wall 32 tries to remain stationary. This establishes a velocity gradient in
the liquid across the annular space 68 between the rotor body 56 and the
inside surface 48 of the side wall 32 to establish shear forces within this
liquid. These shear forces cause the liquid to be heated. The velocity profile
across the annular space 68 is such that the liquid in the annular space 68
remains in the laminar flow region so as to minimize the power consumption of
the liquid heater 25. Thus, it will be seen that the liquid in the annular
space 68 is being moved longitudinally of the annular space 68 by the pump
impeller 58 while the liquid is moving circumferentially about the space 68 by
the rotor body 56. This heats the liquid in the annular space 68 as it flows
therealong and then flows out of the heating chamber 40 into the pumping
chamber 41 where the pump impeller 58 pumps the liquid through the heat
exchanger 12 so that the heat from the liquid can be transferred to the air
passing through the heat exchanger 12.
It has been found that the temperature to which the liquid can be heated in
the annular space 68 is dependent on the relative velocity of the cylindrical
peripheral surface 66 with respect to the inside surface 48 on the side wall
32. When water is used as the liquid, rotating surface 66 at a velocity of
about 1150 feet per minute heats the water to a temperature of about 140° F.,
rotating surface 66 at a velocity of about 1800 feet per minute heats the
water to about 165° F., and rotating surface 66 at a velocity of about 2550
feet per minute heats the water to a temperature of about 210° F. Thus, it
will be seen that the temperature to which the water can be heated can be
adjusted by adjusting the rotational speed of the rotor body 56 to adjust the
velocity of the peripheral surface 66 on the rotor body 56.
The radial distance d.sub.4 of the annual space 68 affects the volume of
liquid that will be heated by the rotating rotor body 56 at any one time.
Distances of 0.06-1.0 inch for the distances d.sub.4 have been found practical
to reasonably heat the liquid passing through the annular space 68. A distance
d.sub.4 of about 0.75 inch has been found preferable to heat the liquid at a
flow rate of about two gallons per minute.
The heating rate capacity of the liquid heater 25 is also dependent on the
velocity of the cylindrical peripheral surface 66 on the rotor body 56. When
water was used as the liquid to be heated, a velocity of about 1800 feet per
minute generated about 19,000 BTU per hour whereas rotating the surface 66 at
a velocity of about 2550 feet per minute generated about 25,500 BTU per hour.
The volume of liquid in the liquid heater 25 and the system of the heat
exchanger 12 and the liquid heater 25 should be such that the air passing
through the heat exchanger 12 at a prescribed volumetric rate can be heated
over the desired temperature differential. It is found that liquid heater 25
holding about one gallon of liquid with the system holding about three gallons
of liquid is sufficient to heat air passing through the heat exchanger 12 at a
volumetric rate of about 300 cfm about 40°-80° F. with a temperature
differential in the liquid passing through the heat exchanger 12 of about
15°-20° F.
In the system illustrated, the diameter d.sub.1 is about 5.5 inches, the
diameter d.sub.3 is about 4 inches, and the length L.sub.2 of the surface 66
is about 6 inches. The drive motor 26 operates from a 115 volt power source
and draws about 5.5 amps to rotate the rotor assembly 31 at about 2400 rpm to
move the peripheral surface 66 on the rotor body 56 at a velocity of about
2550 feet per minute. Thus, the drive motor 26 has a power consumption of
about 0.6 kilowatt per hour to produce a heating output of about 25,500 BTU
per hour. In the above system, the fan 15 was operated to force air through
the heat exchanger 12 at a flow rate of about 300 cfm. With the rotor assembly
31 rotating at about 2400 rpm, the air passing through the heat exchanger 12
was heated from a temperature of about 60.degree. F. to a temperature of
100°-145° F. while the water temperature supplied to the heat exchanger 12
from the liquid heater 25 was at a temperature of about 210° F. and the
temperature of the water returned to the liquid heater 25 from the heat
exchanger 12 is at a temperature of about 185° F. At this rotational speed,
the pump impeller 58 was pumping the water at a flow rate of about 2 gpm with
a pressure differential of about 0.5 psi across the impeller 58. The
thermostatic switch 22 in the space to be conditioned was set to maintain the
temperature of the air in the space at about 71° F. while the thermostatic
switch 24 was set to start operation of the liquid heater 25 when the
temperature of the air exiting the heat exchanger 12 dropped to about 100° F.
and to stop operation of the liquid heater 25 when the temperature of the air
exiting the heat exchanger 12 reached about 140° F. Typically, the operating
cycle for the fan 15 was about 10-12 minutes with the liquid heater 25 being
operated for about two cycles of 1-2 minutes each during each operating cycle
of the fan.
