 |
Description  |
|
|
REFERENCE TO RELATED APPLICATIONS
This application is related to the following commonly-assigned applications
filed concurrently herewith:
______________________________________
TITLE SERIAL NO.
______________________________________
APPARATUS FOR MOUNTING SURFACE
08/245,497
MOUNT DEVICES TO A CIRCUIT BOARD
METHOD FOR SURFACE MOUNTING
08/245,260
SURFACE MOUNT DEVICES TO A
CIRCUIT BOARD
APPARATUS FOR COOLING MULTIPLE
08/245,258
PRINTED CIRCUIT BOARD MOUNTED
ELECTRICAL COMPONENTS
HEAT CONDUCTIVE APPARATUS
08/245,490
FOR ENCAPSULATED ELECTRICAL
COMPONENTS
______________________________________
BACKGROUND OF THE INVENTION
The present invention is directed generally to attaching electrical
components to printed circuit boards. The invention relates more
particularly to removably affixing surface mount integrated circuit
packages with splayed leads to printed circuit boards, wire boards, or
wire assemblies.
An integrated circuit combines numerous active and passive electrical
circuitry elements on a single device called a die or chip. Commonly, the
electronics industry, including the computer, communications, and consumer
electronics sub-industries, interconnect integrated circuits by attaching
them to printed circuit boards. Because chips are inherently small and
fragile devices, they are usually imbedded in a substrate called a chip
carrier or package before they are attached to the printed circuit board.
Protruding from the package are a number of electrical conducting leads.
The manner in which the leads protrude vary; they may extend through the
bottom of the package (e.g. pin or pad grid arrays), they may be arrayed
along two edges of the package (e.g. dual in-line pins), or they may fan
out from the edges of the package (e.g. gull wing and J pins).
The wiring on printed circuit boards comprises thin metallic strips
embedded in an insulating material. These strips interconnect leads
protruding from one integrated circuit package to leads protruding from
other circuit packages mounted on the same board. To make all the required
connections, the boards typically have several layers of interconnect
wiring. The wiring determines the placement of the integrated circuit
packages on the board and routes electrical signals among the integrated
circuits. Leads may connect to the wiring in a variety of ways. One method
involves drilling holes in the board and through the wiring at appropriate
locations, inserting leads through the holes, and making mechanical and
electrical attachments among the leads, the wiring, and the holes.
Another, increasing popular technique, is called surface mount technology.
This method involves arranging contact pads on the surface of the printed
circuit board. The pads provide paths for electrical signals from leads to
the appropriate embedded wires. Leads are placed on top of the pads and
mechanically and electrically attached.
There are a number of methods of mechanically and electrically attaching
integrated circuit package leads to printed circuit boards. The most
commonly used techniques in the electronics industry require lead based
solder. There are many environmental and economic disadvantages in using
solder:
Lead is a well known hazardous material linked to numerous serious human
ailments. While there is no evidence that lead in electrical solder has an
impact on worker health or the environment, both the Environmental
Protection Agency and members of Congress have expressed concern. Industry
research in this area is ongoing.
Depositing lead onto printed circuit boards often requires fluxes which
must be cleansed. The preferred method of cleansing involves freon which
is known to persist for long periods in the atmosphere and is known to
deplete ozone. Other methods of cleansing boards involve using water which
creates problems of treating waste water to eliminate pollutants.
The heat required to deposit solder on boards may damage the components
being attached or the board itself.
Repairing a board or component requires reheating which may cause further
damage and require scrapping an expensive assembly.
Equipment to mass produce soldered boards (e.g. wave solder and flux
cleansing machines) is expensive and cumbersome.
Metal solders frequently short (or bridge) leads to one another, adversely
alter the electrical characteristics of integrated circuits (e.g.
increasing capacitance between leads), limit how closely leads may be
spaced (thus limiting the density of integrated circuits on a board), and
result in defective connections (e.g. cold solder joints) which may be
difficult or impossible to detect before a board is placed in service.
Soldered bonds between boards and components may be broken when the board
is dropped, flexed, or otherwise vibrated.
Many soldered assemblies require gold connections instead of less expensive
metals with superior or nearly equivalent electrical properties (i.e.
silver, copper, and aluminum), because of gold's thermal expansion and
anti-corrosive properties.
Surface mount technologies generally involve applying a 3- to 4-mil coating
of solder paste onto contact pads on the surface of the board, placing
package leads onto the pads, and melting (reflowing) the solder. This is
sufficient mechanically to hold the package in place while making the
proper electrical connection. Surface mounting has a number of advantages
over older techniques; it saves board space by permitting dies to be
mounted in small packages with closely spaced leads, it reduces the number
of levels of embedded wires in a board, and it allows components to be
mounted on both sides of a board. However, surface mount technology not
only entails the known problem of using solder, but also raises the
additional problem of properly aligning the leads onto the contact pads.
Proper placement of surface mount components generally requires
specialized computer controlled equipment.
Because of the advantages of surface mount technology, much research has
been directed at the soldering and alignment problems. Two alternatives to
soldering are the subject of most research; attaching packages to boards
with adhesive (e.g. TAB technology) or applying sufficient pressure to
packages to make a mechanical and electrical connection. Proper placement
of components for TAB and pressure mount techniques has proved to be as
difficult a problem as it is for solder techniques.
Most pressure techniques involve compressing the package itself to the
board. A compressible pad with embedded electrically conductive material,
known as a Z-axis connector, is frequently placed between the package and
the board. Usually only leadless packages with contact pads on the bottom
surface may be used; pressure on the top of a package would stress the
point at which splayed leads, such as gull wings, attach to the package.
In any event, applying pressure to the top of any package sufficient to
maintain a proper electrical connection between the leads and the board
contact pads, may damage the package.
OBJECTS AND SUMMARY OF THE INVENTION
Accordingly, it is an object of the present invention to provide a less
complex method and apparatus for cooling a plurality of electrical
components mounted to a printed circuit board.
Broadly, the invention comprises a base plate mounted on a circuit board
proximate to and in registered relation to a number of component sites
formed on the circuit board and defined, for each connection site, by a
number of connection pads at which an electrical component is to
installed. A chuck assembly, configured to receive and carry the
electronic components, is affixed to the base plate in a manner that
aligns and registers each of the electrical components to the chuck
assembly to the base plate, and thereby to the component sites. The manner
of alignment locates electrical leads (connections) of each of the
electrical components in electrical contact relation with corresponding
ones of pads of corresponding component sites.
Where the electrical components are not sealed within the chuck assemblies,
a fan blowing air across the circuit board may suffice. A more advanced
approach is where, again, the electrical components are not sealed within
the chuck assemblies, but the circuit board, or a portion thereof, is
sealed within a chamber formed by the circuit board, the base plate, and
the alignment plate. In this implementation, the base plate is configured
to form a side wall that, with the circuit board and the alignment plate,
forms a cooling chamber that enclosed the plurality of electrical
components. Ingress and egress, for example through aperatures formed in
the base plate, is provided to circulate either the cooled or warmed gas
or fluid actively circulated through the chamber, around and across the
chuck assemblies and over the circuit board.
Another method of controlling the temperature of components on the circuit
board involves an alignment plate with cavities therein. Coolant would be
circulated through the cavities. In addition, heat sink devices could
extend from the cavities into the chucks as described above. This
configuration would also permit cooling components cryogenically. That is,
for components which incorporate superconducting material, supercooled gas
or fluid may be circulated through the cavities to keep the components
below their transition temperatures.
The above, and other objects, features and advantages of the present
invention will become apparent from the following description read in
conjunction with the accompanying drawings, in which like reference
numerals designate the same elements.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is an exploded perspective view of an alignment device constructed
according to the present invention;
FIG. 2 is an exploded cross-sectional view of the relative position of
assembly components in the production process;
FIG. 3 is a cross-sectional view of assembled device of FIG. 1;
FIG. 4 is an upside-down perspective view of a number of assembled devices,
including their relative positions as mounted on an alignment plate, and
the relative position of the alignment plate to a base plate and printed
circuit board;
FIG. 5 illustrates the assembled alignment device of FIG. 3, modified to
include a heat sync device; and
FIG. 6 illustrates multiple components mounted to a circuit board, each
having the heat sync device shown in FIG. 5, and all enclosed in a
manifold.
DESCRIPTION OF THE PREFERRED EMBODIMENT
The present invention may be used to mount a variety of electrical or
electronic components to circuit boards (or comparable method), but has
particular application to mounting surface mount technology (SMT) devices
of the type having gull wing or "J" connector leads, such as illustrated
in the figures. (Unless otherwise noted, the term "lead" is used herein to
refer to the portion of an electrical or electronic component through
which an electrical connection is made between the component and a circuit
board or other component. It will be obvious to those skilled in this art
the examples of lead configurations include, in addition to the to the
gull wing or J leads illustrated in the Figures, dual inline pins, pin
grid arrays, and pad grid arrays.) As indicated above, certain prior art
mounting methods for circuit components rely upon pressure applied
directly to the device itself (or, more accurately the carrier which
houses the chip or electronic component of the device). This can tend to
impose stresses sufficient to fracture the device, leading to device
failure. The present invention operates to capture and hold only the leads
of the device.
Turning now to FIGS. 1-3, one embodiment of the mounting method of the
present invention, designated generally by the reference numeral 10 is
illustrated. As FIG. 1 shows, the mounting apparatus 10, comprising a base
plate 12 and chuck assembly 14 (shown here in exploded fashion), is
installed on a surface of a circuit board 20, proximate a SMT connection
site 22. The SMT connection site 22 of the circuit board 20 is
conventional in the sense that it is defined by a number of conductive
pads 24 that are oriented to receive and connect to corresponding ones of
printed leads 26 of an engineering sample SMT device 30 placed thereat.
The pads 24, in turn, connect to circuit traces 32 formed on the circuit
board 20 for communicating signals to other devices or components (not
shown) that may be also mounted on the circuit board 20.
The base plate 12 is formed with alignment holes 36 that align with
underlying apertures 36a (FIG. 2) formed in the circuit board 20 to
register the base plate 12 to the SMT connection site 22. A central
opening 40 is formed in the base plate 12 so that, when mounted to the
circuit board 20, access to the SMT connection site 22 is provided
therethrough. The central opening 40 of the base plate 12 has alignment
slots 42 formed in opposing wall portions thereof for receiving the
terminal ends 16 of an alignment plate 15 that forms a part of the chuck
assembly 14. The dimensions of the alignment slots 42, together with those
of the alignment plate 15 (or, at least the terminal ends 16 of the
alignment plate 15), are such that minimal, if any, movement (lateral or
longitudinal) is permitted the alignment plate 15 while so installed on
the base plate 42.
The chuck assembly 14 includes, in addition to the alignment plate 15, a
chuck 18, having a recess 19 formed in an undersurface thereof. The
configuration and dimensions of the recess 19 are such that it can snugly
receive and hold the sample device 30. As shown, the sample device 30 has
J or gull wing leads; however, as indicated above, the present invention
may be used with any type of SMT device lead. Here, the recess 19 formed
in the chuck 18 is sized to capture the SMT device by the leads; and the
bottom peripheral surface 21 of the chuck 18 captures the
lateral-extending terminal portions of the SMT device leads between the
peripheral surface 21 and the pads 24 of the circuit board 20. It will be
evident, however, that if the sample SMT device 30 uses other types of
connective leads (e.g., those that linearly extend laterally away from the
carrier itself, or those that use a pin array protrusion-type leads formed
on the underside of the sample SMT device 30), the recess 19 would be
configured to snugly receive the periphery of the chip carrier that forms
up the sample SMT device 30.
The chuck 18 is affixed to the alignment plate 15, forming the chuck
assembly 14. An alignment process locates the chuck 18 on the alignment
plate 15 so that, when the completed chuck assembly 14 is mounted to the
base plate 12 with the alignment slots 42 receiving the terminal ends 16
of the alignment plate 15, the recess 19 (and the sample AMT device 30 it
holds) will be substantially registered to the SMT connection site 22.
This registration also places the leads 26 of the SMT device 30 in
substantial alignment with the pads 24, to be held thereagainst in place
by the bottom peripheral surface 21 of the chuck 18 for electrical mating.
The alignment process establishes the location and position of the chuck
18 relative to the to the base plate 12 through the alignment plate 15.
The base plate 12, in turn, is registered to the SMT connection site 22 of
the circuit board 20 by alignment of the alignment holes 36 formed in the
base plate 12 with the alignment holes 36a formed in the circuit board 20.
Thereby, any SMT device carried by the chuck assembly 14 is registered to
the SMT connection site 22 for electrical connective engagement of the SMT
device leads with corresponding pads 24 when the chuck assembly 14 is
mounted to the base plate 12.
The alignment process used to register the chuck 18 to the base plate 12,
and thereby to the SMT connection site 22, is as follows. Referring to
FIGS. 2 and 3, a sample SMT device 30 is first conventionally mounted to
the circuit board 20 (e.g., by wave-soldering, adhesive, or any other
technique customarily used to affix SMT devices to circuit boards) with
the leads of the sample SMT device 30 in contact with the corresponding
pads 24 that form and define the SMT connection site 22. The base plate 12
is attached to the circuit board in proximate relation to the SMT
connection site 22 by alignment of attachment holes 36 with corresponding
attachment holes 36a formed in the circuit board 20 (and a 37 or other
attachment mechanism inserted therethrough). Alternatively, the base plate
12 may be adhesively attached to the circuit board 20, using other
techniques to maintain registration. It will soon be evident that in this
alignment process, no particular relative relation between the base plate
12 and the SMT connection site 22 is initially necessary other than that
site 22 be accessible to the chuck assembly 14 through the central opening
40 of the base plate 12. Also, the base plate 12 should be mounted to the
circuit board 20 to that a center line extending between the two alignment
slots 42 pass over (or at least proximate) the SMT connection site 22 for
reasons that will understood below. However, once registration is
established between the chuck 18 and the SMT connection site 22 (through
the alignment plate 15, base plate 12 and aligned holes 36, 36a) the
relative location and position of the base plate 12 with the SMT
connection site 22 must be maintained.
The chuck 15 is then inserted through the central opening of the base plate
40, fitted over the sample device 30 so that the device seats in the
recess 19. An adhesive is spread over the top surface 18a of the chuck 18,
and the alignment plate 15 is placed in alignment slots 42 of the base
plate 12. The dimensions of the alignment plate 15, the chuck 18, and the
alignment slots 42 are such that the underside of the alignment plate 15
contacts the top surface 18a of the chuck 18, allowing the adhesive to
bind the two together, forming the chuck assembly 14. Further, it can now
be seen why the placement of the base plate 12 should be such that a
center line extending between the alignment slots pass at least near the
SMT connection site 22.
At this point, the alignment process is complete. The chuck 18, now forming
a part of the chuck assembly 14, is now registered to the SMT connection
site 22.
The present invention may now be used in a manufacturing environment. Here,
production versions of the circuit board 20 are provided, identical to
that used to align and register the chuck 15 to the SMT connection site
22, including of course the SMT connection site 22 and alignment apertures
36a formed therein to align with the apertures 36 formed in the base plate
12. Further, these alignment apertures 36a formed in production versions
of the circuit board 20 should positioned relative to the SMT connection
site substantially identical to those formed on the circuit board 20 used
in the alignment process, so that when a base plate 12 is installed
thereon, and its apertures 36 aligned with those (36a) of the production
version of circuit board 20, the base plate 12 will be registered to the
SMT connection site 22 of the production circuit board 20 substantially
identical to the registration of the base plate 12 to the SMT connection
site 22 of circuit board 20 used in the alignment process.
Thus, as illustrated in FIG. 2 and 3, mounting the SMT device 30 on a
production version of the circuit board 20 (with an empty SMT connection
site 22) merely requires the base plate 12 to be attached by bolts 17
inserted through the attachment holes 36 and aligned attachment holes 36a
of the production version of the circuit board 20. So attached, the base
plate 12 is registered to the SMT connection site 22 of the production
version circuit board 20 in the same manner as it was in the alignment
process. An SMT device 30 is fitted to the chuck assembly 14 so that it
seats in, and is held by, the recess 19 of the chuck 15. The terminal ends
16 of the alignment plate 15 of the chuck assembly 14 are then inserted
into alignment slots 42 and attached to the base plate 12. For this
purpose attachment holes 109 (FIGS. 2 and 3) and attachment receiving
holes 109a are respectively formed in the chuck assembly and the base
plate to receive screws of bolts 112 (FIG. 3) to hold the chuck assembly
14 in place to the base plate 12. As FIGS. 2 and 3 illustrate, the
laterally extending terminal portions of the leads 26 emanating from the
SMT device 30 are captured and held between the bottom peripheral surface
21 of the chuck 18 and pads 24 of the circuit board 20.
The leads 26 of the SMT device 30 may make direct contact with the pads 24.
However, optionally, as shown in FIGS. 2 and 3, a Z-axis connector 110 may
be placed between leads 26 and circuit pads 24. Thereby, the SMT device
leads 26 are kept in mechanical and electrical connection by the
attachment of the chuck assembly 14 to the base plate 12, and the
attachment of the base plate 12 to the production version of the circuit
board 20. One advantage of using Z-axis connector 110 is that the chuck
assembly 14 containing the SMT device 30 may be mounted and demounted
repeatedly with minimum wear on circuit pads 24. If the Z-axis connector
110 is omitted, the mechanical and electrical connection may be still
maintained, but repeated removal and insertion of the SMT device can act
to abrade and score the pads 24 and/or SMT device leads.
Other connection options include using electrically conductive solder or
adhesive instead of pressure. The chuck assembly 14 and the base plate 12
would be employed as previously indicated to align and register the SMT
device 30 to its proper position relative to the SMT connection site 22 on
the circuit board 20. However, the solder or adhesive is spread between
the leads 26 and the circuit pads 24 before they are forced into
electrical contact. In the preferred embodiment, the circuit board 20 and
the SMT device 30 then are tested for electrical and mechanical faults. If
the combination fails any of the tests, the chuck assembly 14 containing
the SMT device 30 is demated from the base plate 12 and the problem is
corrected before the combination is mated again. Once the combination
passes the tests, the solder is melted or the adhesive is activated. The
chuck assembly 14 and the base plate 12 may then be removed from the
production circuit board 20 leaving the SMT device 30 mounted in its
proper location.
The foregoing discussion has taught a method and apparatus for mounting a
single SMT device to a circuit board. It should be evident to those
skilled in the art that the invention can be extended to mounting multiple
SMT devices. Thus, referring now to FIG. 4, a method and apparatus for
mounting multiple SMT devices is illustrated. FIG. 4 shows a circuit board
70, having a plurality of SMT connector sites (not shown) formed on a
planar surface 72 of the circuit board 70. Affixed to the circuit board
70, such as discussed above, with alignment holes and bolts, or by an
adhesive, or any other attachment technique, is a base plate 76 that
encircles the SMT connection sites (not shown) on the circuit board 70.
The base plate 76 has formed, in opposing wall portions 78 (only one of
which can be seen in FIG. 4) alignment elements in the form of notches 80.
A broad, planar alignment plate 84 is formed and configured with alignment
tabs 86 to mount to the base plate 76 so that the alignment tabs 86 engage
notches 80 to register the alignment plate 84 to the base plate 76.
The alignment phase is essentially the same as that described with respect
to mounting a single SMT device. First, a "stand-in" circuit board is
used, having SMT devices mounted at the SMT connection sites formed on the
circuit board. Chucks 90 are configured with recesses 91 to be fitted over
the mounted sample SMT devices, and the alignment plate 84 mounted to the
base plate 76 and affixed to the chucks 90, forming a chuck assembly 92.
The individual chucks 90 are now registered to the SMT connection sites
through the alignment plate 84, base plate 76, and its placement on the
circuit board 72.
As above, production models of the circuit board 72, identically structured
and configured to that used during the alignment process, including
placement of the SMT connection sites (not shown) and the installation of
the base plate 70, can now be produced. Similarly, production models of
the chuck assembly 92, with the chucks 90 oriented according to the
alignment process and affixed to the alignment plate 84 can be
constructed. Assembly merely requires installing in each of the chuck
assemblies 90 a production versions of the SMT devices 100 by inserting
them in the recesses 91 of the chucks 90. The chuck assembly 92 is then
installed on the base plate 76 with the tabs 86 received in the notches or
grooves 80, thereby registering the SMT devices 100 to the SMT connection
sites formed on the surface 72 of the circuit board 70 through (1)
orientation and placement of the chucks 90 on the alignment plate 84, (2)
registration of the alignment plate 84 to the circuit board 72 through (3)
the base plate 76.
Referring specifically to FIG. 3, an embodiment of the invention is shown
employing a Z-axis connector 110 on top of which sits SMT device 30. Leads
26 of gull wing configuration extend from the SMT device 30, and are
captured between inner edge of the chuck assembly 14 and the Z-access
connector 110. The chuck assembly does not touch the SMT device package,
but leaves a space 19 between the SMT device 30 package and the chuck
assembly 14 inner surface. This volume may now be sealed and made gas
tight. That is, a container for the SMT device 30 is formed by the chuck
assembly 14, leads 26, and the Z-axis connector 110.
One advantage of sealing the SMT device 30 is that it is then protected
from dust and caustic fluids and gases. Thus, the invention may be
employed in hostile environments. In addition, it is preferrable to fill
the space 19 with a substantially inert gas (e.g. helium, nitrogen, argon)
or fluid. It is also preferrable to pressurize the gas to test and insure
the seal and to prevent ingassing. Then a wide choice of materials such as
silver, copper, and aluminum may be employed for the leads 26 and the
circuit pads 24. These materials, especially silver, have excellant
electrical properties, but are infrequently employed because of their
tendency to corrode. If sealed in a substantially inert volume, the
corrosion problem is moot.
Many different types of SMT devices are available. Some of them, such as
those containing CPU chips, generate a significant amount of heat which
must be dissipated. Depending on the amount of heat generated, different
passive and active heat dissipation techniques and devices may be
employed.
One solution is to use the leads 26 and the chuck assembly 14 as heat sink
devices. Much of the heat the SMT device 30 generates flows to the leads
26. The chuck assembly 14 may draw off this heat if it is constructed of
material which is a good thermal conductor while substantially an
electrical insulator. A number of such materials are known including
ceramic substrate materials incorporating aluminum oxide, aluminum
nitride, beryllium oxide, silicon carbide, or boron nitride (BN).
The effectiveness of the above heat dissipating technique and device may be
enhanced in different ways. For example, cylinders or slugs of thermally
conductive metal, such as copper, may be embedded in the chuck assembly
14. Heat sink devices may be attached to the top of the chuck assembly 14
or may extend into the chuck assembly 14 body. If the volume containing
the SMT device 30 is sealed, space 19 may be filled with a relatively good
thermally conductive gas, such as hydrogen or helium, or fluid.
Alternatively, the chuck assembly 14 may contain openings permitting air
to flow around the chuck 18 and/or the SMT device 30 itself.
Another embodiment involves imbedding a screen or other material in the
chuck assembly 14 to intercept electromagnetic interference (EMI) either
originating from external sources or from the SMT device 30 itself.
One technique to dissipate heat is to make a central opening in the top of
the chuck assembly 14. By itself, this opening may allow sufficient air to
circulate to keep the SMT device 30 within its operating temperature
range. For additional passive cooling, as illustrated in FIG. 5, thermal
grease 168 (a heat conducting adhesive) may be placed on a top surface of
the SMT device 30 package. Then a heat sink device 150 may be placed on
the thermal grease 168.
In some applications, it may be desirable to combine sealing the SMT device
30 with placing a heat sink device 150 on top of the SMT device 30
package. However, it is important that for gull wing or J leads 26, that
the heat sink device does not apply pressure to the package; anything more
than a small downward force upon the package will deform the leads 26 and
stress the point at which they connect to the package of the SMT device
30.
The solution, as illustrated in FIG. 5, is to make an opening 160 just
large enough in the top of the chuck assembly 14 to accomodate a heat sink
device. In assembling the production device, thermal grease 168 is first
placed on top of the SMT device 30 package. Then the chuck assembly 14 and
the SMT device 30 are fitted together. A Z-axis connector 110 is placed
upon conductive pads 24; the chuck assembly 14 containing the SMT device
30 is then mounted as described above. The chuck assembly 14 is sealed to
the circuit board 20. Next, the heat sink device 150 is placed through the
opening 160 in the top of the chuck assembly 14 far enough to adhere to
the thermal grease 168 on top of the SMT device 30.
An inert gas or fluid may be introduced into the volume defined by the
chuck assembly 14, heat sink device 150, and circuit board 20. Finally,
any gap between the opening in the top of the chuck assembly and the heat
sink device is closed (with an O ring 164 for example) to complete the
seal of the volume containing the SMT device 30.
Given a sealed device as illustrated in FIG. 6, active cooling systems may
be employed. Such a system may be needed for devices which generate
prodigious amounts of heat, or, for example, to cryogenically cool SMT
devices 30 below threshold temperatures for electrical superconductivity.
The principle is to refrigerate cryogenical cooled gases, such as helium,
or fluids, such as liquid nitrogen or liquid helium, and pump them around
the heat sink devices 150. If the need for cooling is not as great, a
fluid such as water may be employed.
The sealed device serves as a bottom portion of a manifold. A top portion
(FIG. 6) completes the manifold; it is a cover 170 surro | | |
