This application is a divisional application of the invention application entitled "cooling system for computer system" filed on 6.5.2005 and having application number 200580050009.2.
Disclosure of Invention
It may be an object of the invention to provide a small and compact liquid cooling solution which is more efficient than existing air cooling devices and which can be produced at low cost and which can increase the production volume. Another object of the invention may be to produce a liquid cooling device which is easy to use and apply and which requires a low level of maintenance or no maintenance at all. It may be a further object of the invention to produce a liquid cooling device which can be used with existing CPU types and which can be used in existing computer systems.
This object is achieved by a cooling system for a computer, said computer system comprising:
at least one unit generating thermal energy, such as a Central Processing Unit (CPU), and said cooling system for cooling said at least one processor,
a reservoir having a volume of cooling fluid for concentrating and transferring thermal energy dissipated to the cooling fluid from the processor,
a heat exchange interface for providing thermal contact between the processor and the cooling liquid to dissipate heat from the processor to the cooling liquid,
a pump provided as part of an integrated component comprising the heat exchanging interface, the reservoir and the pump,
the pump for pumping the cooling liquid into the reservoir, through the reservoir and from the reservoir to a heat sink,
the heat sink is used for radiating heat energy dissipated to the cooling liquid from the cooling liquid to the periphery of the heat sink.
By providing integrated components, it is possible to limit the number of individual components of the system. However, there is no practical need to limit the number of components, as there is typically sufficient space within the chassis of a computer system to accommodate the different individual components of a cooling system. Thus, surprisingly, there has been no attempt at all to integrate some of the components.
In a possible embodiment according to this aspect of the invention the entire pump is placed inside the reservoir with at least one inlet or outlet to the cooling liquid in the reservoir. In an alternative embodiment the pump is placed outside the reservoir in the vicinity of the reservoir, wherein at least one inlet or outlet is directly connected to the cooling liquid in the reservoir. By placing the pump inside the reservoir or directly adjacent to the outside of the reservoir, an integration of the combined reservoir, heat exchanger and pump is obtained, so that the components are easy to apply in new and existing computer systems, in particular in mainstream computer systems.
In a preferred embodiment, the pumping member of the pump and the motor-driven part of the pump (e.g. the rotor of the electric motor) are placed inside the reservoir, submerged in the cooling liquid, and wherein the stationary part of the motor of the pump (e.g. the stator of the electric motor) is placed outside the reservoir. By placing the driving part of the motor inside the reservoir, submerged in the cooling liquid, and placing the stationary part of the motor outside the reservoir, there is no need to encapsulate the stationary part in a liquid tight isolation. But a problem may arise in that the fixed part drives the driving part. The present invention provides a means to achieve such an effect, though not at all apparent as to how to solve the problem.
The object is also achieved by a cooling system for a computer system, the computer system comprising:
at least one unit generating thermal energy, such as a Central Processing Unit (CPU), the cooling system for cooling the at least one processor,
a reservoir having a volume of cooling fluid for concentrating and transferring thermal energy dissipated to the cooling fluid from the processor,
a heat exchange interface for providing thermal contact between the processor and a cooling liquid for dissipating heat from the processor to the cooling liquid,
a pump for pumping the cooling liquid into the reservoir, through the reservoir and from the reservoir to a heat sink, and
the cooling system is adapted to be in thermal contact with the processor via an existing fixture associated with the processor, an
The heat sink is used for radiating the heat energy dissipated to the cooling liquid from the cooling liquid to the periphery of the heat sink.
The use of existing fixing means has the advantage that: the installation of the cooling system is quick and simple. Again, however, it will be clear to the skilled person that the use of a specially adapted mounting device for any component of the cooling system is not problematic, since there are many possibilities for mounting any kind of any number of components and cooling system components in an existing cabinet of a computer system.
In a preferred embodiment according to this aspect of the invention, there is a fixing means for fixing the heat sink to the processor, or there is a fixing means for fixing the cooling fan to the processor, or there is a fixing means for fixing the heat sink together with the cooling fan to the processor. Existing fixtures of the kind mentioned are typically used for air cooling of computer system CPUs, but air cooling devices are much simpler than liquid cooling systems. However, it has been possible to delicately develop a complex and efficient coolant system that is capable of applying such existing fixtures for simple but inefficient air cooling devices.
According to an aspect of the invention, the pump is selected from the following types:
bellows pumps, centrifugal pumps, diaphragm pumps, rotary pumps, flexible line pumps, flexible vane pumps, gear pumps, peristaltic tubing pumps, piston pumps, screw pumps, pressure washer pumps, rotary lobe pumps, rotary vane pumps, and electrokinetic pumps. By using one or more of the solutions of the invention, a wide variety of pumps can be used without departing from the scope of the invention.
According to another aspect of the invention, the drive means for driving the pump are selected from the following drive means: an electric rotating machine, a piezoelectric drive motor, a permanent magnet drive motor, a hydraulic motor, and a capacitor drive motor. By employing one or more of the solutions of the invention when selecting a pump for pumping a liquid, a variety of pumps may be used without departing from the scope of the invention.
The object of the invention is also achieved by a cooling system for a computer system, the computer system comprising:
at least one unit generating thermal energy, such as a Central Processing Unit (CPU), and a cooling system for cooling the at least one processor,
a reservoir having a volume of cooling fluid for concentrating and transferring thermal energy dissipated to the cooling fluid from the processor,
a heat exchange interface for providing thermal contact between the processor and the cooling liquid for dissipating heat from the processor to the cooling liquid,
a pump for pumping the cooling liquid into the reservoir, through the reservoir and from the reservoir to a heat sink, and
the cooling system further comprising a pump, wherein the pump is driven by an AC motor that is powered by a DC power source of the computer system,
wherein at least a portion of the power from the power source is used to convert to alternating current to the motor.
It is advantageous to use an ac motor, for example a 12V ac motor, to drive the pump to obtain a stable unit that may have to operate 24 hours a day, 365 days a year. However, those skilled in the art will find that the example of a 12V motor need not be employed because high voltages, such as 220V or 110V, are readily available because this is the voltage used to power the voltage supply of the computer system itself. While the 12V motor was chosen for the pump, those skilled in the art would not have chosen an ac motor in the future. The voltage provided by the voltage supply of the computer system itself is a dc voltage and thus this will be the type of voltage chosen by the skilled person.
In a preferred embodiment according to any aspect of the invention, the motor is adapted to simultaneously drive a pump for pumping liquid and a fan for creating an air flow around the reservoir, or the motor is adapted to simultaneously drive a pump for pumping liquid and a fan for creating an air flow around the heat sink, or the motor is adapted to simultaneously drive a pump for pumping liquid and a fan for creating an air flow around the reservoir and a fan for creating an air flow around the heat sink.
By utilizing a single motor driving more than one component of the cooling system according to any aspect of the present invention, the cooling system will be further facilitated to be simpler and more reliable.
The heat exchanging interface may be a component separate from the reservoir, wherein the heat exchanging interface is fixed to the reservoir in such a way that: the heat exchanging interface constitutes a part of the reservoir when being fixed to the reservoir. Optionally, the heat exchanging interface constitutes an integrated surface of the reservoir, and wherein the heat exchanging surface extends along a surface area of the reservoir, which surface area is designed to face the processing unit and which surface area is designed to be in close thermal contact with the processing unit. Also optionally, the heat exchanging interface is constituted by a free surface of the processing unit, and wherein the free surface is capable of forming heat dissipation between the processing unit and the cooling liquid through an aperture provided in the reservoir, and wherein the aperture extends along a surface area of the reservoir, the surface being designed to face the processing unit.
Possibly, uneven surfaces, such as pins and fins extending from the copper sheet provide a network of channels across the inner surface of the heat exchanging interface. As long as the cooling fluid is in thermal contact with the heat exchanging interface, the network of channels ensures that the cooling fluid is transported along the inner surface of the interface, e.g. a copper sheet, in such a way that: maximizing the retention time of the cooling liquid along the heat exchange interface and being conveyed in such a way that: optimizing heat exchange between the heat exchange interface and the cooling fluid.
Possibly, the cooling system may be provided with: a heat exchange interface for providing thermal contact between the processor and the cooling liquid to dissipate heat from the processor to the cooling liquid,
a pumping means for pumping the cooling liquid into the reservoir, through the reservoir and from the reservoir to a heat sink,
the heat sink is configured to dissipate thermal energy dissipated to the cooling fluid from the cooling fluid to a surrounding of the heat sink,
the heat exchange interface constitutes a heat exchange surface made of a material suitable for heat conduction, and
has a first side facing the heat exchanging surface of the cpu which is substantially planar, and
has a second side of said heat exchanging surface facing said cooling liquid, which is substantially planar, and
the reservoir is manufactured from plastic and channels or segments provided in the reservoir are intended to form a certain flow path for the cooling liquid through the reservoir.
Providing a planar heat exchange surface, the first, inner side being in thermal contact with the cooling liquid and the second, outer side being in contact with the heat generating processor, results in a reduction of the cost of manufacturing the heat exchange surface to an absolute minimum.
According to a possible solution described above, the inlet of the pumping means is arranged in the vicinity of the heat exchange interface, to thereby obtain a vortex of the flow of the cooling liquid in the vicinity of the heat exchange interface. The flow vortices are advantageous for obtaining heat dissipation. If the heat exchange interface is planar, the pump inlet is arranged as described above, which may result in a vortex of flow occurring along the heat exchange interface, at least in the vicinity of the pump inlet, but possibly also away from the inlet.
Alternatively or additionally, the outlet of the pumping device is arranged in the vicinity of the heat exchange interface to thereby obtain a vortex of the flow of the cooling liquid in the vicinity of the heat exchange interface. The flow vortices are advantageous for obtaining heat dissipation. If the heat exchange interface is planar and the outlet of the pump is arranged as described above, this may result in a vortex of flow occurring along the heat exchange interface, at least in the vicinity of the inlet of the pump, but possibly also away from the inlet.
However, a flat first, inner surface may also result in the coolant passing the heat exchanging surface too fast. This can be remedied by providing grooves along the inner surface, thereby providing flow paths at the heat exchanging surface. But this leads to an increased cost of manufacturing the heat exchanging surface.
A solution to this problem has been solved by providing instead channels or segments in the reservoir housing. The reservoir housing may be manufactured by injection moulding or casting, depending on the material from which the reservoir housing is made. Providing channels or segments during the moulding or casting of the reservoir housing is more cost-effective than machining grooves along the inner surface of the heat exchanging surface.
Possibly, the cooling system may be provided with at least one reservoir, mainly for dissipating or radiating heat, which is collected and transported by said cooling liquid,
the cooling system is adapted to provide transfer of the heat from the heat dissipation surface to the heat radiation surface, wherein
The at least one reservoir is provided with an aperture for being closed by placing the aperture partly covering or optionally entirely covering the at least one processor, the aperture being placed in such a way that: the free surface of the disposer is in direct heat exchange contact with the interior of the reservoir through the aperture and thus with the cooling liquid in the reservoir.
The heat dissipation from the processor to the cooling liquid must be very efficient to ensure proper cooling of the processor, especially in the case of a CPU, where the heat dissipation surface is limited by the surface area of the CPU. This can be remedied by utilizing a heat exchange surface made of a material with high thermal conductivity, such as copper or aluminum, and ensuring proper thermal bonding between the heat exchange interface and the CPU.
However, in a possible embodiment according to the features in the above paragraph, the heat dissipation takes place directly between the disposer and the cooling liquid by providing an aperture in the reservoir, said aperture being adapted to occupy a free surface of the disposer. Thereby, the free surface of the processing unit extends into the reservoir or forms part of the boundary of the reservoir, and the cooling liquid has a free surface close to the processing unit.
A possible heat exchange interface may be, for example, direct contact between a heat generator of the CPU and a cooling liquid, wherein
At least one unit, such as a Central Processing Unit (CPU), generates thermal energy and the cooling system is for cooling the at least one processor, the cooling system comprising:
at least one reservoir, primarily for dissipating or radiating heat, which is collected and transported by the cooling liquid,
the cooling system is adapted to provide transfer of the heat from the heat dissipation interface to the heat radiation surface, wherein
The at least one reservoir is provided with an aperture for being closed by placing the aperture partly covering or optionally entirely covering the at least one processor, the aperture being placed in such a way that: the free surface of the disposer is in direct heat exchange contact with the interior of the reservoir through the aperture and thus with the cooling liquid in the reservoir.
The aperture of the reservoir is designed to be closed by fixing a border of the aperture to a free surface of the disposer, which border, when fixed to the free surface of the disposer, is liquid tight so that liquid can flow freely across the free surface without risk of the liquid dissipating through the border. In addition, by forming the same technical effect, the aperture of the reservoir is designed to be closed by fixing the border of the aperture along the free surface of the disposer.
If the heat sink is arranged to assist heat dissipation from e.g. a CPU heat generating unit, the aperture of the reservoir may be adapted to be closed by fixing the border of the aperture to the free surface of the heat sink. Further, the aperture of the reservoir may be adapted to be closed by fixing the border of the aperture along the free border of the heat sink. In addition, the first and second substrates are,
possibly, the heat exchange interface may be arranged to:
a first reservoir for being closed by affixing a boundary of an aperture in said first reservoir to a free surface of said treatment device, optionally along said free surface of said treatment device, and
a second reservoir for being closed by fixing a boundary of an aperture in said second reservoir to a free surface of a heat sink, optionally along the free surface of the heat sink, and
liquid conducting means arranged between the first reservoir and the second reservoir.
The first reservoir may be closed by fixing the first reservoir to a part of a heat exchanging surface in close thermal contact with the processor for dissipating heat from the processor to the cooling liquid in the first reservoir, and wherein the second reservoir is closed by fixing the second reservoir to a surface of a heat sink for radiating heat from the cooling liquid in the second reservoir to the external environment.
Furthermore, the first reservoir and the second reservoir may be provided as a monolithic structure comprising the first reservoir and the second reservoir, and wherein heat is dissipated both from the processor to the cooling liquid in the first reservoir and from the cooling liquid in the second reservoir to the external environment by means of the monolithic structure. The monolithic structure may preferably be manufactured at least partially, preferably completely, from plastic, and the monolithic structure is manufactured by injection molding.
The transfer of the cooling liquid from the outlet of the first reservoir to the inlet of the second reservoir and from the outlet of the second reservoir to the inlet of the first reservoir, and the circulation of the cooling liquid within the liquid conducting means is provided by a pumping means for pumping the cooling liquid.
One of the reservoirs of the monolithic structure may comprise the pumping means.
The inlet and/or outlet of the pumping device and/or pumping members may be arranged near the substantially planar side to provide a vortex of flow and thereby improve the heat exchange between the cooling liquid and the substantially planar side, and the inlet of the pumping device may be arranged in the first reservoir and the outlet may be arranged in the second reservoir.
According to an aspect of the invention, a method is proposed, the method of cooling a computer system comprising at least one unit, such as a Central Processing Unit (CPU), generating thermal energy, and the method utilizing a cooling system for cooling the at least one processor, the cooling system comprising a reservoir, at least one heat exchanging interface and a pumping device, the cooling method comprising the steps of:
establishing, defining or selecting an operational state of the pumping device,
controlling operation of the motor of the pumping device in response to: a direction of movement required to obtain a pumping action of a pumping member of said pumping means, a possible direction of movement of a motor-driven part of said pumping means, and
controlling the operation of the computer system to achieve a desired direction of movement of the drive portion of the motor in accordance with the established, defined or selected operating state, thereby establishing a desired direction of movement to achieve the pumping action of the pumping member.
There may be a pumping arrangement in which the pumping member operates in only one direction, but in which the motor driving the pumping member operates in both directions. The solution to this problem is to select either a pumping member that can operate in both directions or a motor that operates only in one direction. According to the invention, a solution is provided in which the unidirectional pumping member is operable by a bidirectional motor. Despite the contradictory nature of this solution, advantages are still shown.
As an example, the method is used in a cooling system, wherein the pumping means is a rotating impeller having a unidirectional direction to obtain a pumping action, and wherein the motor of said pumping device is an alternating current motor having a rotor constituting a motor driving part, and wherein said method comprises the steps of: the rotational position of the rotor of the alternating current motor is established, defined or selected and at least one half-wave of the alternating current signal is applied to the stator of the alternating current motor before the full-wave alternating current signal is applied.
As an alternative example, the method is applied to a cooling system, wherein the pumping member is a rotating impeller having a unidirectional direction to obtain a pumping action, and wherein the motor of the pumping device is an alternating current motor having a rotor constituting a motor driving part, and the method comprises the steps of: establishing, defining or selecting a rotational position of a rotor of the alternating current electric machine and applying at least one half-wave of an alternating current signal to a stator of the alternating current electric machine after applying a full-wave alternating current signal.
In both of the above-mentioned examples, the advantages from a conventional dc pump one-way impeller and the advantages from a conventional ac pump motor are obtained in the above-mentioned solutions. The performance of a dc pump impeller is better than that of an ac pump impeller. The motor from the ac pump is more reliable than the motor from the dc pump. The advantage thus obtained is a synergistic behavior, since the different advantages of the impeller of the direct current pump and the motor of the alternating current pump are different in nature.
According to another aspect of the invention, a method of cooling a computer system comprising at least one unit, such as a Central Processing Unit (CPU), generating thermal energy, and utilizing a cooling system for cooling the at least one processor, the cooling system comprising a reservoir, at least one heat exchanging interface, a blower fan and a pumping device, the cooling method comprising the steps of:
applying one of the following possibilities how to operate the computer system:
establishing, defining or selecting an operational state of the computer system,
controlling operation of the computer system of at least one of the following devices: said pumping device and blowing fan being responsive to at least one of the following parameters: a surface temperature of the heat-generating processor, an internal temperature of the heat-generating processor, or a processing load of the CPU, and
controlling the operation of the computer system in accordance with the established, defined or selected operational state to achieve at least one of the following: a certain cooling performance of the cooling system, a certain electrical consumption of the cooling system, and a certain noise level of the cooling system.
During use of the computer system, the method described above is applied to ensure operation of the computer system, depending on the selected performance. For some applications, cooling performance is important, for example when graphic files may be processed or when large files are downloaded from a network, during which the processor is loaded highly and thus generates more heat. For other applications, power consumption is more important, for example when using home computer systems or in large office building environments where the power grid may be insufficient, for example in third countries. In other applications, the noise generated by the cooling system should be reduced to a certain level, which may be in a large office building or at home where white-collar workers only work, if the home computer is likely to be placed in the living room, or any other location where external factors have to be taken into account.
According to another aspect of the invention, a method is proposed for use with a cooling system further comprising a pumping device having an impeller for pumping the cooling liquid through a pumping housing, the pumping device being driven by an alternating current motor having a stator and a rotor, and the pumping device having means for sensing the position of the rotor, wherein the method comprises the steps of:
initially, establishing a preferred direction of rotation of the motor rotor,
sensing an angular position of the rotor prior to start-up of the motor,
during start-up, applying an alternating voltage signal to the motor and selecting the signal value, positive or negative of the alternating voltage signal at start-up of the motor,
said selection being made according to said preferred direction of rotation, an
The application of the alternating current signal, e.g., alternating voltage, by the computer system for applying an alternating current signal, e.g., alternating voltage, from the computer system power supply during the conversion of the direct current signal, e.g., direct current voltage, of the power supply into an alternating current signal, e.g., alternating voltage, for the motor.
The use of the above-described method according to the invention ensures the most efficient circulation of the cooling liquid in the cooling system and at the same time the lowest possible energy consumption of the motor driving the impeller. The efficient circulation of the cooling liquid is obtained by means of an impeller designed to rotate in a single direction of rotation, thus optimizing the impeller design with only one direction of rotation, as opposed to a double direction of rotation. The lowest energy consumption can be achieved because of the optimized impeller design, thus limiting the rotational speed required for the impeller to achieve a certain amount of flow of the cooling liquid through the cooling system. The unexpected effect of obtaining the lowest possible energy loss is that the lowest possible noise level of the pump is also obtained. The noise level of the pump depends, among other parameters, on the design and rotational speed of the impeller. Thus, the optimized impeller design and impeller speed reduces the noise level to the lowest possible in view of ensuring a certain cooling capacity.
Detailed Description
FIG. 1 is an exploded view of an embodiment of a prior art cooling device for a computer system. Typical components of an air-cooled CPU cooling arrangement are shown. The figure shows a prior art heat sink 4 for air cooling and having fins crossed by gaps, and a prior art fan to be mounted on top of the heat sink by using fastening means 3 and 6.
The fastening means comprises a frame 3 with holes for bolts, screws, rivets or other suitable fastening means (not shown) thereby fixing the frame to the motherboard 2 of the CPU 1 or another processor of the computer system. The frame 3 also has mortises provided in vertically extending uprights in each corner of the frame for occupying the tenons of a pair of uprights. The struts 6 serve to enclose the heat sink 4 and the fan 5, so that the fan and the heat sink are thereby fixed to the frame. When the frame is secured to the CPU board of another processor, and when the tenons of the struts are inserted into the mortises of the frame, using a suitable securing mechanism, the fan and heat exchanger are pressed towards the CPU using a force perpendicular to the CPU surface, which is provided by the lever arm.
FIG. 2 shows the components of the exemplary air-cooled CPU cooling arrangement of FIG. 1 when assembled. The components are fixed to each other and will be mounted on top of a CPU on a computer system motherboard (not shown).
FIG. 3 shows another embodiment of a prior art cooling system. Typical components in a liquid-cooled CPU cooling arrangement are shown. A prior art heat exchanger 7 is shown connected together with a prior art reservoir 8, a prior art liquid pump 9 and a heat radiator 11, and a fan 10 is arranged together with the heat radiator. A prior art heat exchanger 7, which may be mounted on a CPU (not shown), is connected to the radiator and the reservoir, respectively. The reservoir serves as a storage unit for excess liquid that cannot be contained in other components. The reservoir is also designed to serve as a means for venting any residual gas in the system and a means for filling the system with liquid. The heat radiator 11 serves as a means for removing heat from the liquid by blowing air through the heat radiator by means of the fan 10. All components are connected to each other by conducting liquid pipes serving as cooling medium.
FIG. 4 is an exploded view of a cooling system according to an embodiment of the present invention. Components not part of the cooling system are also shown. The figure shows a central processing unit CPU 1 mounted on the motherboard of a computer system 2. The figure also shows a part of the existing fastening means, i.e. the frame 3 or the like with mortises provided in vertically extending studs in each corner of the frame. The existing fastening means, i.e. the frame 3 and the stanchions 6, will be fixed to the main board 2 during use by bolts, screws, rivets or other suitable fastening means extending through four holes provided in each corner of the frame and through corresponding holes in the CPU board. The frame 3 will still provide an opening for the CPU to enable the CPU to extend through the frame.
The heat exchanging interface 4 is a separate component, made of a heat conducting material, such as copper or aluminum, having a relatively high heat conductivity, and will be in thermal contact with the CPU 1 when the cooling system is fixed to the CPU board 2. The heat exchanging surface constitutes a part of the reservoir housing 14 and thus the heat exchanger 4 constitutes a part of the reservoir housing 14 facing the CPU. The reservoir may for example be made of plastic or metal. The reservoir, which may be manufactured from a plastic material, or any other part of the cooling system, may be metallized to minimize liquid diffusion or evaporation of the liquid. The metal may be provided as a thin layer of metal coating disposed on either or both of the inside or outside of the plastic part.
If the reservoir is made of metal or any other material having a relatively high heat conductivity compared to e.g. plastic, the heat exchanging interface as a separate element may be excluded, since the reservoir itself may constitute a heat exchanger over the entire area of the reservoir in thermal contact with the processor. Alternatively, the heat exchanging interface constitutes the reservoir housing, which may then be fastened to the heat exchanging interface by means of screws, glue, solder, brazing or the like for fixing the heat exchanging interface to the housing and vice versa, possibly with a sealant 5 arranged between the housing and the heat exchanging interface.
Alternatively, the heat exchanging interface may be integrated with the reservoir containing the cooling liquid, it being possible to remove the heat exchanger and provide another means for dissipating heat from the processor to the cooling liquid in the reservoir. Another means is an aperture provided in the reservoir for directing the fluid towards the disposer. The boundary of the aperture will be sealed towards the boundary of the disposer or will be sealed on top of the disposer to thereby prevent leakage of cooling liquid from the reservoir. The only requirement for sealing is to provide a liquid tight connection between the aperture and the peripheral border of the processor or processor, e.g. the processor carrying card.
By removing the heat exchanger, a more efficient heat dissipation from the processor to the reservoir cooling liquid will be provided, since the intermediate elements of the heat exchanger are removed. The only obstacle in this sense is the provision of a liquid tight seal against leakage of the cooling liquid in the reservoir.
The heat exchanging surface 4 is typically a copper sheet. When removing the heat exchanging surface 4, which may be used not only in the embodiment shown in fig. 4, but also in all embodiments of the invention, it is then necessary to provide the CPU with a robust surface that will prevent evaporation of the cooling liquid and/or any damaging effects that the cooling liquid may exert on the CPU. The robust surface may be provided to the CPU by the CPU manufacturer or may be used later. The solid surface to be later used may be, for example, a layer such as an adhesive tape provided on the CPU. The adhesive tape may be made of, for example, a thin metal layer to prevent evaporation of the cooling liquid and/or any degradation of the CPU itself.
Within the reservoir, a liquid pump (not shown) is arranged for pumping cooling liquid from an inlet tubing connection 15 connected to the reservoir housing, through the reservoir, through a heat exchanger in thermal contact with the CPU, to an outlet tubing connection 16 also connected to the reservoir housing. The fastening means present comprise a pillar 6 with four tenons and a frame 3 with four corresponding mortises for fixing the reservoir and the heat exchanger to the CPU board. When the two parts of the fastener device present are fastened to each other, the fastening will, by means of the lever arm 18, generate a force to ensure thermal contact between the CPU 1 mounted on the motherboard and the heat exchanger arranged facing the CPU.
The cooling liquid of the cooling system may be any type of cooling liquid, such as water, an aqueous solution with, for example, an antimicrobial additive, an aqueous solution with an additive for improving heat transfer, or other cooling liquids of special composition, such as a non-conductive liquid or with a lubricant additive or an anti-corrosion additive.
Fig. 5 shows the components shown in fig. 4 assembled and fixed to the CPU board of the computer system 2. The heat exchanger and the CPU are in intimate thermal contact with each other. The heat exchanger and the rest of the reservoir 14 are fixed to the main board 2 by the presence of fastening means fixed to the CPU main board and by the force generated by the lever arm 18 of the presence of fastening means. A pipe inlet interface 15 and a pipe outlet interface 16 are provided to enable connection of a pipe to the interfaces.
Fig. 6 is an exploded view of the reservoir as previously shown in fig. 4 and 5 and seen from the opposite position, and also shows a pump 21 arranged inside the reservoir. Eight screws are provided for fixing the heat exchanging surface to the rest of the reservoir. The heat exchanging surface is preferably made of a copper sheet having a flat outer surface as shown in the figure, said outer surface being intended to abut the free surface of a heat generating component, such as a CPU (see fig. 4). However, the inner surface (not shown, see fig. 7) facing the reservoir is also plane. Thus, the copper sheet does not require any machining other than shaping the outer boundary for the octagonal shape shown in this embodiment and drilling a hole for inserting the bolt. No grinding of the inner and/or outer surface is required.
A sealing layer in the form of a gasket 13 is intended to be connected between the reservoir 14 and the heat exchanging surface forming a liquid tight connection. The pump is adapted to be placed in the reservoir. The pump has a pump inlet 20 through which the cooling liquid flows from the reservoir into the pump, and the pump has a pump outlet 19 through which the cooling liquid is pumped from the pump to the outlet connection. The figure also shows a cover 17 for the reservoir. The non-smooth inner surface of the reservoir and the fact that the pump is arranged inside the reservoir will provide a vortex of cooling liquid inside the reservoir.
However, instead of the non-smooth wall of the reservoir and the pump being arranged inside the reservoir, the reservoir may have channels or segments for forming a certain flow path of the cooling liquid through the reservoir (see fig. 9-10 and 15). The channels or segments are especially needed when the inner surface of the heat exchanging surface is plane and/or when the inner wall of the reservoir is smooth and/or if the pump is not arranged inside the reservoir. In each of the mentioned cases, the flow of the cooling liquid inside the reservoir may result in the cooling liquid passing the reservoir too fast and not residing in the reservoir for a sufficiently long time to absorb a sufficient amount of heat from the heat exchanging surface. By providing a reservoir interior channel or partition, a flow will be provided which forces the cooling liquid past the heat exchanging surface, increasing the amount of time the cooling liquid resides inside the reservoir, thus improving heat dissipation. If the channels or segments are arranged inside the reservoir, the shape of the channels and segments may be determined by whether the reservoir is made of plastic, possibly by injection moulding, or metal, e.g. aluminium, possibly by die casting.
The cooling liquid enters the reservoir through the tubing inlet connection 15 and enters the pump inlet 20 and is pumped out of the pump outlet 19 connected to the outlet connection 16. The connections between the reservoir and the inlet and outlet pipe connections form a liquid seal, respectively. The pump may not only be a separate pumping device but may also be integrated into the reservoir, thus forming an integrated part of the reservoir and the pumping device. The single integrated element of the reservoir and the pumping device may also be integrated, so that the reservoir, the pumping device and the heat exchanging surface may form a single integrated unit. This is possible, for example, if the reservoir is made of metal, such as aluminium. The choice of material thus offers the possibility of constructing the reservoir and the heat exchanging surface with a relatively high thermal conductivity, and also the possibility of providing bearings and similar structural elements for the motor and the pumping wheel as part of the pumping device.
In another embodiment the pump is placed in the vicinity of the reservoir, but outside the reservoir. By the pump being placed outside, but in the vicinity of the reservoir, an integrated component is still obtained. The pump or the inlet or outlet is preferably arranged to obtain a swirling flow of the flow in the vicinity of the heat exchange interface, thereby promoting an improved heat dissipation between the heat exchange interface and the cooling liquid. Even in alternative embodiments, pumping means such as impellers (see fig. 15-16) are provided adjacent the heat exchange surface. The pumping member itself typically introduces turbulence to the flow, thereby promoting improved heat dissipation regardless of the position of the pump itself or the inlet or outlet to the reservoir or pump.
The pump may be driven by an ac or dc motor. When driven by an ac motor, this can be done by converting the dc power supplied by the computer system to ac power for the pump, although there is no technology and power required in the computer system. The pump may be driven by a motor of any voltage common to a utility grid, e.g. 110V or 220V. However, in the illustrated embodiment, the pump is driven by a 12V AC motor.
In case the pump is driven by an ac motor, the control of the pump is preferably performed by the operating system of the computer system itself or similar, wherein the computer system comprises means for measuring the CPU load and/or the CPU temperature. Measurements made using the operating system of a computer system or similar system eliminate the need for special equipment for operating the pump. Communication between the operating system or similar system for operating the pump and the processor may be along a communication connection, such as a USB connection, that has been established in the computer system. Thus, real-time communication between the cooling system and the operating system can be set without a dedicated device for establishing communication.
In the case where the motor for driving the pump is an ac motor, the above-described method of controlling the pump may be used in combination with a method of: the pumping device is provided with means for sensing the position of the rotor of the electric machine and employs the steps of: firstly establishing a preferred direction of rotation of the rotor of the electric machine, before starting the electric machine, sensing the angular position of the rotor, during starting, applying an alternating voltage to the electric machine, and selecting the signal value, positive and negative, of the alternating voltage at the start of the electric machine, said selection being made according to the preferred direction of rotation, and said application of the alternating voltage being made by a computer system for supplying the alternating voltage from a power supply of the computer system during conversion of the direct voltage of the power supply into an alternating voltage for the electric machine. By generating an alternating voltage for the motor by the operating system of the computer system itself, the direction of rotation of the pump can be selected uniquely by the computer system, independent of the voltage applied by the utility grid supplying the computer system.
Further control strategies using the operating system of the computer system or similar systems may involve balancing the rotational speed of the pump as a function of the required cooling capacity. If a low cooling capacity is required, the rotational speed of the pump may be limited, thereby limiting the noise generated by the motor driving the pump.
In the case of a fan arranged in conjunction with a heat sink as shown in fig. 1 or a fan arranged in conjunction with a heat radiator, the operating system or similar system of the computer system may be designed to regulate the rotational speed of the pump, and thus the speed of the motor driving the pump, and to regulate the rotational speed of the fan, and thus the speed of the motor driving the fan. The adjustment should take into account the required cooling capacity, but the adjustment should take into account both cooling devices, i.e. the pump and the fan, which produces the most noise. Thus, if the fan is generating more noise overall than the pump, the adjustment will reduce the rotational speed of the fan before reducing the rotational speed of the pump when the required cooling capacity is low. Thereby, the noise level of the entire cooling system is reduced as much as possible. In the opposite case, i.e. the pump as a whole generates more noise than the fan, the rotational speed of the pump is reduced before the rotational speed of the fan is reduced.
An even further control strategy involves controlling the cooling capacity according to the type of processing performed by the computer. Some computer processes, such as word processing, impose less work on a processing unit, such as a CPU, than other types of computer processes, such as image processing. Thus, the type of processing performed on the computer system may be used as an indicator of cooling capacity. Depending on the type of treatment the user will handle, it may even be possible to establish some cooling mode as part of the operating system or similar. If the user selects, for example, word processing, then some cooling strategy is employed based on the need to limit cooling. If the user chooses to, for example, image processing, a cooling strategy is employed based on the need to increase cooling. Two or more different cooling modes may be established based on the capacity and control possibilities and the capabilities of the cooling system, and based on the intended use of the computer system selected by the user during use of the computer system or when hardware is selected during set-up of the computer system, i.e. before actual use of the computer system.
The pump is not limited to mechanical means but may be of any form capable of pumping coolant through the system. However, the pump is preferably one of the following types of mechanical pumps: bellows pumps, centrifugal pumps, diaphragm pumps, rotary pumps, flexible line pumps, flexible vane pumps, gear pumps, peristaltic tubing pumps, piston pumps, screw pumps, pressure washer pumps, rotary lobe pumps, rotary vane pumps, and electrokinetic pumps. Similarly, the motor driving the pump assembly need not be electric, but may also be a piezoelectric drive motor, a permanent magnet drive motor, a hydraulic drive motor, or a capacitive drive motor. The choice of pump and the choice of motor driving the pump depend on many different parameters and the type of pump and the type of motor are chosen by the person skilled in the art according to the specific application. For example, some pumps and some motors are better suited for small computer systems such as lab-tops, some pumps and some motors are better suited to create a high flow of cooling fluid, thus achieving high cooling efficiency, and even some pumps and motors are better suited to ensure low noise operation of the cooling system. Fig. 7 is a cut-away view inside the reservoir when the reservoir and the heat exchanging surface 4 are assembled and the pump 21 is placed inside the reservoir. The reservoir is provided with a tube inlet connection (not visible in the figure) through which the cooling liquid enters the reservoir. Subsequently, the cooling liquid flows through the reservoir via the heat exchanging surface and enters the inlet of the pump. After having passed the pump, the cooling liquid flows out of the outlet of the pump and further out through the pipe outlet interface 16. The figure also shows a cover 17 for the reservoir. The flow of the cooling liquid inside the reservoir and through the pump may be further optimized to use as low energy as possible to pump the cooling liquid, but still have a sufficient amount of heat dissipated from the heat exchanging surface into the cooling liquid. This further optimization may be established by varying the length and shape of the inlet of the pipe interface within the reservoir and/or by varying the position of the pump inlet and/or by having a pumping device placed near or around in thermal contact with the heat exchanging surface and/or by having channels or segments inside the reservoir.
In this case, the increased turbulence generated by the pumping means is used to improve the heat exchange between the heat exchange surface and the cooling liquid. Another or other way of improving the heat exchange is to force the cooling liquid through specially adapted channels or segments arranged inside the reservoir, or by making the surface of the heat exchanging surface sheet inside the reservoir uneven, or by using shaped fins with segments. In the shown figure the inner surface of the heat exchanging surface facing the reservoir is plane.
Fig. 8 is a perspective view showing a cooling system of the reservoir 14, said reservoir 14 having a heat exchanging surface (not shown) and a pump (not shown) inside the reservoir. The pipe inlet connection and the pipe outlet connection are connected to the heat radiator by means of connecting pipes 24 and 25, through which connecting pipes 24 and 25 the cooling liquid flows in and out of the reservoir and the heat radiator, respectively. In the heat radiator 11 the cooling liquid passes a number of pipes for heat dissipation, which is dissipated to the cooling liquid inside the reservoir and around the heat exchanger. The fan 10 blows air to the channels of the heat radiator to cool the heat radiator and thereby cool the cooling liquid flowing inside the channels past the heat radiator and back into the reservoir.
According to the invention, the heat radiator 11 can be arranged selectively. An alternative heat radiator consists of a heat sink, for example a standard heat sink made of extruded aluminum with fins on a first side and substantially planar on a second side. The fan may be disposed along the first side in connection with the fin. The reservoir is arranged along the second side of the heat sink and has at least one aperture for closing by covering said aperture over a part, or alternatively over the whole, of the substantially planar side of the heat sink. When closing the reservoir in this way, the heat sink surface is in direct heat exchanging contact with the interior of the reservoir and thus with the cooling liquid in the reservoir through the at least one aperture. This alternative method of arranging the heat radiator can be used in the embodiment shown in fig. 8 or as a heat radiator of another use and/or another embodiment of the invention.
The pumping means for pumping the cooling liquid through the reservoir may or may not be arranged inside the reservoir at the heat sink. The reservoir may have channels or segments for forming a certain flow path of the cooling liquid through the reservoir. The channels or segments are especially needed when the inner surface of the heat exchanging surface is plane and/or when the inner wall of the reservoir is smooth and/or if the pump is not placed inside the reservoir. In each of the mentioned cases, the flow of the cooling liquid inside the reservoir may cause the cooling liquid to flow through the reservoir too quickly, not being resident in the reservoir long enough to absorb enough heat from the heat exchanging surface. If the channels or segments in the reservoir are arranged inside the reservoir, the shape of the channels or segments may determine whether the reservoir is made of plastic, possibly injection moulded, or metal, e.g. aluminium, possibly die cast.
By means of the alternative heat radiator, the heat radiator 11 is not provided with the excessively expensive pipe structure shown in the figures, which guides the cooling liquid along the ribs for the improved surface of the structure, which ribs are connected to the channels. Instead, the heat radiator is provided as a unit as described consisting of a heat sink with or without a fan and a reservoir, and thereby a simpler and thus cheaper heat radiator than the heat radiator 11 shown in the figure is provided.
An alternative heat radiator provided as a unit consisting of a heat sink and a reservoir, with or without a pump inside the reservoir and with or without segments or channels, may be used alone, said heat radiator being intended to be placed in direct or indirect thermal contact with a heat generating processor, e.g. a CPU, or a heat exchanging surface, respectively. These embodiments of the invention may for example be used in a reservoir where the cooling liquid along a first side in the reservoir is in direct heat exchanging contact with a heat generating processor, such as a CPU, and the cooling liquid along a second side in the reservoir is in direct heat exchanging contact with a heat sink. Such a reservoir may be formed such that the heat exchanging surface area towards a heat generating processor, e.g. a CPU, is larger than the heat exchanging surface area towards a heat sink. This may for example have the effect of increasing the heat exchanging surface area to achieve an improved heat dissipation from e.g. a CPU to the heat sink compared to conventional heat sinks without a fixed reservoir. Conventional heat sinks typically only exchange heat with the CPU through the area provided by the area of the top side of the CPU. It has been found that a system comprising a reservoir and a heat sink with a fan is a simple, cost-optimized system with improved heat dissipation compared to a standard heat sink with a fan without a reservoir. In another embodiment of the invention, which can be derived from fig. 8, the fan and the heat radiator are placed directly in line with the reservoir. Thereby, the reservoir 14, the fan 10 and the heat radiator 11 constitute an integrated unit. Such an embodiment may provide the possibility to omit the connecting tubing and to transport the cooling liquid directly from the heat radiator to the reservoir via the inlet interface of the reservoir and directly from the reservoir to the heat radiator via the outlet interface of the reservoir. Such an embodiment may even provide the possibility that the pumping means of the liquid pump inside the reservoir and the motor for the blades of the fan 23 of the heat radiator 11 are driven by the same motor, thus making this motor the only one motor in the cooling system.
When the heat radiator is placed on top of the fan, which is at this time placed in direct alignment with the reservoir, and the heat radiator is connected directly to the inlet and outlet connections of the reservoir, no ducts are needed. However, if the heat radiator and the reservoir are not directly aligned with each other, a pipe is still required, and not only a pipe, possibly made of metal, such as copper or aluminium, may be used, such a pipe preventing any possible evaporation of the cooling liquid. Furthermore, the connection between such a tube and the heat radiator and the reservoir, respectively, may be a weld, so that the connection prevents evaporation of the cooling liquid.
In the embodiments derived from the description, the integrated unit of the reservoir, the heat exchanging surface and the pumping device will give a structure which improves the heat radiating properties, which is established in that the air flow of the fan can also be directed along the outer surface of the reservoir. If the reservoir is made of metal, the metal will be cooled by air that has passed the reservoir before or through the heat radiator. Such cooling of the reservoir by air will be further enhanced if the reservoir is made of metal and if the reservoir is provided with segments at the outer surface of the reservoir. The integrated unit just described will thus provide improved heat radiation performance, the heat radiation function normally performed by the heat radiator thus being supplemented by one or more other elements of the cooling system, i.e. the reservoir, the heat exchanging surface, the liquid pump and the fan.
Fig. 9-10 show an embodiment of the reservoir housing 14 in which channels 25 are provided inside the reservoir for forcing a cooling liquid to flow inside the reservoir. A channel 25 in the reservoir 14 leads from the inlet 15 to the outlet 16, like a labyrinth between the inlet and the outlet. The reservoir 14 is provided with an aperture 27, which aperture 27 has an outer dimension corresponding to the dimension of the free surface of the processing unit 1 to be cooled. In the embodiment shown, the processor to be cooled is CPU 1.
When the channel 26 is arranged inside the reservoir, the shape of the channel may determine whether the reservoir is made of plastic, possibly manufactured by injection moulding, or metal, such as aluminium, possibly manufactured by extrusion or die casting.
The reservoir 14, which may be manufactured from a plastic material, or any other part of the cooling system, may be metallized to minimize liquid diffusion or evaporation of the liquid. The metal may be provided as a thin metal coating, either or both on the inside or outside of the plastic part.
The CPU 1 is adapted to be disposed in the aperture 27 as shown in fig. 10 such that the outer boundary of the CPU engages the boundary of the aperture. It is possible that a sealant (not shown) may be provided along the CPU and aperture boundaries to ensure a liquid tight engagement between the CPU boundaries and the aperture boundaries. When the CPU 1 is arranged in the aperture 27, the free surface (not shown) of the CPU faces the reservoir, i.e. the part of the reservoir where the channels are arranged. Thus, when arranged in the hole 27 (see fig. 10), the free surface of the CPU 1 is in direct contact with the cooling liquid flowing through the channel 26 in the reservoir.
When the cooling fluid is forced to flow from the inlet 15 to the outlet 16 along the channel 26, the entire free surface of the CPU 1 will be traversed by the cooling fluid, thus ensuring proper and maximum cooling of the CPU. The channel structure may be designed and selected according to any one or more of the specifications, i.e. high heat dissipation, certain flow characteristics and ease of manufacture, etc. Thus, the channels may have other designs according to any need or requirement, and according to the type of CPU and the shape and size of the CPU free surface. Also, unlike CPUs, other processors may exhibit different heat dissipation needs, and may exhibit other shapes and sizes out of the surface, resulting in other configurations requiring channels. If the processor is elongate, for example an array of microprocessors, one or more parallel channels may be provided, possibly having only a common inlet and a common outlet.
Fig. 11-13 show an embodiment of the heat sink 4 in which the partitions 28 are arranged at a first side 4A of the heat sink and fins 29 for dissipating heat to the surroundings are arranged at the other, second side 4B of the heat sink. The intermediate reservoir 30 is provided with a recessed reservoir at the side facing the first side 4A of the heat sink. The recessed reservoir 30 has an inlet 31 and an outlet 32 at the other, opposite side facing the heat sink 4.
When the segments 28 are arranged on the first side 4A of the heat sink, the shape of the segments may determine whether the reservoir, which is made of a metal, such as aluminium or copper, is to be manufactured by extrusion or by other manufacturing methods, such as casting. Especially when the segments 28 are linear and parallel to the fins 29, extrusion is a feasible and cost-effective way of manufacturing the heat sink 4, as shown in the embodiments.
The intermediate reservoir 30, which may be manufactured from a plastic material, or any other component of the cooling system, may be metallized to minimize liquid diffusion or evaporation of the liquid. The metal may be provided as a thin metal coating provided on either or both of the inside or outside of the plastic part.
The recessed reservoir is provided with a kind of serrated edge 33 along opposite sides of the reservoir and the inlet 31 and the outlet 32 are provided at opposite corners of the intermediate reservoir 30, respectively. A partition 28 is placed at the first side 4A of the heat sink, i.e. the side facing the intermediate reservoir 30, so that when the heat sink is assembled with the intermediate reservoir housing (see fig. 13), the partition 29 extends from one serrated edge side of the reservoir to the other serrated edge side of the reservoir.
When the cooling liquid is forced to flow from the inlet 31 through the reservoir to the outlet 32 along the channels (not shown) formed by the segments 29 of the heat sink 4, the entire first side 4A of the heat sink will be passed by the cooling liquid, thus ensuring a normal and maximum heat dissipation between the cooling liquid and the heat sink. The structure of the segments on the first side 4A of the heat sink and the structure of the serrated edge side of the intermediate reservoir housing may be designed and selected according to any rules. Thus, the segments may have other designs, possibly wavy, or also jagged, depending on any desired flow characteristics of the cooling liquid, and depending on the type of heat sink and the size and shape of the reservoir.
Other types of heat sinks, which may be round heat sinks, may also exhibit different needs for heat dissipation, may exhibit other sizes and shapes of the free surface, leading to the need for other structures of the segments and the intermediate reservoir. If the heat sink and the reservoir are circular or oval, a spiral-shaped partition or a radially extending partition may be provided, possibly with an inlet and an outlet in the middle of the reservoir. If a pump impeller is provided, as shown in fig. 15-16, the pump impeller may be provided in the middle of the spiral-shaped segment or in the middle of the radially extending segment.
Fig. 14 shows the reservoir 14 shown in fig. 9-10 and the heat sink 4 and the intermediate reservoir 30 shown in fig. 11-13 assembled together to thereby constitute an integrated monolithic structure. It is not absolutely necessary to assemble the reservoir 14 of fig. 9-10 and the heat sink 4 and the intermediate reservoir 30 of fig. 11-13 together to obtain a proper cooling system function. The inlet 15 and the outlet 16 of the reservoir 14 of fig. 9-10 may be connected to the outlet 32 and the inlet 31, respectively, of the intermediate reservoir of fig. 11-13 by means of a tube or a conduit.
The reservoir 14 of fig. 9-10 and the heat sink 4 and the intermediate reservoir 30 of fig. 11-13 may then be arranged at different locations in the computer system. However, by assembling the reservoir 14 of fig. 9-10 and the heat sink 4 and the intermediate reservoir 30 of fig. 11-13 together, a compact monolithic unit is obtained, and the need for tubing or tubing is avoided. Pipes or conduits may involve an increased risk of coolant leakage, or may require welding or other special machining to eliminate the risk of coolant leakage. By eliminating the need for pipes or conduits, any leakage and any additional machining can be avoided when assembling the cooling system.
Fig. 15-16 show possible embodiments of the reservoir according to the invention. The reservoir is substantially identical to the reservoir shown in fig. 9-10. But the impeller 33 of the cooling system pump is arranged in direct communication with the channel 26. Furthermore, in the shown embodiment a heat exchanging interface 4, e.g. a surface made of a sheet of copper or other material with a high thermal conductivity, is arranged between the channel 26 inside the reservoir and the CPU 1 as a processor.
The heat exchanging surface 4 is preferably made of a copper sheet having a flat outer surface as shown, which outer surface is intended to abut the free surface of a heat generating component, such as a CPU 1 (see fig. 4). However, the inner surface (not shown, see fig. 7) facing the reservoir is also plane. Thus, the copper sheet does not require machining except to shape the outer boundary into a specifically adapted shape for the embodiment shown and drill holes for inserting bolts. It is not necessary to provide grinding of the inner and/or outer surfaces.
The provision of a heat exchanging surface 4 is not necessarily a preferred embodiment, since the solution in combination with the holes (see fig. 9-10) results in a direct heat exchange between the free surface of the CPU or other processing unit and the cooling liquid flowing along the channels in the reservoir. However, the heat exchanging surface enables the use of the cooling system independent of the type and size of the free surface of the CPU or other processor. Moreover, the heat exchanging surfaces enable replacement, repair, or other management of the cooling system without risk of coolant entering the computer system, which may not require complete or partial pumping of the cooling system.
In the shown embodiment the heat exchanging surface 4 is fixed to the reservoir by means of bolts. Other convenient fastening means may also be used. The heat exchanging surface 4, and thus the reservoir 14, may be fixed to the CPU 1 or other processor by any suitable means, such as welding, soldering or thermal bonding by bonding using glue. Optionally, special means (not shown) may be provided for ensuring thermal contact between the free surface of the CPU or other processor and the heat exchanging surface. One of such devices may be the stationary device shown in fig. 4 and 5 or a stationary device already provided as part of a computer.
When the channel 26 is arranged inside the reservoir 14, the shape of the channel may determine whether the reservoir is made of plastic, possibly injection moulded, or metal, possibly die cast, e.g. aluminium.
The reservoir 14, which may be manufactured from a plastic material, or any other part of the cooling system, may be metallized to minimize liquid diffusion or evaporation of the liquid. The metal may be provided as a thin metal coating provided on either or both of the inside and outside of the plastic part.
The impeller 33 of the pump (see fig. 4) is arranged in a separate recess of the channel 26, said separate recess having a size corresponding to the diameter of the impeller of the pump. The recess is provided with an inlet 34 and an outlet 35, respectively arranged opposite the inlet 31 and the outlet 32 of the cooling liquid into the channel 26 or out of the channel 26. The impeller 33 of the pump has a shape and design for only one-way rotation, in the embodiment shown only clockwise rotation. Thereby, the efficiency of the pump impeller is greatly improved compared to impellers capable of and intended for both clockwise and counter-clockwise rotation.
The increased efficiency of the impeller design results in the motor (not shown) driving the pump impeller being likely to be smaller than other impellers required to establish a normal and sufficient flow of coolant through the channel. In the preferred embodiment, the impeller is for a dc motor, although the motor is an ac motor, preferably a 12V ac motor. The paradox of using an ac motor to drive a dc impeller results in the use of a smaller motor that may be required to establish a normal and adequate flow of coolant through the channel.
The impeller may be driven by a motor of any voltage, e.g. 110V or 120V, common in the utility grid. The power supply of the computer system converts high voltage ac power to low voltage dc power. Thus, the impeller of the pump may be driven by an ac or dc motor. As mentioned above, the impeller of the pump is preferably driven by an ac motor. This can be accomplished by converting a portion of the dc power supplied by the computer system power supply to ac power for the pump's ac motor, although it is technically not necessary to use an ac motor and the use of an ac motor in a computer system that supplies dc power is detrimental to the use of electricity. However, in the illustrated embodiment, the impeller of the pump is driven by a 12V motor.
Fig. 17 shows a preferred possible embodiment of a reservoir according to the invention. The reservoir 14 has substantially the same features as the reservoir shown in fig. 15-16. In the shown embodiment the reservoir has a substantially conical, circular configuration and is provided with stiffening ribs 36 extending in axial direction outside the reservoir 14.
When designing and possibly injection moulding or casting the reservoir, it is possible to use, for example, a cylindrical, circular or conical rectangular or cylindrical, rectangular or even oval or triangular shape. The dimensions of the illustrated embodiment are about 55 mm in diameter and the axial elongation is also 55 mm.
The reservoir 14 has a recess 40 in the middle of the reservoir. The recess 37 is intended to receive a motor stator 37 which drives the impeller 33 of the pump, said impeller being fixed to a shaft 38 of a motor rotor 39. Recess 40 has a bore 41, four side walls 42, a bottom 43 and a circular sleeve 44 extending from bottom 43 of recess 40 and outwardly towards bore 41 of recess 40. The interior (not shown) of the sleeve 44 is intended to surround the pump rotor 39.
Thus, a liquid-tight partition is formed between the motor rotor 39 and the stator 37 of the pump, said rotor 39 being arranged inside the interior of the sleeve 44 and being immersed in the cooling liquid, said stator 37 being arranged in the recess 40 and surrounding the exterior of the sleeve 44. Thus, the stator 37 does not need to be sealed from the coolant, as the recess 40 and sleeve 44 together ensure that the stator remains dry, separate from the coolant, but the stator 37 is still able to drive the rotor 39 when powered by the power supply (not shown) of the computer system.
Along the outer circumferential enlargement the reservoir 14 is provided with a protrusion 45 extending outwardly from the outer circumferential enlargement. The protrusion is adapted to mate with an annular band (see description below) that secures the reservoir to a computer system CPU or other processor. The protrusions 45 are shown as a plurality of individual protrusions. Alternatively, the protrusion may be only one continuous protrusion extending outwardly and around an increasing portion of the circumference.
The reservoir 14 may also be provided with an inlet (not shown) and an outlet (not shown) for the cooling liquid. The inlet and the outlet are arranged along the downwardly and inwardly facing surface of the reservoir when seen in perspective in the drawings. The inlet and outlet lead to a radiator (not shown) for cooling the cooling liquid (see description below) that has been heated by the processor through the heat exchanging surface.
The radiator may be placed close to or remote from the reservoir 14, depending on the configuration of the computer system. In a possible embodiment the radiator is placed in the vicinity of the reservoir and may thus not comprise a duct extending between the radiator and the inlet and outlet, respectively. Such an embodiment provides a very compact structure of the entire cooling system, i.e. a monolithic structure, wherein all elements required for the cooling system are combined in one unit.
In an alternative embodiment the reservoir 14 itself also constitutes the radiator of the cooling system. In such an embodiment, no inlet and outlet are required. If the reservoir is made of e.g. copper or aluminium or another material with a high thermal conductivity, the cooling liquid may radiate heat through the outer surface of the reservoir 14 itself (see description below) after having been heated by the processor through the heat exchanging surface 8. In such an embodiment, the ribs 36 along the outer surface of the reservoir 14 may also serve as or replace cooling fins. In such an embodiment, the fins would have a smaller dimension than the transverse dimension of the ribs 14 shown in fig. 17, and the number of fins would be greater than the number of fins shown in fig. 17.
The impeller 33 of the cooling system pump is arranged in direct communication with a pump chamber 46 having an outlet 34, said outlet 34 being arranged tangentially to the circumference of the impeller 33. Thus, the pump acts as a centrifugal pump. The inlet to the pump chamber 46 is the entire opening into the pump chamber structure cavity which is also in direct communication with the interior of the reservoir 14. An intermediate member 47 is arranged between the pump chamber 46 together with the interior of the reservoir and the heat exchanging interface 4. The intermediate member 47 is provided with a first passage 48 for leading the cooling liquid from the pump chamber 46 to a heat exchange chamber (not shown) provided opposite the intermediate member 47. The intermediate member 47 is further provided with a second channel 49 for conducting cooling liquid from a heat exchange chamber (not shown) arranged at the opposite side of the intermediate member 47 to the interior of the reservoir 14.
In the embodiment shown, a heat exchange interface 4, for example a surface made of a copper sheet or other sheet of material having a high thermal conductivity, is provided in thermal communication with a heat exchange chamber (not shown) on the opposite side of the intermediate member 47.
The heat exchanging interface 4 is preferably made of a copper sheet having a flat outer surface (not shown) for abutting the free outer surface of a heat generating component, such as a CPU, on the opposite side to the side shown in the figure (see fig. 4). The inner surface facing the heat exchange chamber (not shown) on the opposite side of the intermediate member 47 is provided with pins 4A extending from the bottom of the copper sheet and into the heat exchange chamber (not shown) on the opposite side of the intermediate member 47. The pins 4A constitute an uneven surface and may be provided either in the copper sheet casting process or by grinding or other machining process of the copper sheet. The pins provide a network of channels across the inner surface of the heat exchanging interface along which a cooling fluid can flow.
In addition, the inner surface of the copper plate facing the reservoir is also planar. In this alternative embodiment, the copper sheet does not require machining, except to shape the outer boundary into a specifically adapted shape in the embodiment shown. When both the outer and inner surfaces are planar, no grinding or other machining processes need be provided for the inner and/or outer surfaces.
The provision of a heat exchanging interface 4 is not necessarily a preferred embodiment, since the solution in combination with the holes (see fig. 9-10) results in a direct heat exchange between the free surface of the CPU or other processing unit and the cooling liquid flowing along the channels in the reservoir. However, the thermal exchange interface enables the use of a cooling system that is independent of the type and size of the free surface of the CPU or other processor. Moreover, the heat exchange interface also enables replacement, repair, or other management of the cooling system without risk of coolant entering the computer system, and may not require complete or partial withdrawal of the cooling system.
In the embodiment shown, the heat exchange interface 4 is fixed to the intermediate part 47 by means of glue or other fixing means ensuring a correct and liquid-tight sealing of the heat exchange interface with the intermediate part. Any other suitable and convenient means for ensuring that the heat exchange interface is secured to the intermediate member (not shown) is contemplated.
The heat exchanging interface and thus the reservoir is fixed to the top of the CPU by means of an annular collar 50. The annular band 50 has an annular configuration and has four legs 51 extending axially from the annular configuration. The four legs 51 are provided with supporting feet 52 and the supporting feet 52 are provided with holes 53. The annular band 50 is intended to move around the outside of the reservoir 14 and further axially to the protrusion 45 of the reservoir 14.
After having been placed around the reservoir 14, the annular band 50 is fixed to the main board of the computer system by bolts (not shown) or similar fixing means extending through holes 53 in the supporting feet 52 and further through corresponding holes in the main board. The corresponding holes in the motherboard are preferably holes already present in the motherboard near the CPU and the CPU socket, respectively. Thus, the legs 51 and the support feet 52 of the annular hoop 50 are specially designed according to the holes already provided in the main plate.
Further, the heat exchanging interface 4 and thus the reservoir 14 may be fixed to the CPU or other processor by any other suitable means, such as welding, soldering or thermal bonding by means of bonding glue. In addition, special means (not shown) may be provided for ensuring thermal contact between the free surface of the CPU or other processor and the heat exchanging interface. One such means may be the fixtures shown in fig. 4 and 5 or fixtures already provided as part of the computer system.
When the reinforcing and/or cooling fins 36 are provided outside the reservoir 14, the shape and number of the fins may determine whether the reservoir is made of plastic, possibly injection moulded, or metal, such as aluminium, possibly die-cast. Furthermore, the function of the fins, i.e. to reinforce the reservoir only or to compromise cooling purposes, may determine whether the reservoir is made of plastic, possibly injection moulded, or metal, e.g. aluminium, possibly die-cast.
The reservoir 14, which may be manufactured from a plastic material, or any other part of the cooling system, may be metallized to minimize liquid diffusion or evaporation of the liquid. The metal may be provided as a thin metal coating provided on either or both of the inside or outside of the plastic part.
The impeller 33 of the pump has a shape and design for unidirectional rotation only, in the embodiment shown only clockwise rotation. Thereby, the efficiency of the impeller of the pump is greatly improved compared to impellers capable of and intended for both clockwise and counter-clockwise rotation.
The increased efficiency of the impeller design results in that the motor (not shown) driving the impeller of the pump may be smaller than other motors required for establishing a normal and sufficient flow of coolant through the channels. In the preferred embodiment, the motor is an ac motor, preferably a 12V ac motor, although the impeller is for a dc motor. The use of an ac motor to drive a dc impeller contradicts the possibility of establishing a smaller motor required for proper and sufficient flow of coolant through the channel.
The impeller may be driven by a motor of any voltage, e.g. 110 vac or 220 vac, common in the utility grid. The power supply of the computer system converts the high voltage ac power to low voltage dc power. Thus, the impeller of the pump can be driven by either an ac or dc motor. As mentioned above, preferably the impeller of the pump is driven by an ac motor. While the use of ac motors is technically unnecessary and is detrimental to the use of electricity in computer systems that supply dc power, this can be accomplished by converting a portion of the dc power supplied by the computer system power supply to ac power for the pump ac motor.
In each aspect of the invention, in which an ac motor is used to drive an impeller driven by a dc motor, although this method of constructing the pump is contradictory, the following preferred modes of operation are established to mitigate the disadvantages:
in order to be able to control the direction of rotation of the impeller connected to the rotor and to optimize the condition of the maximum mean torque value during start-up, i.e. from zero speed to synchronous speed, an electronic control circuit is used. The electronic control circuit comprises a processor which drives a static power switch, constituted for example by a triac arranged in series between an alternating power source, obtained by direct current of the computer system, and an alternating current motor. The output of the current monitor is an input signal for an electronic processor.
The electronic control circuit may also comprise several sensors adapted to sense the position and polarity of the permanent magnets included in the rotor of the alternating current motor, when the rotor is in motion, and when it is in certain operating conditions, or when it is stationary or kept at zero speed. The number of position sensors may be hall sensors, encoders or optical or electromechanical sensors capable of establishing and/or measuring the rotor position. The output signals from the number of position sensors are input signals for an electronic processor.
Additionally, the output signal from the position sensor may be phase shifted by an electronic phase shift circuit before the output signal is sent to the electronic processor input.
A third signal may be input to the processor, the third signal being capable of causing the processor unit to sense the polarity of the ac voltage applied to the ac motor. But the third signal is not necessary.
The signal input to the electronic processor is converted to digital form and, after processing by the processor, the output signal is provided by the processor. The output signal is used to close or open a static switch formed by a triac arranged in series with the ac motor.
In an electronic processor, a current signal provided by a current sensor enters a zero-crossing detector which provides an output as a logic "1" signal indicating that the current is near zero, with a positive or negative deviation from the zero value of the current. The deviation depends on the type of motor used and its application, as well as the type of static power switch used. The signal from the position sensor enters a phase shift and processing circuit whose output is either 1 or 0 depending on the position and polarity of the rotor.
In an electronic processor, the phase-shifted position signal and the signal processed from the alternating voltage enter an electronic logic XOR gate, which outputs a 1 signal if the value of the alternating voltage is equal to 0 and the value of the phase-shifted position signal is equal to 1, or the value of the alternating voltage is equal to 1 and the value of the phase-shifted position signal is equal to 0.
The output of the zero crossing detector AND the output of the XOR gate in numerical form thus enter an electronic logic AND gate which provides a control signal output for closing or opening the static power switch.
The AND gate AND the two input AND signal processing system allow two cases to be determined: 1) the alternating voltage signal is positive, the current is close to zero, and the rotor rotation angle is between 0 and 180 degrees; 2) the ac voltage signal is negative, the current is near zero, and the rotor rotation angle is between 180 and 360 degrees. Both cases provide the same direction of rotation of the rotor of the alternating current motor.
Fig. 18 shows an ac motor embodiment in which one stator pole 54 is longer than the other stator pole 55 by an amount denoted by l. By this arrangement, the permanent magnet rotor 39 having an ideal line 56 separating the north and south poles N, S of the rotor is arranged such that the ideal line 56 does not coincide with the centre line axis 57 of the stator 37, but the ideal line 56 is inclined at an angle a with respect to the centre line 57 of the stator 37.
Two field coils 58, 59 are provided on the two poles 54, 55 of the stator 37, respectively, and the field coils are connected in series and are supplied with power from an alternating current power supply via terminals (not shown). With this structure of the alternating current motor, the motor can be started more easily in a predetermined rotation direction of the rotor.
In a preferred embodiment of the invention, the control electronics only provide a half-wave voltage signal to the ac motor during start-up, thereby providing torque pulses to the rotor. Since only the motor half-wave signal is supplied, the torque pulse will always be unidirectional and will thus push the rotor to start rotating in the required direction. The required direction of rotation is determined by the design of the impeller connected to the rotor and the polarity of the half-wave voltage signal.
After a certain amount of time, in which several half-wave voltage signals have been supplied to the motor, the rotor will stop rotating in a certain position, for example as shown in the figure. The rotor thus enters a determined steady-state position independent of its starting position. After this process the ac motor is supplied with a full wave voltage signal which will accelerate the rotor until the motor enters synchronous operation, i.e. when the rotor rotates with the same periodic frequency as the frequency of the ac voltage source.
The initial polarity of the ac voltage signal determines the final direction of rotation of the rotor, so that if the initial voltage is positive, with increasing amplitude, the rotor will start in one direction, whereas if the voltage is negative, with decreasing amplitude, the rotor will rotate in the opposite direction.
The number of half-waves required to bring the rotor into a determined steady state position in which the rotor stops rotating depends on the characteristics of the motor, such as the moment of inertia and the external load applied to the rotor. The number of half-waves required thus depends on empirical analysis of the particular motor, in particular the load situation.
The half-wave voltage signal and the corresponding half-wave current signal supplied to the motor have waveforms as shown in fig. 19.
In another embodiment, the control electronics for driving the ac motor shown in fig. 18 are configured such that the control electronics instruct the ac motor to stop at a predetermined position by supplying the motor with a number of half-wave voltage signals after synchronous operation in which the motor is supplied with a full-wave voltage signal. Thus, when the motor needs to be started again, the rotor is already in a certain position, and therefore only the polarity of the full-wave alternating voltage signal supplied to the motor has to be selected so that the final direction of rotation of the rotor coincides with the end position of the rotor in the last run.
According to the method, there is no need for an initial step of bringing the motor into a determined steady-state position by supplying the motor with several half-wave voltage signals. Even in the above-described another embodiment, it is possible to terminate the full-wave power supply by several half-wave voltage signals and start the full-wave power supply by initially supplying the motor with several half-wave signals. However, this is more cumbersome, yet safer.
Fig. 19 shows the voltage signal V and the current signal I supplied to the ac motor and the position signal of the rotor. Initially the rotor remains still, indicated by the line L. The electronic control circuit controls the static power switch so that the voltage signal V and the current signal I are displayed in half-wave. The rotor thus receives torque pulses due to the current-voltage combination, which pulses are always unidirectional and are used to drive the rotor to start rotating in the desired direction. After the start-up phase, the rotor enters its synchronous running process.
Thus, it is possible to generate an alternating signal, preferably a 12V alternating signal, by means of digital electronic pulses supplied with a 12V direct current from a computer power supply. From the possible sensor outputs related to the impeller position it is determined how to start the alternating current signal, i.e. by the negative or positive half-wave, and by ensuring that the impeller is started in the same rotational direction each time, and thus the performance and efficiency of the alternating current pump is the same as those of the direct current pump.
In addition, the magnetic field sensor may be omitted and instead of reading the position of the impeller, the impeller is forced to start at the same position each time. To ensure that the impeller is in a defined position before it is started, a signal is supplied to the stator of the ac motor for a defined period of time. The signal may be supplied three times in succession according to the power supply curve. The pulses must be within the same half-wave portion of the signal period. Although the frequency of the pulse signal is an arbitrary value, it may be 50/60Hz, although an ac pump driven by an ac signal from the power outlet of the utility grid and converted from 230/115V to 12V cannot operate under normal circumstances, because it is not possible to change the sinusoidal signal from the utility grid.
In this way, the impeller will be forced to the correct polarity before start-up and when a full-wave electrical signal is supplied, the pump will turn the impeller in a defined manner of rotation. The supplied full-wave electrical signal must start with an opposite half-wave amplitude signal compared to the initial half-wave pulse supplied before the full-wave signal starts.
The present invention has been described with reference to particular embodiments and with reference to particular applications, but it is to be noted that different embodiments of the present invention can be manufactured, marketed, sold and used alone or in combination with one another. In the above detailed description of the invention, the description of one embodiment that may refer to one or more figures may be incorporated into the description of an embodiment that may refer to another one or more figures, and vice versa. Thus, any single embodiment or combination of two or more embodiments described in the text and/or drawings falls within the possible scope of the present application.