 |
Description  |
|
|
BACKGROUND OF THE INVENTION
(1) The Field of the Invention comprises heat exchangers for purposes of
recycling waste heat, cooling, and ventilating.
(2) Prior Art
So far as is presently known, conventional heat exchangers employ heating
tubes developed in the last century by A. M. Perkins which, as described
in U.S. Pat. No. 1,872,363 to Thurm, issued Aug. 16, 1932, comprise a
straight evacuated tube containing a suitable heat transfer agent and
hermetically sealed at both ends. By exposing one end of the tube to heat,
the agent is vaporized, the vapors travel to the other cooler end where
they are condensed, giving up heat of condensation which may be recovered.
The heated end is usually termed the evaporator and the cooled end the
condenser; and if the evaporator is lowered below the condenser, the
action is improved by having gravity aid the return of the condensate to
the evaporator. These Perkins tubes are used in present day commercial
heat exchangers, such as those described in U.S. Pat. Nos. 3,788,388 and
3,865,184, which show operation at orientations ranging from the
horizontal to positions in which the evaporator is slightly elevated, or
tilted, above the condenser, these tilt positions being described as
useful to reduce the efficiency of the unit, even to shutting off the
same. Obviously, such elevations are adverse to high heat exchanger
capacity because they retard or eliminate the above mentioned effect of
gravity. The user of these commercial units is instructed that a favorable
tilt, i.e., evaporator below the condenser, is necessary to provide
adequate capacity, or to operate in winter, and that a tilt control
mechanism is required for year round, i.e., both summer and winter,
operation, for frost prevention on the weather face of the exhaust side of
a unit, and for regulating the temperature of supply air leaving the unit
(and entering a building) to avoid over-recovery of heat. Such a
mechanism, which is described in said 3,788,388 patent, comprises at least
eight structures, not including a number of flexible connections between
the unit and the ductwork, and is expensive as well as detractive of the
valued passive quality of the heat exchanger and conducive to a degree of
long term unreliability.
It is clear, therefore, from the foregoing that operation of a heat
exchanger with Perkins tubes at a horizontal orientation will not provide
a desired high heat transport (a result confirmed by work described
below), that for high heat transport, among other cases, a favorable tilt
is required which for automatic operation necessitates use of
tilt-producing and tilt control mechanism, and that operation is reduced
or even shut off at unfavorable or adverse tilts. This last result raises
a separate problem of unavoidable, accidental, and/or initially undetected
adverse tilts which may extend up to 0.5 or 1 inch, or more, and which may
be and frequently are present owing to the difficulty, especially with
tubes of long length, of determining true horizontal and/or of permanently
maintaining it; various reasons may account for the difficulty, such as
imprecise use of levelling tools by workmen, or the presence of slight
undetected damage to the tubes, or the tendency of buildings to shift, and
the like. A heat exchanger employing conventional Perkins tubes is unable
to operate effectively, or even at all, in the presence of accidentally
produced adverse tilts.
The foregoing disadvantages are avoided by the heat exchanger, and elements
therefor, described herein which operates at a horizontal orientation and
produces high heat transport without need for tilt mechanisms, and which
is substantially insensitive to accidentally produced adverse tilts. Among
these and other advantages to be described is the provision of an element
having a significantly greater driving head, and thus better heat
transport, than a Perkins tube.
SUMMARY OF THE INVENTION
Considering the elements which make up the heat exchanger, which is a tube
in the form of a closed loop having a pair of substantially parallel arms
connected to each other at their ends by a pair of end pieces or members,
with one arm disposed above the other, and having a working agent
initially present in the lower arm. A suitable number of loops is disposed
in and supported by a frame, usually of rectangular shape, to form the
heat exchanger. In use, the unit is usually disposable in a wall between
two spaces between which an air interchange is to be made, one of which
spaces may be a building or a room therein and the other the outdoors,
although both may be buildings or rooms either in the same or different
buildings. A first airway is present in the unit for transferring air in
one direction, a second airway for transferring air in the opposite
direction, and means for keeping the airways separate. The loop elements
extend into and between these airways for transferring heat from the air
in one airway to the air in the other. The portions or sections of the
loops disposed in the first airway may comprise the evaporator section for
the agent and receive heat from heated air passing thereover, and the
other portions or sections of the loops disposed in the second airway may
comprise the condenser section for giving up heat to the air passing
thereover. In the evaporator section the agent is vaporizable in the lower
arms of the loops with resulting vapors flowing upwardly through adjacent
end members to the upper arms and thence along said upper arms to the
condenser section where they are condensed and the condensed agent then
flowing down the end members adjacent thereto to the lower arms of the
condenser section and back to the evaporator section, and as may be
apparent, the agent in the forms of vapor and condensate flows cocurrently
through each loop in the same direction. Such flow is distinctive,
contributing greatly to the described advantages. In a Perkins tube on the
other hand the agent vapors flow in one direction and the condensate in
the opposite direction, so that a vapor shear effect is created which, if
the vapor velocity is great enough, will prevent part or all of the
condensate from flowing back to the evaporator, despite the assistance of
gravity.
BRIEF DESCRIPTION OF THE DRAWINGS
The invention is illustrated in the accompanying drawings, which are
diagrammatic, and in which
FIG. 1 is an enlarged broken cross-sectional view of a single loop element
in a heat exchanger environment;
FIG. 2 is an enlarged broken cross-sectional view of a single conventional
Perkins tube;
FIG. 3 is a side view, in section, of a heat exchanger employing the loops
of FIG. 1 and taken along line 3--3 of FIG. 4;
FIG. 4 is a sectional view along line 4--4 of FIG. 3;
FIG. 5 is a sectional view along line 5--5 of FIG. 3;
FIG. 6 is a diagram showing the disposition of the upper and lower arms of
a loop relatively to each other and to the horizontal;
FIG. 7 is an enlarged fragmental cross-sectional view of the fill tube of
FIG. 1 in the open and closed positions;
FIG. 8 is a sketch of test apparatus for Example 1;
FIG. 9 is a sketch of test apparatus for Example 3; and
FIG. 10 is a perspective view of test apparatus for Example 4.
DESCRIPTION OF SPECIFIC EMBODIMENTS
As shown in FIG. 1, the heat exchanger element 10 comprises a tube 11 in
the form of a closed loop consisting of upper and lower arms or legs 12,
13 connected at their ends by end members or pieces in the form of bends
14, 15, with the arm 12 disposed over the arm 13. A working agent 16 is
present, initially lying in the lower arm, but which is shown as having a
different distribution, to be discussed below. End panels of the heat
exchanger 9 are seen at 17, 18 and a central barrier at 19, each of these
being suitably apertured to receive and support the loop. At 20, 21 are
groups of closely spaced apertured fins engaged by the arms of the loop,
and at 22 is a portion of a fill tube for evacuating, charging, and
sealing the loop; it is described in more detail below.
FIG. 3 shows a row of loops, six in all, in the heat exchanger, which form
one vertical row, while FIG. 4 shows five such vertical rows, numbered 1
to 5, all partially enclosed by a frame comprising end walls 25, 26 and
top and bottom walls 27, 28. The sides or faces 23, 24 of the unit are
open. The number of loops per row, and rows per unit, are variable. Fins
20, 21 may be seen to extend from top to bottom, each preferably in the
form of a continuous apertured plate or sheet and spaced about 12 to the
inch, although such spacing is variable. The central thermal barrier 19
divides the unit into two sections or halves 29, 30, the former defined by
end panel 17, barrier 19, and walls 27, 28, and comprising a first airway
for transferring air through the spaces between loops, and the latter
defined by panel 18, barrier 19, and walls 27, 28, and comprising a second
airway. If the section or portion 29 is regarded as the evaporator or
evaporator section 31, and section or portion 30 as the condenser or
condenser section 32, it is apparent that the loops extend into and
between the two airways and are capable of transferring heat from the air
in one to the air in the other.
FIG. 5 shows how the vertical rows 1 to 5 are vertically offset with
respect to each other. Considering loop arms 34, 35, and 36, it may be
seen that they form a triad; similarly, arms 37, 38, and 39 form another
triad; and in fact such triad groups of arms may be discerned throughout
FIG. 5; and if one connects the centers of the arms of each triad to form
a triangle, it will be found to be an equilateral triangle each of whose
angles is 60.degree.. This is a preferred arrangement for good air-loop
contact, although other offset row dispositions are suitable.
Returning to FIG. 1, which is intended to depict steady state operation of
a heat exchange loop, if warm exhaust air is flowing through airway 29,
say in a direction out of the paper, and cool fresh air is flowing through
airway 30, say into the paper, as in winter operation, the agent in the
lower arm 13 of evaporator 31 is heated and vaporized. The resulting
vapors flow upwardly in the adjacent end bend 14, note arrow A, to upper
arm 12, then along such arm to condenser 32, arrow B, where they are
condensed, giving up heat of condensation to the incoming cool fresh air
in airway 30, then the condensate flows downwardly through return bend 15
to the lower arm 13 of the condenser, and then back to the evaporator.
Considering the foregoing circuit in a bit more detail, note that as
vapors from the evaporator reach the bend 14, they rise because lighter
than the bulk liquid, in turn tending to produce a suction effect which
draws more vapors to bend 14. At lower heat inputs, only vapors pass from
bend 14 to upper arm 12, but at higher inputs some liquid may be entrained
by the vapors; however, since arm 12 is also being heated in the
evaporator section, such entrained liquid is converted to vapor. Then in
arm 12 of the condenser, where liquid condensate is formed, the flow of
condensate to the right is favored by the push or movement of the vapors
coming from the evaporator, and such flow together with an assist from
gravity, owing to the liquid in bend 15, helps move the condensate down
the return bend 15 to the lower arm 13 of the condenser. At that position
in the condenser, the condensate may be considered to be a subcooled
liquid, i.e., liquid at a temperature below boiling temperature. Notice
that the continuous condensation of vapors tends to create a suction
effect and thus to require more vapors to be generated in the evaporator.
Notice, too, that the flow in arm 13 is from right to left, that is,
liquid from the condenser is fed to the evaporator from the side opposite
the side where vapors are formed, or to put it another way, from the side
where the vapor velocity is zero. This general movement to the left in arm
13 also favors the above described vapor flow upwardly in bend 14. A study
of the agent flow in a loop is reported in Example 1.
It is evident that both vapors and condensate flow cocurrently in the same
direction through the loop. Such flow, as described, is distinctive by
comparison with that in the conventional horizontally orientated Perkins
tube 40 of FIG. 2, which is intended to have the same length, diameter,
and wall thickness as one of the loop arms of FIG. 1. In tube 40, consider
end 41 as the evaporator heated by warm air and end 42 as the condenser
cooled by cooler air. The agent 43 is vaporized and the vapors are driven
toward the condenser, arrow C, while the returning condensate flows back
toward the evaporator, the two being in counterflow. The vapor exerts a
shearing effect on the condensate, akin to that of a wind blowing over a
lake. If the vapor velocity is high enough, the condensate can be
prevented, partly or completely, from flowing back to the evaporator, a
result particularly noticeable at a horizontal orientation of the Perkins
tube, but also possible at a favorable tilt. The vapor shear simply
dominates and reverses the flow of condensate, regardless of the
beneficial aid of gravity, so that the condensate collects or bunches up
at the condenser end with no place to go. This phenomenon is in contrast
to the vapor flow of the loop of FIG. 1 where condensate returning to the
evaporator for reboiling cannot be impeded by the vapor velocity but
rather is aided by it.
The vapor shear effect in the Perkins tube may be even more readily
appreciated when it is seen that the liquid driving head is equal to only
one tube diameter, i.d., as represented by the distance PD of FIG. 2,
comprising the inside diameter of tube 40, a head which is rather easily
overcome by the countercurrently flowing vapors. It is to increase the
head PD that the art applies a favorable tilt to conventional Perkins
tubes. On the other hand, the liquid driving head in the loop of FIG. 1 is
equal to the distance S plus LAD plus at least half of UAD, where LAD is
the lower arm diameter, i.d., and UAD is the upper arm diameter, i.d. The
level of liquid in the upper arm at the right hand end of condenser 32 is
variable, and as shown, may range from the level at d.sub.1, which is
about 1/2 the arm diameter, to the level at d.sub.2, which is the full
diameter. S is the distance between near adjacent surfaces of the loop
arms. The liquid driving head is thus defined:
Head=S plus 1D plus 0.5 to 1D, or
Head=S plus 1.5 to 2D (1)
where D is the loop arm diameter. Now S may preferably vary from 2 to 25
loop arm inside diameters, so that said Head may range from 2 plus 1.5 to
2D to 25 plus 1.5 to 2D, or from a minimum of 3.5 diameters to a maximum
of 27 diameters. Since D is also equal to the Perkins tube diameter PD, a
comparison may be formulated; if S equals 2 diameters, and assuming a
neglible wall thickness, the ratio of the loop liquid head to the Perkins
tube liquid head is
##EQU1##
or 3.5 to 4 times greater; and if S equals 25 diameters, it is
##EQU2##
or 26.5 to 27 times greater.
The superiority in liquid driving heads is even more marked when it is
observed that in the loop the cocurrent vapor flow reinforces the liquid
head, rather than lessen it as in FIG. 2, so that the total loop driving
head is the liquid head plus the reinforcing vapor flow. Higher driving
head, of course, means higher heat transport and less sensitivity to
unfavorable tilt.
As indicated, the loop arm spacing, S, is preferably 2 to 25 arm diameters;
more particularly, it may range from about 2.4 to 4 or 5 arm diameters. On
this basis, and using equation (1) above, the liquid driving head may vary
from 3.9 to 7 arm diameters, or 3.9 to 7 times greater than that of said
Perkins tube. Whether the broader or narrower range of arm spacing is
considered, it is clear that the liquid head is more than 3 times greater
than the Perkins. If a lower driving head is tolerable, S may also equal
one arm diameter, the liquid head then being 2.5 to 3 diameters or 2.5 to
3 times greater than that of the Perkins tube, which is still a sizable
improvement.
To obtain the advantages of the invention, the upper arm 12 is required to
be above the lower arm 13, as described, and preferably directly above it,
although a measure of variation is possible which may be described by
reference to FIG. 6, showing lower and upper arms 46, 47 of a loop 48.
Lower arm 46 has a horizontal line X'X intersecting its longitudinal
center line at point O. Only the end of said center line is visible at
point O. A horizontal plane passing through line X'X and the arm 46 center
line would have both lines lying in it. Now visualize a plane passing
through the line OA which joins the center lines of arms 46, 47, and
notice that this plane makes an angle AOX with the said horizontal plane.
The value of angle AOX is 60.degree., which represents a preferred angular
relationship, and the latter may range up to 120.degree., this being the
angle BOX formed when the upper arm 47 is to the left of arm 46 in the
position shown at 49. Accordingly, a preferred angular relationship of the
arms relatively to the horizontal is such that a plane passing through the
longitudinal center lines of the arms makes an angle of 60.degree. to
120.degree. with a horizontal plane passing through the longitudinal
center line of the lower arm. A wider useful range is 15.degree. to
165.degree., involving the positions 50, 51 of the upper arm and the
angles COX and DOX. A still more preferred disposition is that described,
where the upper arm is directly over the lower, i.e., 90.degree.,
corresponding to angle YOX.
It may be apparent in FIG. 6 that the lines X'X and Y'Y comprise the X and
Y axes of a set of plane or rectangular coordinates wherein O is the
origin or point of intersection of the axes. Also, as indicated earlier,
the spacing S between loop arms may preferably range from 2 to 25, more
preferably 2.4 to 4 or 5, and as broadly as 1 to 25, arm diameters. These
two quantities, angular disposition of the arms and the spacing
therebetween, serve to define the position of the arms relatively to each
other and relatively to the horizontal. If the center lines of both arms
of a loop lay in the horizontal plane passing through the line X'X, the
advantages of the invention are not obtainable; one would simply have the
Perkins tube of FIG. 2, in closed loop form, with the disadvantages
described herein.
Each loop may be conveniently formed by taking two straight thoroughly
cleaned tubes of the desired dimensions, then belling or flaring both ends
of each, then taking a cleaned return bend, preferably die formed, and
inserting its ends into the adjacent flared ends of the two tubes, taking
a second return bend and similarly engaging it with the opposite flared
ends of the tubes, and brazing the return bends to the tubes, as with
silver solder to get a joint of high melting point, to form a loop with
vacuum-tight joints. However, before attaching one of the bends, a fill
tube is desirably first installed on it, note FIG. 7, by drilling an
opening 54 in the outer wall 55 of the bend 56, and over such opening a
fill tube 57 is attached, so that the bore 59 is in registration with
opening 54. Conveniently, the attachment is by means of brazing or a weld
58. A valve (not shown but illustrated in U.S. Pat. No. 4,050,509) is
attached to the fill tube and its outboard side connected to a high
capacity vacuum pump. With the valve open, the loop is evacuated of all
air and non-condensible gases and until a hard vacuum in the range of
0.00005 to 0.00001 mm. mercury is obtained, after which the valve is
closed, and the pump is disconnected. The loop is then charged with
working agent, using conventional means. The fill tube is pinched shut, as
at 59a, the valve is removed, and the fill tube end is welded at 60 to
permanently close off the same. Each loop is thus hermetically sealed to
prevent agent from leaking out and air from leaking in.
Suitable working agents are those which exhibit a reasonable vapor pressure
of 5 or 10 to 200 psia at room temperature; are reasonably safe to handle;
are compatible with the loop material; are capable of being alternately
vaporized and condensed in the loop under the conditions of use; and are
reasonably priced. They may include the following, alone or in combination
with one or more others:
"Freon" 13B1 CBrF.sub.3
"Freon" 502 CHClF.sub.2 /CClF.sub.2 CF.sub.3
"Freon" 22 CHClF.sub.2
"Freon" 115 CClF.sub.2 CF.sub.3
"Freon" 500 CCl.sub.2 F.sub.2 /CH.sub.3 CHF.sub.2
"Freon" 12 CCl.sub.2 F.sub.2
"Freon" C-318 C.sub.4 F.sub.8 (cyclic)
"Freon" 114 CClF.sub.2 CClF.sub.2
"Freon" 21 CHCl.sub.2 F
Also such fluorinated compounds as 1,1-difluoroethane,
1,1,1-chlorodifluoroethane, and hexafluoroacetone; hydrocarbons like
propane and butane; and compounds such as ammonia, acetone, methyl
chloride, ethyl chloride, methyl formate, ethylamine, and sulfur dioxide.
Careful attention should be paid to the chemical compatibility of the
agent with the loop material as even low rates of chemical reaction may
produce non-condensible gases which, over the lifetime of the loops, may
accumulate to such as extent as to render the condenser inoperative.
Ammonia has been found to have good long term compatibility with carbon
steel and to suitably meet the other described qualifications. "Freon" 22,
b.pt. -41.36.degree. F., is a preferred agent for room temperature
applications. Preferred amounts range from 37.5 to 42% of the loop volume,
although amounts of 34 to 51.3% of loop volume may be used. An amount as
low as 26.8% of loop volume is useful provided some reduction in heat
transfer performance can be tolerated. These volume amounts of agents are
measured at the approximate temperature of operation of the loops. Earlier
boiling of agent may be promoted by addition of a small amount of a lower
boiling agent, say 1 to 13% by weight of the higher boiling agent. For
example, addition of 14.9 gms. (3% by wt.) of Freon-13B1, b.pt.
-71.95.degree. F., to 482 gms. of Freon-22 promotes boiling at a lower
temperature, and thus favors heat transfer by boiling over heat transfer
by conduction, the former being more efficient. In turn, evaporator heat
transfer coefficients are improved. In selecting a lower boiling material
to be added to the working agent, care is required to avoid forming a
maximum boiling azeotropic mixture, although minimum boiling azeotropes
may be useful if they meet the conditions described.
Tests of the performance of a loop are described below but may be
summarized briefly to the effect that the loop did not suffer burnout at a
high adverse tilt of 0.88 inch whereas in a comparable run a Perkins tube
showed burnout at adverse tilts of only 0.09 and 0.22 inch. Burnout, as
described in Example 2, is the point at which the tube will not function.
A loop as long as 20 feet showed no burnout at an even greater adverse
tilt of 0.97 inch. In Example 3 a 20-ft. loop operated at a power more
than two times greater than that of a comparable Perkins tube, and in a
separate test reached a power level of 2000 watts; in fact, only the
limitations of the electric heaters prevented a greater level from being
attained. In the same run, superheating of the vapor was demonstrated,
which is not thought to be possible in a Perkins tube; and also in the
same run, superior heat transfer coefficients were shown in a horizontal
orientation over an adverse tilt.
In the manufacture of the heat exchanger unit the fin plates 20, 21 are
formed full size and apertured to receive the loop arms. If desired,
shallow channels may be pressed into the plate between loops to aid
air-fin contact. The arms of the loops are desirably chosen from small
diameter tubes, 5/8 inch up to 1 inch or more, o.d., for which U.S.
industry is tooled and which are standard in industrial practice, although
larger diameter tubes are useful. The tubes are inserted through the
apertures of the fins, end panels, and barrier, then they are expanded to
produce an intimate contact therewith, after which the end bends (one of
each loop pair having a previously attached fill tube) are attached as by
silver soldering; and evacuation, charging, and sealing are carried out.
Very large heat exchangers may be built using loops of lengths greater
than 12 feet, measured between end bends. Levelling of such long units is
feasible to within 1 inch, but becomes more difficult as the lengths
approach 20 feet. But as shown in Example 4, at adverse tilts of 1 and 2
inches the unit does not become inoperable but rather undergoes some
reduction in efficiency; in fact, at an adverse tilt of 1 inch the
reported efficiences are still respectable at 55 to 57%, and could be
increased by operating at higher power. Other means for increasing the
efficiency include the above described step of using two agents to obtain
earlier boiling; and also conventional methods of improving heat transfer.
The use of the loops permits the unit to be bolted down and rigidly
attached to the air ducts, resulting in a bidirectional and completely
passive unit free from additional structures that detract from
reliability. And while the invention is specially addressed to units of
long length, owing to the problem of adverse tilts associated with them,
it is not limited in this way and may usefully be applied to units of
shorter length, say below 8 ft., or 6 ft., with which it is feasible to
employ loops made from tubes of smaller diameter, say from 1/8 of 1/4 to
5/8 inch, o.d.; further, units are contemplated measuring only about 12
inches in length by 6 inches in height having about 40 loops, suitably
arranged in rows, and whose arms are 1/8 or even 1/16 inch in diameter,
o.d. A further advantage is the freedom of the individual loop from
instabilities, as in situations where one loop may be carrying somewhat
more power than neighbor loops.
The invention is applicable wherever it is desirable to recover thermal
energy exhausted with air from HVAC (heating, ventilating, and cooling)
systems and industrial process equipment, and especially where the mass
flow is large or where significant temperature differences exist betwen
the exhaust product and incoming air. Example include year-round
ventilation systems for office buildings, shops, hospitals, laboratories,
restaurants, churches, schools, hotels, libraries, apartments, courts,
theaters, auditoriums, and the like, and buildings that house processes
like baking ovens, dryers, paint sprayers, etc. Also vehicles like the
passenger or other compartments of trucks, cars, military vehicles, and
including ships and aircraft.
Tests of the loop-containing heat exchanger were made, described in Example
4, and as will be recognized, a meaningful measure of its performance is
heat exchange efficiency. Briefly, it was found that efficiency (as
defined in said example) increased with increasing power transported; that
efficiency was good at an adverse tilt of 1 inch and, as expected, better
at a favorable tilt; and that as between a horizontal orientation and an
adverse tilt of 1 inch, efficiency was substantially insensitive to such
tilt, a result that would be increasingly apparent at smaller adverse
tilts. In example 5 a high efficiency was reached, ascribed to an
unbalanced air flow.
The invention may be illustrated by the following examples.
EXAMPLE 1
The flow of agent in a single loop was studied in the apparatus shown in
FIG. 8, where part of the lower arm 62 of an 8-ft. loop 63 of 5/8-inch
copper tubing was replaced by a transparent glass tube 64 about 32 inches
long around which was wound an electrical resistance wire 65 to supply
heat. Ends of the wire were connected to a power source not shown, and
included in the circuit were means for measuring the power. The loop and
first been thoroughly cleaned, evacuated of all air and non-condensible
gases, and charged with 600 gms. Freon-11, representing 56% of the
internal loop volume. The heated side 66 of the loop comprised the
evaporator, while the other side 67, comprising the condenser, was
disposed in a water bath 68 containing tap water 61 at about 55.degree. F.
A thermocouple was disposed at the point X in the upper arm 69 of the
evaporator. Leakproff hermetic connections between ends of the glass and
copper tubes were made conveniently using a nut having internal threads on
one side for engaging corresponding threads on the copper tube, and having
on the other side a rubber O-ring for engaging the glass tube.
With the loop in a horizontal orientation, as determined by use of a
transit, the agent was observed to completely fill the glass tube before
application of heat, and it was considered that the agent filled both end
bends 70, 71 to levels indicated by broken lines 72, 73. Heat was applied
corresponding to 10 watts of power, but no change in the agent was
observed in the glass tube. At 15 watts, the liquid was seen to suddenly
decrease in volume until it filled only 2/3 of the glass tube, and
intermittent slugs or bodies of liquid from the condenser were seen to
travel slowly in the direction of arrow A to the evaporator. These slugs
were apparently reflected back on meeting the end bend 70, as indicated by
an intermittently rising level. As power was gradually increased to 200
watts, the slugs moved faster but were still appearing intermittently;
between such appearances the liquid was quiescent and, at higher powers,
its level decreased to about 1/3 of the tube diameter, while the slugs
assumed an annular form.
The loop orientation was changed by dropping the evaporator side 1 inch
below horizontal, and the foregoing applications of power and observations
repeated. The direction of flow remained the same, and other observations
were generally unchanged. Then the evaporator was raised 1 inch above
horizontal, and the process repeated, but essentially the same results
were observed although the level of liquid tended to be greater in the
condenser.
The evaporator was next raised 2 inches above horizontal, and at 100 watts
the liquid level between slugs filled the glass tube completely adjacent
the condenser (point B), but only to about 10 to 20% of the tube diameter
adjacent the end bend 70 (point C), indicating considerable vaporization
to be occurring. At 200 watts the slug velocity was estimated to be 1 to 2
ft./sec. and slugs were moving from condenser to evaporator.
As it appeared from the last test that there was considerable vaporization
in the evaporator, and that the condenser was filling to capacity, the
amount of agent in the loop was reduced so that it represented only 34% of
the loop volume at 70.degree. F. At a 2-inch elevation of evaporator above
horizontal, the agent level was quite reduced, so the elevation was
dropped 1 inch. Power was applied as before. Typical slugs were observed,
except that at high powers, such as 200 watts, they were less fully
formed, owing probably to the reduced amount of agent; however, their
velocities at high powers were estimated to be about 10 ft./sec., and they
had some resemblance to an annulus, indicating annular flow. The loop was
then brought to a horizontal position and the testing repeated. At very
low powers, slugs several inches in length tended to form and to move
slowly (less than 1 ft./sec.) in the direction of arrow A; at high powers
the slugging ceased, the glass tube appeared to be about 40% full, and the
loop operated smoothly with agent flow in the direction of arrow A.
In all the foregoing tests, and at varying power levels and different
orientations, it was obvious that the agent flow was in one direction,
that of arrow A. Where vaporization took place, the vapors always flowed
in the direction of arrow D, they were condensed in the condenser, the
condensate returned to the lower arm 67 via return bend 71 and then passed
to the evaporator. Both liquid and vaporous agent were thus flowing in the
same direction, i.e., cocurrently.
EXAMPLE 2
This example compares the behavior, in respect of burnout, of a prior art
Perkins tube and a present loop, both 8 ft. long and of copper and both at
a negative or adverse elevation, i.e., evaporator above the condenser. The
Perkins tube was simply a straight length of tube, while the loop had its
arms 1.5 inches apart, as measured between the longitudinal center lines
of the two arms, or 0.94 inch in terms of the distance S of FIG. 1,
neglecting the tube wall thickness. Both tube and loop had been evacuated
of air and non-condensibles, charged with agent, and sealed, under
comparable conditions. In both cases heat was applied to the evaporator
under comparable conditions. Vapor temperatures were determined by
measuring vapor pressure with a precision pressure gage and then using
Freon tables (DuPont) to convert pressure to temperature.
Additionally, a test was made of a 20-ft. loop, having its arms spaced
apart 2.06 inches between center lines (S of FIG. 1 equal to 1.5 inches),
at an adverse elevation. The results with both loops demonstrate their
ability to transport heat at significant adverse orientations and over
long lengths. The data follows:
______________________________________
Perkins
Tube Loop
Run No. 1 2 3 4
______________________________________
Length, ft. 8 8 8 20
o.d., inch 0.625 0.625 0.625 0.625
i.d., inch 0.555 0.555 0.555 0.555
Adverse elevation,
0.22 0.09 0.88 0.97
inch
Working agent
Freon-12 Freon-12 Freon-12
Freon-12
% of tube or
48.6 48.6 49.4 42.9
loop volume
Vapor temp., .degree.F.
77 137 103 58.4
Power level, watts
334 473 600 590
Burnout Burnout No No
Burnout
Burnout
______________________________________
Comparing Runs 1 and 2 with 3, it will be seen that although the loop was
disposed at a more severe adverse orientation than the Perkins tube, being
4 and 5.5 times greater, no burnout took place, whereas burnout occurred
in the Perkins tube and at considerably lower power levels. Burnout is the
point at which the tube will not function, i.e., its capacity for
transporting heat is so drastically reduced that it is inoperative owing
to the fact that all the agent has flowed out of the evaporator into the
condenser; it is detectable from the fact that the evaporator, or any part
of the same, suddenly increases in temperature to a value substantially
equal to that of the heat source; below burnout temperature the evaporator
temperature is always related linearly to the amount of power carried by
the tube, and above burnout temperature the evaporator temperature is not
related to the amount of power transported. Burnout can be determined by
gradually increasing the applied power, measuring the evaporator wall
temperature and the vapor temperature at each power level, and then
plotting the power versus the difference of the two temperatures to obtain
a curve which, at burnout point, shows an asymptotic increase of the
temperature difference. Burnout point can also be determined by keeping
the power constant and changing the adverse elevation until the
temperature of the evaporator wall increases asymptotically. Although
different agents were used in the foregoing runs, this was not a
significant factor influencing the results because both Freon-12 and
Freon-22 have essentially the same properties at the test temperatures;
they are used interchangeably in low temperature air-to-air heat
exchangers. Also, the difference in agent volumes was not a factor; in
fact, the volume used in the Perkins tube was near optimum for that
device. Run No. 4, together with No. 3, illustrate the insensitivity of
the loops to significant adverse elevations, elevations which are greater
than those unavoidably encountered in commercial manufacturing,
installation, and maintenance practice. These runs further demonstrate the
potential of the loops for use in long heat exchangers.
EXAMPLE 3
This example describes power transport tests on a loop of long length (20
ft.) and presents some date on maximum power and heat transfer
coefficients. A 20-ft. loop (measured from the end of one end bend to the
other) of 5/8-inch o.d. copper tubing, with arms spaced apart 1.5 inches
between center lines (S of FIG. 1 equal to 0.94 inch), after evacuation
and charging with 985 gms. (814.4 cc. or 42.38% of loop volume at
70.degree. F.) Freon-22, was supported in the apparatus shown in FIG. 9
comprising an insulated tank 75, in which was disposed a 10-ft. portion 76
of the loop 77 as the evaporator, and a container 78, in which was
disposed an 8-ft. loop portion as the condenser 79. Tank 75 contained a
50-50 mixture 80 of water and methanol which was heated to vaporization by
electric heaters 81 placed outside and just below the tank, while
container 78 held flowing tap water 74 at about 50.degree. F. to condense
the Freon vapors within condenser 79. A separating space 82 of two feet
length, to which heat was not positively added or subtracted, was kept
between evaporator and condenser to prevent heat flow from one to the
other and to permit more accurate temperature measurements of the outside
loop walls by means of thermocouples located at the points marked with an
x. A pressure gage 83 provided vapor pressure readings in the end bend 84,
and by using conventional Freon-22 (DuPont) pressure-temperature tables,
the corresponding vapor temperatures could be obtained. An absolutely
perfect horizontal orientation of the loop over its length of 20 ft. could
not be obtained, but by means of a calibrated surveyor's transit a
horizontal position was secured accurate to 1/8 inch.
Data for a loop power run follows, and for contrast there is presented
along side comparable data thought to be typical of operation in a 20-ft.
Perkins tube made of 5/8-inch copper tubing.
______________________________________
Loop Perkins Tube
______________________________________
Length, ft. 20 20
o.d., inch 0.625 0.625
i.d., inch 0.555 0.555
Orientation Horizontal Horizontal
.+-. 1/8 inch
Agent Freon-22 Freon-22
% of tube or 48.38 40
loop volume
Vapor temperature,
69 50
.degree.F.
Power, watts 955.4 211
Maximum power, 2000 211
watts
h.sub.e, BTU/hr-ft.sup.2 -.degree.F.
204 --
h.sub.c, BTU/hr-ft.sup.2 -.degree.F.
1037 --
Superheat in vapor,
4.4 --
.degree.F.
______________________________________
Commenting on the above data, it is evident that higher power (measured
from current and voltage readings in the electric heater circuit and using
the equation, P equals EI) was obtained with the loop, 955.4 watts with no
burnout as against 211 watts with burnout for the Perkins tube; and even
if the latter value is doubled to take into account the use of two arms in
the loop, the comparison for the loop is quite favorable, 955.4 watts vs.
422 watts. The lower power in the Perkins tube is attributed, at least | | |
