Patent Application
20100006132
The invention is directed, in general, to thermoelectric modules.
Thermoelectric modules (TEMs) are a class of semiconductor-based devices that may be used to, e.g., heat or cool an object, or may be used to generate power when placed between a heat source and a heat sink. Generally, semiconductor pellets of alternating doping type are arranged in series electrically and in parallel thermally. As current flows through the pellets, one side of the TEM becomes colder, and the other warmer. Conversely, when placed in a thermal gradient, the TEM may drive a current through a load. A TEM may be used to cool or heat a device, or to maintain an operating temperature with the aid of a feedback control loop.
One embodiment is an apparatus including a first thermally conductive body having a plurality of fingers and a second thermally conductive body having a plurality of fingers. The first and second bodies are configured such that the fingers of the first body are interdigitated with the fingers of the second body. Each of a plurality of thermoelectric modules has a first major surface and an opposing second major surface. The first major surface is in thermal contact with one of the fingers of the first body, and the second major surface is in thermal contact with one of the fingers of the second body.
Another embodiment is a method that includes providing a first thermally conductive body having a plurality of fingers and a second thermally conductive body having a plurality of fingers. The first and second bodies are configured such that the fingers of the first body are interdigitated with the fingers of the second body. Each of a plurality of thermoelectric modules has a first major surface and an opposing second major surface. The first major surface is configured to be in thermal contact with the one of the fingers of the first body, and the second major surface is configured to be in thermal contact with one of the fingers of the second body.
Another embodiment is a system that includes a first thermally conductive body having a plurality of fingers and a second thermally conductive body having a plurality of fingers. The bodies are configured such that the fingers of the first body are interdigitated with the fingers of the second body. Each of a plurality of thermoelectric modules has a first major surface and an opposing second major surface. The first major surface is in thermal contact with one of the fingers of the first body and the second major surface being in thermal contact with one of the fingers of the second body. A controller is configured to control a rate of heat transport between the first body and the second body through the plurality of thermoelectric modules.
Various embodiments are understood from the following detailed description, when read with the accompanying figures. Various features may not be drawn to scale and may be arbitrarily increased or reduced in size for clarity of discussion. Various features in figures may be described as “vertical” or “horizontal” for convenience in referring to those features. Such descriptions do not limit the orientation of such features with respect to the natural horizon or gravity. Reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
A thermoelectric module (TEM) typically is assembled from a number of n-doped and p-doped pellets connected electrically in series and thermally in parallel. When an externally applied potential causes electrons and/or holes to flow through the pellets, a temperature difference is maintained across the pellets. The pellets are electrically connected to each other in series in alternating fashion such that one side of the TEM becomes cooler in response to current flow, and the other side becomes warmer. Thus the TEM may be used as a heat pump. Operation of the TEM is reversible in the sense that when a thermal gradient is imposed across the pellets of a TEM, a potential is produced that may be harnessed to produce work.
As current flows through the pellets, Joule heating (I2R) causes power to dissipate in the pellets. The dissipated power raises the temperature of the pellets, reducing pumping efficiency. The dissipated power also increases the heat to be removed by the TEM. Thus, in general, the efficiency of a TEM is greater for a lower pumping current than for a higher pumping current. However, the lower the pumping current, the lower the rate of heat transfer. This tradeoff typically leads to a design compromise particular to the requirements of a specific system design.
Efficiency of a TEM may be increased by increasing the area of the TEM so the current through each pellet is reduced, and/or thermally isolating, e.g., insulating, the hot and cold sides of the pellets to a greater degree. U.S. patent application Ser. No. 12/128,478 to Hodes, et al., incorporated herein by reference as if reproduced herein in its entirety, discloses an apparatus in which heat from a heat source is spread laterally using a thermally conductive substrate such as, e.g., a vapor chamber. The laterally conductive substrate allows the use of a larger TEM, which may reduce the heat flux per unit area (unit heat flux) through the TEM, thus allowing the TEM to operate with greater efficiency. However, this approach may require the use of greater area of a circuit assembly substrate (e.g., a printed circuit board) in a cooling or temperature control application, or in a waste heat recovery application, such as, e.g., from an automobile exhaust system. In such cases it may be prohibitively expensive and/or impractical to provide greater area to the TEM to achieve greater efficiency of operation. What is needed is an alternative method to reduce the current density through TEM pellets that consumes less area on a circuit assembly substrate.
The described embodiments benefit from the recognition that TEMs may be stacked, but operated thermally in parallel to decrease the unit heat flux through the TEMs while reducing the required circuit assembly substrate area. Thus, the effective area used to spread heat from, e.g., a heat-producing device, is increased by extending the TEMs vertically with respect to the substrate rather than horizontally. The vertical assembly utilizes a thermally conductive body to configure the TEMs in parallel thermally, as described further below.
Turning initially to
The heat source 110 may be any source of heat. In some embodiments, the heat source 110 is an electronic device configured to produce heat when energized, such as, e.g., a microprocessor, power amplifier, or high power laser. In other embodiments, the heat source may be a source of waste heat such as, e.g., a smoke stack or a catalytic converter of an automobile. The apparatus 100 may be configured to cool the heat source 110, to maintain a temperature thereof, or to recover power from the waste heat. If a temperature is maintained, a controller including, e.g., an active feedback loop, may be used. In still other embodiments, the heat source 110 may be a passive device, such as a sensor, e.g., that does not dissipate heat, but is maintained by the apparatus 100 at a desired operating temperature.
The thermally conductive body 130 includes fingers, e.g., fingers 135a, 135b, 135c (collectively referred to as fingers 135). The thermally conductive body 140 also includes fingers, e.g., fingers 145a, 145b, 145c (collectively fingers 145). While the thermally conductive bodies 130, 140 are each shown having three fingers, embodiments are contemplated having fewer or more fingers. The thermally conductive bodies 130, 140 may have the same number of fingers, as illustrated, but embodiments having unequal numbers of fingers are within the scope of this disclosure. The fingers 135, 145 are interdigitated, as described in greater detail below. A TEM 150a is located between the fingers 135a, 145a. Similarly, TEMs 150b, 150c, 150d, 150e are located between the remaining fingers 135, 145 as illustrated. (TEMs 150a, 150b, 150c, 150d, 150e are referred to collectively as TEMs 150.) Thus, the heat sinks 150 are stacked vertically with respect to the heat sink 120. A space 180 is not occupied by either a TEM or a thermally conductive body.
The combination of thermally conductive paths provided by the thermally conductive bodies, and thermal insulation provided by the TEMs results in a compact, space filling assembly. The thermally conductive paths may be chosen to conduct heat in a predetermined three-dimensional thermal circuit to provide a larger TEM surface area to pump heat, or recover waste heat, than would otherwise be possible in the footprint of the assembly. This configuration in effect provides, for the case of cooling the heat source 110, a single evaporator at the portion of the thermally conductive body 130 in thermal contact with the heat source 110, and multiple condensers at the portions of the thermally conductive body 140 that are in thermal contact with the TEMs 150. For the case that heat is transferred to the heat source 110, the situation is reversed-in effect the portions of the thermally conductive body 140 in thermal contact with the heat sink 120 act as a single evaporator, and the portions of the thermally conductive body 130 in thermal contact with the TEMs 150 act as multiple condensers.
Turning briefly to
Returning to
The thermally conductive bodies 130, 140 may be formed from any material generally accepted as thermally conductive, typically about 200 W/m-K or greater. The body may include a layer designed to increase thermal conductivity in one or more directions relative to the surface of the thermally conductive bodies 130, 140. Examples materials include, e.g., metals such as aluminum, copper, silver, and gold; composites such as Al/SiC; ceramics such as beryllium oxide (beryllia); and carbon-based thermal conductors such as diamond films and pyrolitic graphite. In some cases, the thermal conductivity is about 400 W/m-K or greater. In some embodiments, described further below, the thermally conductive bodies 130, 140 may include a vapor chamber or heat pipe, in which case the effective thermal conductivity may be about 5000 W/m-K or greater in at least some directions. The thermally conductive bodies 130, 140 may formed of the same or different materials, may have a same or different thermal conductivity characteristic, and may have fingers 135, 145 of a same or different geometry.
In the apparatus 100 and the apparatus 170, any TEM may optionally be replaced with a thermally insulating material if desired. For example, it may be desired to design a specific heat flux value through one or several TEMs 150. Moreover, insulation may optionally be placed in a void space, e.g., the space 180, to increase thermal isolation of the thermally conductive bodies 130, 140. Examples of insulating materials include expanded polystyrene and aerogel.
Turning to
The operation of a vapor chamber is described, e.g., in the '478 application, and is summarized here using the thermally conductive body 330 as an example for the case the TEMs 350 are configured to pump heat from the thermally conductive body 330 to the thermally conductive body 340. The vapor chamber 360 includes a wick 362 lining one or more interior surfaces of a chamber that is otherwise hollow, with the exception of any needed structural supports. The wick 362 is wetted with a working fluid such as alcohol or water. When an exterior surface of the chamber is in contact with a heat source, the working fluid in the vicinity of the contact area evaporates into the vapor chamber 360 and is transported away from the heated area. The phase change carries the heat of evaporation away from the heated area. The vapor may then condense on the wick in a cooler region of the chamber. The phase change releases the heat of condensation in the cooler region. In this manner, the effective lateral thermal conductivity (e.g., parallel to the lined interior surface) may be 10×-100× the thermal conductivity of a solid metal thermal conductor. Thermal conductivity ranging from 5,000-20,000 W/m-K is possible.
Heat from the heat source 110 flows to the thermally conductive body 330. The working fluid in the wick 362 proximate the heat source 110 vaporizes and enters the vapor chamber 360 open volume of the vapor chamber 360. The vapor diffuses through the vapor chamber 360 to the portions thereof proximate the TEMs 350a, 350b, 350c. Because the surface of each TEM 350a, 350b, 350c in contact with the thermally conductive body 330 is cold, relative to the surface in contact with the thermally conductive body 340, the vapor condenses on the wick 362 proximate the TEMs 350a, 350b, 350c. The heat of condensation is pumped by the TEMs 350a, 350b, 350c to the thermally conductive body 340.
In the thermally conductive body 340, the heat transported by the TEMs 350a, 350b, 350c causes the working fluid in a wick 370 to vaporize. The vapor diffuses through the chamber 365 and condenses on the wick 370 proximate the heat sink 120. The heat sink 120 may be cooled by, e.g., air or a liquid to remove the heat of condensation from the thermally conductive body 330. The heat sink 120 may have a larger footprint than the TEMs 350. The thermally conductive body 340 and the heat sink 120 may be configured to provide as large an area to dissipate heat as desired, as limited by available area in the system design.
Turning to
The size of a TEM may is limited to a maximum, such as, e.g., about 5 cm on a side, due to differential thermal expansion of the warm and cold sides. In the example embodiments described herein, when a design requires a TEM with a dimension greater than the allowable maximum for the type of TEM employed, a number TEMs with dimensions of smaller than the applicable maximum may be used. In such embodiments, each of the multiple TEMs is in thermal contact with, e.g., both the thermally conductive body 330 and the thermally conductive body 340.
The TEM 350b is configured such that a current I produces a thermal gradient 382 that, e.g., cools an upper major surface 450 of the TEM 350b and warms an opposing lower major surface 460. The major surfaces are the surfaces through which heat is transported when the TEM 350b is operating. The n-pellets 410 and p-pellets 420 have relatively low thermal conductivity, e.g., 10-20 W/m-k. Thus, the TEM 350b acts as a thermal insulator through which the heat flux may be modulated by the current I.
In an example embodiment, one or both thermally conductive bodies 330, 340 are mated to the TEM 350b to form a single integrated structure. In other words, the thermally conductive body 330, e.g., acts as the substrate for the pellets 410, 420 and the conductors 430. In such an embodiment, electrical isolation of the conductors 430 from the vapor chamber may be provided by, e.g., a thin polymer layer. Additional details on integration of a vapor chamber and a TEM are provided in U.S. patent application Ser. No. 12/128,478 to Hodes, et al., incorporated herein by reference as if reproduced herein in its entirety.
In the apparatus 300, e.g., the upper major surface 450 is placed in thermal contact with a lower surface 470 of the finger 432. Similarly, the lower major surface 460 is placed in thermal contact with an upper surface 480 of the finger 434. Optionally, a thermal conduction aid such as, e.g., thermal grease, may be used between the TEM 350b and the fingers 432, 434.
Turning to
Because the heat flux through each TEM 510a, 510b, 510c is lower in the power-generation embodiment, the temperature difference between the two sides of the TEM is reduced relative to a higher flux case. Thus, an electrical component, e.g., being cooled by the apparatus 500 may operate at a lower temperature than if a single TEM with a higher heat flux were used to generate power. In other words, the apparatus 500, by virtue of spreading the heat flux over a larger number of pellets, allows an electrical component to operate at a lower temperature for the same total heat flux from the electrical component through the several TEMs. This aspect may allow a greater portion of the heat dissipated by the electrical component to be harnessed to produce power than would otherwise be possible without risking reducing the lifetime of the component due to high temperature operation. This aspect is in contrast to the current state of the art in thermal power recovery, in which typically a relatively small portion of waste heat may be recovered from an electrical component because of the thermally insulating characteristic of the TEM.
The TEMs 510 may be connected electrically in series, e.g., to a load R to produce I2R watts of power. As described previously, the heat source 520 may be an electrical component or a conduit for hot exhaust. The scavenged power may be used for any desired purpose. Configuring the TEMs 510 in the manner described in the various embodiments herein provides that a greater fraction of the available waste heat is converted to useful power. In other embodiments, not shown, the TEMs 510 are connected electrically in parallel. In such embodiments the recovered power is provided at higher current and lower voltage than in the illustrated series configuration.
The working fluid and vapor circulate independently in each heat pipe 610a, 610b, 610c. This aspect of the embodiment provides a means to tailor the heat flow from different areas of a TEM in thermal contact with a thermally conductive block 620. For example, the heat pipe 610b may be configured to transport heat at a greater rate than the heat pipes 610a, 610c to remove heat from an interior region of a TEM away from the edges thereof. The heat pipe 610b may be, e.g., configured with a different diameter than the heat pipes 610a, 610c.
Finally, turning to
Those skilled in the art to which the invention relates will appreciate that other and further additions, deletions, substitutions and modifications may be made to the described embodiments without departing from the scope of the invention.
