LOOP HEAT PIPE AND STARTUP METHOD FOR THE SAME

TECHNICAL FIELD

The present invention is related to a loop heat pipe and a startup method for the same.

BACKGROUND ART

Heat pipes are used for cooling electronic devices. A heat pipe is a heat transfer device that transports heat by utilizing the phase change of the working fluid sealed therein.

In order to enhance the cooling capability for cooling electronic devices, a heat pipe known as a loop heat pile has been developed that can transport a larger heat load over a longer distance.

The loop heat pipe includes an evaporator which receives heat from a heat source and vaporizes a liquid-phase working fluid, and a condenser which condenses the vapor-phase working fluid by giving off heat. The loop heat pipe further includes a vapor line through which the working fluid converted to the vapor phase by the evaporator is transported to the condenser, and a liquid line through which the working fluid converted to the liquid phase by the condenser is transported to the evaporator. The loop heat pipe has a loop structure in which the evaporator, the evaporator line, the condenser, and the liquid line are connected in series, and the working fluid is sealed therein.

In recent years, a blade server of the type that has two CPUs on one blade has been developed in order to enhance the processing capability of the server.

If two CPUs are to be cooled during operation by using a loop heat pipe, there arises a need to provide two evaporators in order to receive heat from the respective CPUs, which means that two loop heat pipes have to be incorporated into the blade server.

In order to incorporate two loop heat pipes into the blade server, an area for accommodating the two loop heat pipes needs to be provided on the substrate.

However, since the blade server was originally developed as a server more compact in volume than the conventional server, electronic devices including CPUs are packed at high density on the substrate.

There are therefore cases in which it is difficult to secure an area for accommodating two loop heat pipes on the substrate.

On the other hand, a loop heat pipe equipped with two evaporators has been proposed. A loop heat pipe of this type is depicted in FIG. 1.

The loop heat pipe 110 includes a first evaporator 111A and a condenser 112. The loop heat pipe 110 further includes a first liquid line 114A through which the working fluid converted to the liquid phase by the condenser 112 is transported to the first evaporator 111A, and a vapor line 113 through which the working fluid converted to the vapor phase by the first evaporator 111A is transported to the condenser 112.

Further, as depicted in FIG. 1, the loop heat pipe 110 includes a second evaporator 111B which assists in transporting the liquid-phase working fluid into the first evaporator 111A at the time of startup. A portion of the liquid-phase working fluid passed through the first liquid line 114A flows into the second evaporator 111B through a second liquid line 114B and through the condenser 112. The working fluid converted to the vapor phase by the second evaporator 111B merges with the working fluid flowing in the vapor line 113 and is transported to the condenser 12. The working fluid transported through the second liquid line 114B is passed through the condenser 12 and flows into the second evaporator 111B without merging with the working fluid flowing in the first liquid line 114A.

When starting up the loop heat pipe 110, the liquid-phase working fluid is quickly fed into the second evaporator 111B disposed near the condenser 112, thus starting the circulation of the working fluid through the loop and causing the liquid-phase working fluid to flow into the first evaporator 111A. The second evaporator 111B is an auxiliary evaporator provided to assist the startup of the loop heat pipe 110. Therefore, the second evaporator 111B has a smaller size and lower cooling capacity than the first evaporator 111A.

If such a loop heat pipe 110 having two evaporators 111A and 111B is used for cooling two CPUs substantially equal in heat load, the circulation of the working fluid through the loop tends to become unsteady because of the difference in cooling capacity between the two evaporators 111A and 111B and because of the arrangement of the fluid lines.

[Patent Document 1] Japanese Unexamined Patent Publication No. 2008-85112
[Patent Document 2] U.S. Patent Application No. 2004/0182550

SUMMARY

According to an aspect of the embodiment disclosed in this specification to solve the above problem, there is provided a loop heat pipe which includes: a first evaporator and a second evaporator each of which vaporizes a liquid-phase working fluid by receiving heat from a heat source and thereby converts the liquid-phase working fluid to a vapor-phase working fluid; a first condenser and a second condenser each of which condenses the vapor-phase working fluid by giving off heat and thereby converts the vapor-phase working fluid back to the liquid-phase working fluid; a first vapor line through which the working fluid converted to the vapor phase by the first evaporator is transported to the first condenser; a first liquid line through which the working fluid converted to the liquid phase by the first condenser is transported to the second evaporator; a second vapor line through which the working fluid converted to the vapor phase by the second evaporator is transported to the second condenser; and a second liquid line through which the working fluid converted to the liquid phase by the second condenser is transported to the first evaporator.

The object and advantages of the embodiments will be realized and attained by means of the elements and combinations particularly pointed out in the claims.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram illustrating a loop heat pipe according to the related art;

FIG. 2 is a diagram illustrating a first embodiment of a loop heat pipe disclosed in this specification;

FIG. 3 is a diagram illustrating a blade server in which the loop heat pipe of FIG. 2 is incorporated;

FIG. 4 is an enlarged longitudinal cross-sectional view of an evaporator in the loop heat pipe of FIG. 2;

FIG. 5 is an enlarged lateral cross-sectional view of the evaporator in the loop heat pipe of FIG. 2;

FIGS. 6(A) to 6(D) are diagrams illustrating the operation of the loop heat pipe of FIG. 2;

FIG. 7 is a diagram illustrating how the amount of received heat becomes unbalanced;

FIG. 8 is a diagram illustrating a second embodiment of the loop heat pipe disclosed in this specification;

FIGS. 9(A) to 9(D) are diagrams illustrating the operation of the loop heat pipe of FIG. 8;

FIG. 10 is a diagram illustrating a third embodiment of the loop heat pipe disclosed in this specification;

FIG. 11 is a diagram illustrating a blade server in which a fourth embodiment of the loop heat pipe disclosed in this specification is incorporated;

FIG. 12 is a diagram illustrating Working Examples 1 to 14 of the loop heat pipe disclosed in this specification;

FIG. 13 is a diagram illustrating Working Example 15 of the loop heat pipe disclosed in this specification;

FIG. 14 is a diagram illustrating Working Example 16 of the loop heat pipe disclosed in this specification;

FIG. 15 is a diagram illustrating Working Example 17 of the loop heat pipe disclosed in this specification; and

FIG. 16 is a diagram illustrating Working Example 18 of the loop heat pipe disclosed in this specification.

DESCRIPTION OF EMBODIMENTS

A first preferred embodiment of a loop heat pipe disclosed in this specification will be described below with reference to drawings. It will, however, be noted that the technical scope of the present invention is not limited to the specific embodiments disclosed herein, but extends to the inventions described in the appended claims and their equivalents.

FIG. 2 is a diagram illustrating the first embodiment of the loop heat pipe disclosed in this specification. FIG. 3 is a diagram illustrating a blade server in which the loop heat pipe of FIG. 2 is incorporated. FIG. 4 is an enlarged longitudinal cross-sectional view of an evaporator in the loop heat pipe of FIG. 2. FIG. 5 is an enlarged lateral cross-sectional view of the evaporator in the loop heat pipe of FIG. 2.

As illustrated in FIG. 2, the loop heat pipe 10 of the present embodiment includes a first evaporator 11A and a second evaporator 11B each of which vaporizes a liquid-phase working fluid 16 by receiving heat from a heat source and thereby converts the liquid-phase working fluid 16 to a vapor-phase working fluid 16. The loop heat pipe 10 further includes a first condenser 12A and a second condenser 12B each of which condenses the vapor-phase working fluid 16 by giving off heat and thereby converts the vapor-phase working fluid 16 back to the liquid-phase working fluid 16. Further, the loop heat pipe 10 includes a first vapor line 13A through which the working fluid 16 converted to the vapor phase by the first evaporator 11A is transported to the first condenser 12A, and a first liquid line 14A through which the working fluid 16 converted to the liquid phase by the first condenser 12A is transported to the second evaporator 11B. Furthermore, the loop heat pipe 10 includes a second vapor line 13B through which the working fluid 16 converted to the vapor phase by the second evaporator 11B is transported to the second condenser 12B, and a second liquid line 14B through which the working fluid 16 converted to the liquid phase by the second condenser 12B is transported to the first evaporator 11A.

In the loop heat pipe 10, a loop flow passage is formed by connecting in series the first evaporator 11A, the first vapor line 13A, the first condenser 12A, the first liquid line 14A, the second evaporator 11B, the second vapor line 13B, the second condenser 12B, and the second liquid line 14B.

The working fluid is hermetically sealed in the loop flow passage. The working fluid 16 transports heat while undergoing a phase change between the liquid phase and the vapor phase in the loop heat pipe 10. The working fluid 16 is hermetically sealed in the loop heat pipe 10 at saturated vapor pressure.

As the working fluid 16, for example, water, alcohol, ammonia, fluorocarbon, or the like, may be used.

In use, the loop heat pipe 10 is incorporated, for example, in a blade server 20, as illustrated in FIG. 3.

The blade server 20 is equipped with two CPUs 21A and 21B. The first evaporator 11A of the loop heat pipe 10 is disposed in thermal contact with the CPU 21A. The second evaporator 11B is disposed in thermal contact with the CPU 21B.

In many cases, the blade server 20 has an elongated rectangular shape as depicted in FIG. 3. The blade server 20 is usually oriented in such a manner that its width direction extending at right angles to its longitudinal direction coincides with the plumbline direction. Typically, the CPU 21A is located upwardly of the CPU 21B, as viewed along the plumbline direction.

Accordingly, in the loop heat pipe 10, the first evaporator 11A that receives heat from the CPU 21A is located upwardly of the second evaporator 11B that receives heat from the CPU 21B, as viewed along the plumbline direction.

A main fan 22 delivers air to the first and second condensers 12A and 12B to promote heat dissipation.

While FIG. 3 has depicted an example in which the loop heat pipe 10 is incorporated in the blade server, the loop heat pipe 10 may be incorporated in other electronic apparatus having a heat source to be cooled.

Next, the first evaporator 11A will be described in further detail below with reference to FIGS. 4 and 5. Since the second evaporator 11B is identical in structure to the first evaporator 11A, the description hereinafter given of the first evaporator 11A applies as well to the second evaporator 11B.

As illustrated in FIG. 4, the first evaporator 11A has a longitudinally extending shape. The longitudinal direction of the first evaporator 11A coincides with the direction in which the working fluid 16 flows through the flow passage of the loop heat pipe 10. In FIG. 4, the flow direction of the working fluid 16 is indicated by arrows.

As illustrated in FIGS. 4 and 5, the first evaporator 11A includes a longitudinally extending housing 30, a metal block 31 centrally located within the housing 30, a metal tube 32 located in a hollow space within the metal block 31, and a wick 33 located within the metal tube 32.

The housing 30, the metal block 31, and the metal tube 32 are each formed from a metal having high thermal conductivity such as copper.

The longitudinal direction of the housing 30 coincides with the longitudinal direction of the first evaporator 11A. The second liquid line 14B is connected to one longitudinal end of the housing 30. The first vapor line 13A is connected to the other longitudinal end of the housing 30.

The heat source such as the CPU 21A is thermally coupled to the housing 30 by means of a thermal bonding material such as thermal grease (not depicted).

The metal block 31 is in intimate contact with the inner surface of the housing 30 and is thus thermally coupled to the housing 30. The metal block 31 has a cylindrically shaped hollow core. The longitudinal direction of the hollow core coincides with the longitudinal direction of the first evaporator 11A. The metal block 31 quickly conducts the heat, received from the heat source 21A via the housing 30, to the metal tube 32 located in the hollow core.

The metal tube 32 has a longitudinally elongated cylindrical shape. The metal tube 32 is located in the hollow core of the metal block 31. The longitudinal direction of the metal tube 32 coincides with the longitudinal direction of the first evaporator 11A. The outer surface of the metal tube 32 is in intimate contact with the inner surface of the hollow core of the metal block 31, and thus the metal tube 32 is thermally coupled to the metal block 31.

As illustrated in FIG. 5, a plurality of projections 34a and depressions 34b are formed on the inner surface of the metal tube 32 at a prescribed pitch in the circumferential direction thereof. The projections 34a and depressions 34b are formed along the entire longitudinal direction of the metal tube 32. The groove-like space formed between the wick 33 and the depressions 34b provides a passage for the working fluid 16.

The wick 33 has a longitudinally elongated cylindrical shape, as depicted in FIG. 4. The wick 33 is open at one end thereof that faces the second liquid line 14B and closed at the other end thereof that faces the first vapor line 13A.

The wick 33 is inserted in the metal tube 32 with its closed end facing toward the first vapor line 13A. As illustrated in FIG. 5, the outer surface of the wick 33 contacts the tips of the plurality of projections 34a formed on the inner surface of the metal tube 32, and thus the wick 33 is thermally coupled to the metal tube 32.

The wick 33 is formed of a porous material. For example, the wick 33 is constructed from a porous member formed by sintering copper powder. Preferably, the hollow interior space of the wick 33 is made to communicate with the exterior thereof by means of numerous fine pores of diameters about 10 μm to 50 μm.

When the liquid-phase working fluid 16 flows into the first evaporator 11A from the second liquid line 14B, the working fluid 16 infiltrates into the wick 33 by capillary action, and the wick 33 is thus wetted with the working fluid 16. The liquid-phase working fluid 16 infiltrated into the wick 33 is heated and vaporized by the heat received from the heat source such as the CPU 21A.

The vapor-phase working fluid 16 existing in the wick 33 itself or on its surface or in the hollow interior space of the wick 33 is vented from the hollow interior space to the outside through the fine pores formed in the wick 33.

The housing 30 of the first evaporator 11A having the above structure may be chosen to have dimensions measuring 50 mm vertically, 50 mm horizontally, and 20 mm in height, compared with the CPU as the heat source which measures, for example, 30 mm vertically and 30 mm horizontally. The metal block 31 may be chosen to have dimensions measuring 40 mm vertically, 40 mm horizontally, and 20 mm in height. The metal tube 32 may be chosen to have an outer diameter of 14 mm and an inner diameter of 10 mm (tube wall thickness of 2 mm). The depressions 34b of depth 1 mm are formed in the inner surface of the metal tube 32, for example, at a pitch of 2 mm. The wick 33 may be chosen to have an outer diameter of 10 mm and an inner diameter of 4 mm.

Next, the first condenser 12A will be described in further detail below with reference to FIGS. 2 and 3. Since the second condenser 12B is identical in structure to the first condenser 12A, the description hereinafter given of the first condenser 12A applies as well to the second condenser 12B.

As depicted in FIG. 2, the first condenser 12A includes a first condenser line 40A and a plurality of first heat sinking plates 41A coupled to the first condenser line 40A.

The first vapor line 13A is connected to one end of the first condenser line 40A. The first liquid line 14A is connected to the other end of the first condenser line 40A.

The plurality of first heat sinking plates 41A are thermally coupled to the first condenser line 40A, and the heat of the working fluid 16 passing through the first condenser line 40A is dissipated via the plurality of first heat sinking plates 41A.

As depicted in FIG. 3, it is preferable to blow air over the plurality of first heat sinking plates 41A of the first condenser 12A by means of the main fan 22 or the like in order to dissipate heat and facilitate the phase change of the working fluid 16 from the vapor phase to the liquid phase.

Next, the first vapor line 13A will be described in further detail below with reference to FIGS. 2 and 3. Since the second vapor line 13B is identical in structure to the first vapor line 13A, the description hereinafter given of the first vapor line 13A applies as well to the second vapor line 13B.

One end of the first vapor line 13A is connected to the first evaporator 11A. The other end of the first vapor line 13A is connected to the first condenser 12A.

All of the working fluid 16 flowing in the first vapor line 13A is not necessarily in vapor phase. Depending on the operating conditions or installation environment of the loop heat pipe 10, the working fluid 16 may turn into liquid phase during passage between the first evaporator 11A and the first condenser 12A, so that the working fluid 16 partly in liquid phase and partly in vapor phase may flow in the first vapor line 13A.

The first vapor line 13A is formed from a metal having high thermal conductivity such as copper.

Next, the first liquid line 14A will be described in further detail below with reference to FIGS. 2 and 3. Since the second liquid line 14B is identical in structure to the first liquid line 14A, the description hereinafter given of the first liquid line 14A applies as well to the second liquid line 14B.

One end of the first liquid line 14A is connected to the first condenser 12A. The other end of the first liquid line 14A is connected to the second evaporator 11B.

All of the working fluid 16 flowing in the first liquid line 14A is not necessarily in liquid phase. Depending on the operating conditions or installation environment of the loop heat pipe 10, the working fluid 16 may turn into vapor phase during passage between the first condenser 12A and the second evaporator 11B, so that the working fluid 16 partly in liquid phase and partly in vapor phase may flow in the first liquid line 14A.

The first liquid line 14A is formed from a metal having high thermal conductivity such as copper.

The working fluid 16 is sealed in the loop heat pipe 10 preferably in such an amount that the liquid-phase working fluid 16 fills the first evaporator 11A, the second liquid line 14B, the second evaporator 11B, and the first liquid line 14A. Also preferably, the volume of this working fluid 16 is a little larger than one half of the volume of the flow passage in the loop heat pipe 10. If the volume of the working fluid 16 is larger than this specific volume, the flow resistance increases, and the thermal resistance thus increases. On the other hand, if the volume of the working fluid 16 is smaller than this specific volume, the operation of the loop heat pipe 10 may become unstable.

Next, the operation of the loop heat pipe 10 will be described below with reference to FIGS. 6(A) to 6(D). FIGS. 6(A) to 6(D) are diagrams illustrating the operation of the loop heat pipe.

First, as illustrated in FIG. 6(A), in the loop heat pipe 10, the first evaporator 11A is located upwardly of the second evaporator 11B, as viewed along the plumbline direction. Accordingly, in the pre-startup stage, the liquid-phase working fluid 16 is collected in the lower part of the loop heat pipe 10, and the inside of the second evaporator 11B is filled with the liquid-phase working fluid 16. The fine pores of the wick 33 inside the second evaporator 11B are thus impregnated with the liquid-phase working fluid 16.

The upper part of the loop heat pipe 10 is filled with the vapor-phase working fluid 16. Accordingly, the inside of the first evaporator 11A is filled with the vapor-phase working fluid 16. That is, the wick 33 inside the first evaporator 11A is in a dry condition, and the first evaporator 11A is thus in the so-called dry-out condition.

When starting up the loop heat pipe 10, first the second evaporator 11B is started to receive heat. In the example illustrated in FIG. 3, only the CPU 21B is put into operation, and the second evaporator 11B begins to receives heat from the CPU 21B which is the heat source.

The first evaporator 11A is started to receive heat after a predetermined length of time has elapsed from the time the second evaporator 11B began to receive heat. This predetermined length of time is determined based on the time taken for the liquid-phase working fluid 16 to begin to flow into the first evaporator 11A.

In the second evaporator 11B that received heat from the heat source, first the housing 30 is heated by the heat from the heat source, and the heat thus applied to the housing 30 is transferred to the metal block 31. The heat transferred to the metal block 31 is then transferred to the metal tube 32, and the heat transferred to the metal tube 32 is further transferred via the projections 34a of the metal tube 32 to the wick 33 which is thus heated.

When the temperature of the wick 33 rises as it is heated, the liquid-phase working fluid 16 filled into the fine pores of the wick 33 boils and vaporizes. Since the pressure inside the fine pores increases as the working fluid 16 in the fine pores of the wick 33 turns into vapor phase, the vapor-phase working fluid 16 is forced out onto the outer surface of the wick 33.

The vapor-phase working fluid 16 forced out onto the outer surface of the wick 33 passes, for example, through the space formed by the depressions 34b of the metal tube 32 and flows into the interior space of the housing 30 on the second evaporator 13B side. The vapor-phase working fluid 16 then flows into the second vapor line 13B.

In the operating condition after the startup of the loop heat pipe 10, some of the vapor-phase working fluid 16 may remain inside the metal tube 32 of the second evaporator 11B. This vapor-phase working fluid 16 is also forced out onto the outer surface of the wick 33 as the pressure increases due to the vaporization of the working fluid 16.

Next, as illustrated in FIG. 6(B), as the pressure rises inside the housing 30 of the second evaporator 11B, the liquid-phase working fluid 16 remaining in the second evaporator line 13B is forced into the second condenser 12B. The liquid-phase working fluid 16 is further forced from the second condenser 12B into the second liquid line 14B, and the fluid level in the second liquid line 14B thus rises.

Then, the vapor-phase working fluid 16 pushed by the liquid-phase working fluid 16 is forced to pass through the first evaporator 11A and then through the first vapor line 13A, and finally flows into the first condenser 12A. The vapor-phase working fluid 16 flowing into the first condenser 12A is condensed by giving off heat and changes to the liquid phase. The heat contained in the working fluid 16 is transferred via the first condenser line 40A to the first heat sinking plates 41A from which the heat is dissipated.

In this way, in the first condenser 12A, the vapor-phase working fluid 16 is cooled, and all or part of it changes to the liquid phase. As a result, the liquid-phase working fluid 16 accumulates in the first condenser 12A and the first vapor line 13A, and the fluid level thus rises.

Next, as illustrated in FIG. 6(C), the liquid-phase working fluid 16 forced into the second liquid line 14B by being pushed from the second condenser 12B side begins to flow into the first evaporator 11A.

At this point in time, the first evaporator 11A starts to receive heat. For example, as illustrated in FIG. 3, the CPU 21A is put into operation, and the first evaporator 11A begins to receive heat from the CPU 21A which is the heat source.

The liquid-phase working fluid 16 flowing into the first evaporator 11A changes to the vapor phase, and the vapor-phase working fluid 16 flows into the first vapor line 13A.

Next, as illustrated in FIG. 6(D), the interior space of the wick 33 in the first evaporator 11A is substantially filled with the liquid-phase working fluid 16, and the operation of the loop heat pipe 10 thus becomes stable. When the operation of the loop heat pipe 10 becomes stable, the first evaporator 11A, the second evaporator 11B, the first liquid line 14A, and the second liquid line 14B are substantially filled with the liquid-phase working fluid 16. The other portions of the loop heat pipe 10 are filled with the vapor-phase working fluid 16.

In this way, the two heat sources are cooled stably by the loop heat pipe 10.

According to the loop heat pipe 10 described above, since the loop heat pipe is constructed from a single loop flow passage, the overall dimensions of the structure can be reduced. Furthermore, because of the provision of two evaporators, the loop heat pipe 10 can cool two heat sources.

According to the startup method of the loop heat pipe 10 described above, since the evaporator filled with the liquid-phase working fluid 16 is first started to receive heat, the loop heat pipe 10 can be started up in a reliable manner.

For example, in the case of a blade server equipped with two CPUs, if the two CPUs can both be mounted in the lower part of the substrate (lower as viewed along the plumbline direction), the two evaporators can both be filled with the liquid-phase working fluid at the time of the loop heat pipe startup. However, such an arrangement greatly constrains the blade server construction, and is therefore difficult to implement in practice.

Further, in the case of a blade server equipped with two CPUs, the two CPUs are often mounted in different positions as viewed along the plumbline direction. In this case, if different loop heat pipes are provided for the two respective CPUs, the evaporator thermally coupled to the CPU mounted in the upper position as viewed along the plumbline direction may not be able to be filled with the liquid-phase working fluid at the time of startup. If the loop heat pipe is to be started up with no liquid-phase working fluid in the evaporator, the loop heat pipe will not start up because the evaporator is in the dry-out condition and is therefore unable to cause the liquid-phase working fluid to change to the vapor phase.

By contrast, according to the loop heat pipe 10 described above, since two evaporators are provided within a single loop, the evaporator located in the lower part as viewed along the plumbline direction can be easily filled with the liquid-phase working fluid at the time of startup. Further, according to the loop heat pipe startup method described above, the evaporator located in the lower part as viewed along the plumbline direction is first started to receive heat, and after the liquid-phase working fluid has begun to flow into the evaporator located in the upper part as viewed along the plumbline direction, the evaporator located in the upper part is started to receive heat. In this way, the loop heat pipe 10 can be started up in a reliable manner.

The loop heat pipe 10 described above operates stably when the amount of received heat is equal between the first evaporator 11A and the second evaporator 11B. However, if the amount of received heat is not equal between the first evaporator 11A and the second evaporator 11B, the amount of the working fluid 16 that changes from liquid phase to vapor phase becomes different between the first evaporator 11A and the second evaporator 11B; as a result, the distribution of the working fluid 16 in the flow passage becomes uneven, and the circulation of the working fluid 16 may become unstable or may stop.

An example of this will be described with reference to FIG. 7. FIG. 7 is a diagram illustrating how the amount of received heat becomes unbalanced.

In the loop heat pipe 10 depicted in FIG. 7, the amount of received heat in the first evaporator 11A has increased, and on the other hand, the amount of received heat in the second evaporator 11B has decreased, resulting in a situation where the amount of received heat is unbalanced. In the example of the blade server depicted in FIG. 3, this corresponds to the situation where the usage rate of the CPU 21A has increased and its temperature has risen, while the usage rate of the CPU 21B has decreased and its temperature has decreased.

In the loop heat pipe 10, the vaporization rate of the working fluid 16 is higher in the first evaporator 11A where the amount of received heat has increased than in the second evaporator 11B where the amount of received heat has decreased.

As a result, the amount of the liquid-phase working fluid 16 in the second liquid line 14B decreases, while the amount of the liquid-phase working fluid 16 in the first liquid line 14A increases. FIG. 7 shows the condition in which the fluid level of the liquid-phase working fluid 16 has risen into the first vapor line 13A.

If this condition further continues, the first evaporator 11A eventually runs out of the liquid-phase working fluid 16 and is forced into the dry-out condition, and the circulation of the working fluid 16 stops.

Such a phenomenon tends to occur, in particular, when the flow resistance of the working fluid 16 is relatively large, for example, when the distance between the evaporator and the condenser is large or when the evaporator is located in a position lower than the condenser.

It is therefore preferable to design the loop heat pipe so that it can operate stably even when the amount of received heat becomes unbalanced between the two evaporators.

In view of the above, loop heat pipes according to second to fourth embodiments will be described below with reference to drawings as examples of the loop heat pipe that can operate stably even when the amount of received heat becomes unbalanced between the two evaporators. The detailed description of the first embodiment given above essentially applies to those parts of the second to fourth embodiments that are not specifically described herein. Further, in FIGS. 8 to 11, the same component elements as those in FIGS. 2 to 7 are designated by the same reference numerals.

FIG. 8 is a diagram illustrating the loop heat pipe 50 of the second embodiment disclosed in this specification.

The loop heat pipe 50 includes a bypass line 15 which connects between the first vapor line 13A and the second vapor line 13B. The bypass line 15 has the function of diverting the flow of the working fluid 16 and thereby bringing the loop heat pipe 50 back into the stable operating condition when the distribution of the working fluid 16 in the flow passage has become uneven because, for example, the amount of received heat has become unbalanced between the two evaporators.

It is preferable for the bypass line 15 to be provided so as to connect between a portion of the first vapor line 13A in the vicinity of the first condenser 12A and a portion of the second vapor line 13B in the vicinity of the second condenser 12B. For example, it is preferable for the bypass line 15 to connect between the portion of the first vapor line 13A that is located 1 to 3 cm away from the first condenser 12A and the portion of the second vapor line 13B that is located 1 to 3 cm away from the second condenser 12B.

Preferably, the cross-sectional area of the section of the bypass line 15 through which the working fluid 16 passes is not larger than the cross-sectional area of the section of the first vapor line 13A or the second vapor line 13B through which the working fluid 16 passes. Also preferably, the pressure'loss of the working fluid 16 in the bypass line 15 is larger than that in the liquid line or the vapor line.

The reason is that the flow resistance of the working fluid 16 through the bypass line 15 needs to be increased to prevent the working fluid 16 from easily flowing into the bypass line 15 when the loop heat pipe 50 is operating stably.

Next, a description will be given of the preferred relationship between the cross-sectional area of the fluid flow section of the bypass line 15 and the cross-sectional area of the fluid flow section of the first vapor line 13A or the second vapor line 13B.

That is, the ratio of the cross-sectional area of the fluid flow section of the bypass line 15 to the cross-sectional area of the fluid flow section of the first vapor line 13A or the second vapor line 13B is preferably in the range of 0.1 to 1, and more preferably in the range of 0.4 to 0.6.

The cross-sectional area ratio of 0.1 or larger is preferable from the standpoint of quickly diverting the flow of the working fluid 16 and bringing the loop heat pipe back into the stable operating condition when the distribution of the working fluid 16 in the flow passage has become uneven. If the cross-sectional area ratio is smaller than 0.1, the pressure loss through the bypass line 15 becomes too large, and the flow of the working fluid 16 through the bypass line 15 is impeded.

On the other hand, the cross-sectional area ratio of 1 or smaller is preferable from the standpoint of preventing the working fluid 16 from preferentially flowing into the bypass line 15 when the loop heat pipe 50 is operating stably. Further, when the cross-sectional area ratio is 1 or smaller, the liquid-phase working fluid 16 can be caused to flow into the bypass line 15 by utilizing capillary forces.

The length of the bypass line 15 is suitably chosen according to the configuration of the loop heat pipe 50.

The bypass line 15 may be provided with a loop section, a bent section, etc. in order to increase the pressure loss of the working fluid 16.

The structure of the other portions of the loop heat pipe 50 is the same as that of the foregoing first embodiment.

Next, the operation of the loop heat pipe 50 will be described below with reference to FIGS. 9(A) to 9(D). FIGS. 9(A) to 9(D) are diagrams illustrating the operation of the loop heat pipe 50.

First, in FIG. 9(A), the loop heat pipe 50 is operating in a stable condition. The process from the time the loop heat pipe 50 is started up until it reaches the stable operating condition is the same as that described in the first embodiment.

Next, suppose that the amount of received heat in the first evaporator 11A has increased and the amount of received heat in the second evaporator 11B has decreased, thus putting the loop heat pipe 50 in a situation where the amount of received heat is unbalanced, as illustrated in FIG. 9(B).

In the loop heat pipe 50, the vaporization rate of the working fluid 16 is higher in the first evaporator 11A where the amount of received heat has increased than in the second evaporator 11B where the amount of received heat has decreased.

As a result, the amount of the liquid-phase working fluid 16 in the second liquid line 14B decreases. Here, since the amount of the working fluid 16 in the flow passage is constant, the amount of the liquid-phase working fluid 16 in the first liquid line 14A increases. FIG. 9(B) shows the condition in which the fluid level of the liquid-phase working fluid 16 has risen into the first condenser 12A.

As a result, the pressure of the vapor phase portion of the working fluid 16 in the second liquid line 14B decreases, while the pressure in the first vapor line 13A increases. As the pressure in the second liquid line 14B decreases, the pressure in the second condenser 12B as well as the pressure in the second vapor line 13B decreases.

Thereupon, the vapor-phase working fluid 16 in the first vapor line 13A flows through the bypass line 15 into the second vapor line 13B. The working fluid 16 flowing into the second vapor line 13B enters the second condenser 12B where it is converted to the liquid-phase working fluid 16 which then flows into the second liquid line 14B. If any liquid-phase working fluid 16 exists in the first vapor line 13A, the liquid-phase working fluid 16 may also flow into the bypass line 15.

As a result, the amount of the liquid-phase working fluid 16 in the second liquid line 14B increases, while the amount of the liquid-phase working fluid 16 in the first liquid line 14A decreases. In this way, the distribution of the working fluid 16 in the loop heat pipe 50 is automatically brought back to the condition illustrated in FIG. 9(A). The loop heat pipe 50 is thus restored to the stable operating condition.

However, if the rate of increase in the amount of received heat in the first evaporator 11A and the rate of decrease in the amount of received heat in the second evaporator 11B are large, the amount of received heat becomes further unbalanced, and the distribution of the working fluid 16 in the loop heat pipe 50 changes to the condition illustrated in FIG. 9(C).

FIG. 9(C) illustrates the condition in which the amount of the liquid-phase working fluid 16 in the second liquid line 14B has further decreased and the amount of the liquid-phase working fluid 16 in the first liquid line 14A has further increased. In the condition illustrated in FIG. 9(C), the fluid level of the liquid-phase working fluid 16 has risen into the first vapor line 13A.

When the fluid level of the working fluid 16 thereafter reaches the portion connected to the bypass line 15, the liquid-phase working fluid 16 in the first vapor line 13A flows through the bypass line 15 into the second vapor line 13B due to the pressure difference and capillary forces, as illustrated in FIG. 9(D).

The working fluid 16 flowing into the second vapor line 13B is passed through the second condenser 12B and flows into the second liquid line 14B.

The operation of the loop heat pipe 50 has been described above by taking as an example the case where the amount of received heat in the first evaporator 11A increases and the amount of received heat in the second evaporator 11B decreases. However, when the amount of received heat increases only in the first evaporator 11A and the amount of received heat remains unchanged in the second evaporator 11B, or when the amount of received heat remains unchanged in the first evaporator 11A but the amount of received heat decreases in the second evaporator 11B, the loop heat pipe 50 can also be restored to the stable operating condition.

In this way, when a relative change occurs in the amount of received heat between the first evaporator 11A and the second evaporator 11B, the distribution of the working fluid 16 in the flow passage is brought back to the normal condition, and the loop heat pipe 50 is thus restored to the stable operating condition.

Further, when a relative change occurs in cooling capability between the first condenser 12A and the second condenser 12B, the distribution of the working fluid 16 in the flow passage is also brought back to the normal condition, and the loop heat pipe 50 is thus restored to the stable operating condition.

According to the loop heat pipe 50 described above, when the distribution of the working fluid 16 in the flow passage becomes uneven, the working fluid 16 is caused to flow from the first vapor line 13A to the second vapor line 13B through the bypass line 15, so that the loop heat pipe 50 can be restored to the stable operating condition.

Accordingly, even when the amount of received heat becomes unbalanced between the two evaporators, the loop heat pipe 50 can be made to operate stably.

Furthermore, since any uneven distribution of the working fluid 16 occurring in the flow passage can be resolved without using external energy such as electric power, the loop heat pipe 50 is of an energy saving design.

Next, the loop heat pipe of the third embodiment will be described below with reference to FIG. 10. FIG. 10 is a diagram illustrating the loop heat pipe 60 of the third embodiment disclosed in this specification.

In the loop heat pipe 60, the first evaporator 11A and the second evaporator 11B are differently sized. For example, the second evaporator 11B may be made two times as long as the first evaporator 11A.

The loop heat pipe 60 can be used to cool two heat sources having different heat loads. The loop heat pipe 60 can also be used to cool two heat sources having different sizes.

For example, the loop heat pipe 60 can be used to cool a CPU and a chip controller mounted in a server.

Generally, the CPU has a larger size and a larger heat load that the chip controller.

The metal block 31 in the first evaporator 11A may be chosen to have dimensions measuring 30 mm vertically, 30 mm horizontally, and 20 mm in height, compared with the chip controller as one heat source which measures, for example, 20 mm vertically and 20 mm horizontally. On the other hand, the metal block 31 in the second evaporator 11B may be chosen to have dimensions measuring 50 mm vertically, 50 mm horizontally, and 20 mm in height, compared with the CPU as the other heat source which measures, for example, 30 mm vertically and 30 mm horizontally.

The structure of the other portions of the loop heat pipe 60 is the same as that of the foregoing second embodiment.

According to the loop heat pipe 60 described above, the heat sources can be efficiently cooled by using the evaporators each designed to match the size and heat load of the heat source to be cooled.

Next, the loop heat pipe of the fourth embodiment will be described below with reference to FIG. 11. FIG. 11 is a diagram illustrating a blade server 80 in which the loop heat pipe 70 of the fourth embodiment disclosed in this specification is incorporated.

In the loop heat pipe 70, the first and second condensers are constructed in integral fashion, as illustrated in FIG. 11.

More specifically, a plurality of heat sinking plates 41 are coupled in common to both the first condenser line 40A in the first condenser and the second condenser line 40B in the second condenser.

The structure of the other portions of the loop heat pipe 70 is the same as that of the earlier described second embodiment.

According to the loop heat pipe 70 described above, the overall dimensions can be further reduced because the first and second condensers are constructed in integral fashion.

In the present embodiment, the loop heat pipe of each of the above embodiments and its startup method can be modified in various ways without departing from the spirit and purpose of the present invention.

For example, while the first evaporator 11A has been described in each of the above embodiments as being disposed upwardly of the second evaporator 11B as viewed along the plumbline direction, the second evaporator 11B may be disposed upwardly of the first evaporator 11A as viewed along the plumbline direction.

In this case, if it is not possible to identify, at the time of manufacture of the loop heat pipe, the plumbline direction by reference to which the loop heat pipe 10 is to be oriented for use, a component element such as described below may be provided. 1. An acceleration sensor is attached to the loop heat pipe 10 so that the plumbline direction can be identified. 2. Upon startup of the loop heat pipe 10, the temperatures of the two heat sources such as CPUs are monitored, and the heat source whose temperature rise is larger is identified. Then, power supply to the heat source whose temperature rise is larger is slowed down for a prescribed period of time, to provide a startup time difference between the two heat sources. By providing such a component element, the evaporator located in the lower part as viewed along the plumbline direction can be started to receive heat earlier than the other.

In each of the above embodiments, the first vapor line 13A, the second vapor line 13B, the first liquid line 14A, and the second liquid line 14B have been described as having the same diameter, but each line may have a different diameter.

Each of the above embodiments has been described based on the schematically illustrated loop heat pipe, and it will be recognized that each component element is not limited in its structure, arrangement, configuration, etc. to the specific example illustrated herein. For example, the arrangement of the evaporators or condensers and the configuration (layout) of the vapor lines and liquid lines connecting between them can be modified as desired according to the internal configuration of the electronic device in which the loop heat pipe is to be incorporated. Further, other component elements such as startup radiating fins or a startup fan may be provided as desired according to the arrangement and configuration of the evaporators, condensers, vapor lines, or liquid lines.

Next, the operation and effect of the loop heat pipe disclosed in this specification will be further described below with reference to working examples. However, the present invention is not limited by the working examples described herein.

WORKING EXAMPLES
Working Example 1

First, a loop heat pipe of the structure depicted in FIG. 8 was fabricated. Then, the loop heat pipe was assembled into a blade server such as depicted in FIG. 3. The blade server was set with its substrate surface oriented perpendicularly to the plumbline direction so that the two evaporators were arranged in a horizontal plane. The heat load of the CPU A whose heat was to be transferred to the first evaporator was 0 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 60 W; the loop heat pipe thus fabricated was taken as Working Example 1.

Working Example 2

Working Example 2 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 20 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 60 W.

Working Example 3

Working Example 3 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 40 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 60 W.

Working Example 4

Working Example 4 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 60 W.

Working Example 5

Working Example 5 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 40 W.

Working Example 6

Working Example 6 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 20 W.

Working Example 7

Working Example 7 was fabricated in the same manner as Working Example 1, except that the heat load of the CPU A whose heat was to be transferred to the first evaporator was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator was 0 W.

Working Example 8

The blade server was set with its substrate surface oriented in parallel to the plumbline direction so that the two evaporators were arranged in a vertical plane. Further, the heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 0 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 60 W. Otherwise, Working Example 8 was fabricated in the same manner as Working Example 1.

Working Example 9

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 20 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 60 W. Otherwise, Working Example 9 was fabricated in the same manner as Working Example 8.

Working Example 10

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 40 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 60 W. Otherwise, Working Example 10 was fabricated in the same manner as Working Example 8.

Working Example 11

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 60 W. Otherwise, Working Example 11 was fabricated in the same manner as Working Example 8.

Working Example 12

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 40 W. Otherwise, Working Example 12 was fabricated in the same manner as Working Example 8.

Working Example 13

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 20 W. Otherwise, Working Example 13 was fabricated in the same manner as Working Example 8.

Working Example 14

The heat load of the CPU A whose heat was to be transferred to the first evaporator located in the upper part as viewed along the plumbline direction was 60 W, and the heat load of the CPU B whose heat was to be transferred to the second evaporator located in the lower part as viewed along the plumbline direction was 0 W. Otherwise, Working Example 14 was fabricated in the same manner as Working Example 8.

Working Examples 1 to 14 were operated as described below, and the temperatures of the CPUs A and B were measured.

First, power was turned off to the blade server, and the blade server, including the CPUs and the loop heat pipe, was allowed to cool down for a sufficient period of time until the entire structure reached room temperature. Next, power was turned on to the blade server, and the temperatures that the CPUs A and B finally reached were measured.

The results of the measurements are given in FIG. 12.

In Working Example 14, the temperature of the CPU A reached 80° about one minute after power was turned on to the blade server. It was found that when the evaporators were vertically arranged, the loop heat pipe did not start up if the second evaporator located in the lower part as viewed along the plumbline direction did not receive heat.

On the other hand, in Working Examples 1 to 13, the final temperatures of the CPUs A and B were both lower than 60° C. That is, it was found that even when the two evaporators were vertically arranged, if the second evaporator located in the lower part as viewed along the plumbline direction received heat, the loop heat pipe started up to cool the CPUs A and B. It was also found that when the two evaporators were horizontally arranged, the loop heat pipe started up to cool the CPUs A and B, even if one or the other of the evaporators did not receive heat.

Further, using the loop heat pipe of the structure depicted in FIG. 11, measurements similar to those of Working Examples 1 to 14 were made; in this case also, similar results were obtained.

Next, using a two-phase fluid simulator, computer experiments were conducted to simulate the operation of the loop heat pipes having the structures depicted in FIGS. 2 and 8, respectively, and Working Examples 15 to 18 were obtained. SINDA/FLUINT (available from C&R TECHNOLOGIES) was used as the two-phase fluid simulator.

Working Example 15

The loop heat pipe of the structure depicted in FIG. 2 was used. A non-CFC refrigerant R141b, or more specifically, HCFC (hydrochlorofluorocarbon), was used as the refrigerant. The first and second liquid lines and the first and second vapor lines were each 4.5 mm in inner diameter. The first and second liquid lines were each 1.0 m in length. The first and second vapor lines were each 1.0 m in length. The first and second condensers were each 1.0 m in length. The first and second evaporators were each heated by a heat source having an output of 150 W. The first and second evaporators were arranged horizontally relative to the plumbline direction. The loop heat pipe had a first set formed by the second condenser, the second liquid line, the first evaporator, and the first vapor line, and a second set formed by the first condenser, the first liquid line, the second evaporator, and the second vapor line. In each set, the liquid line was divided into eight grids, the evaporator was divided into two grids, the vapor line was divided into 12 grids, and the condenser was divided into eight grids. Then, in the steady-state condition of the loop heat pipe, the proportion of the vapor phase of the working fluid was calculated for each grid in each set. The results of the calculations are given in FIG. 13.

Working Example 16

Working Example 16 was obtained in the same manner as Working Example 15, except that the loop heat pipe of the structure depicted in FIG. 8 was used. The inner diameter of the bypass line was 4.5 mm, and the length of the bypass line was 1.4 mm. The results of the calculations are given in FIG. 14.

Working Example 17

Working Example 17 was obtained in the same manner as Working Example 15, except that the first evaporator in the first set was heated by a heat source having an output of 50 W and the second evaporator in the second set was heated by a heat source having an output of 150 W. The results of the calculations are given in FIG. 15.

Working Example 18

Working Example 18 was obtained in the same manner as Working Example 16, except that the loop heat pipe of the structure depicted in FIG. 8 was used and that the first evaporator in the first set was heated by a heat source having an output of 50 W and the second evaporator in the second set was heated by a heat source having an output of 150 W. The results of the calculations are given in FIG. 16.

As can be seen from FIGS. 13 and 14, in Working Examples 15 and 16 in which the two evaporators were equally heated, all of the working fluid in each vapor line was in vapor phase, and all of the working fluid in each liquid line was in liquid phase.

As can be seen from FIG. 15, in Working Example 17, the circulation of the working fluid did occur in the loop heat pipe, but in the first vapor line in the first set, about 40% of the working fluid was in vapor phase and about 60% was in liquid phase. On the other hand, in the second set, all of the working fluid in the second vapor line was in vapor phase.

As can be seen from FIG. 16, in Working Example 18, all of the working fluid in each vapor line was in vapor phase, and all of the working fluid in each liquid line was in liquid phase. It was thus found that when the bypass line was added in the loop heat pipe structure of Working Example 17, all of the working fluid flowing in the vapor line of the first set also was maintained in vapor phase.

All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the invention and the concepts contributed by the inventor to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions, nor does the organization of such examples in the specification relate to a showing of superiority and inferiority of the invention. Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

	Number	Date	Country
Parent	PCT/JP2010/056093	Apr 2010	US
Child	13323973		US

LOOP HEAT PIPE AND STARTUP METHOD FOR THE SAME

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

US Classifications

International Classifications

Abstract

Description

Claims

Priority Claims (1)

CROSS-REFERENCE TO RELATED APPLICATIONS

Continuations (1)