Fundamentals of Heat

Pipes
With Applications to Electronics
Cooling
-- Widah Saied

Introduction
Things to be discussed:
 Basic components
 Advantages
 Ideal thermodynamic cycle
 Applications
 Types
 Heat transfer limitations
 Resistance network
 Wick design
 Choosing the working fluid
 Container design
 Heat pipes in electronics cooling
 Current research in electronics cooling


Basic Components


condenser
evaporator
wick
http://www.lightstreamphotonics.com/images/tech_orangecontainer_small.png
Adiabatic section
Advantages of Heat Pipes
 Very high thermal conductivity. Less
temperature difference needed to transport
heat than traditional materials (thermal
conductivity up to 90 times greater than
copper for the same size) (Faghiri, 1995) resulting, in
low thermal resistance. (Peterson,1994)

 Power flattening. A constant condenser heat
flux can be maintained while the evaporator
experiences variable heat fluxes. (Faghiri, 1995)


 Efficient transport of concentrated heat. (Faghiri, 1995)


Advantages of Heat Pipes
 Temperature Control. The evaporator and
condenser temperature can remain nearly
constant (at T
sat
) while heat flux into the
evaporator may vary (Faghiri, 1995) .

 Geometry control. The condenser and
evaporator can have different areas to fit
variable area spaces (Faghiri, 1995) . High heat flux
inputs can be dissipated with low heat flux
outputs only using natural or forced convection
(Peterson,1994) .
Thermodynamic Cycle
 1-2 Heat applied to evaporator through external
sources vaporizes working fluid to a saturated(2‟) or
superheated (2) vapor.
 2-3 Vapor pressure drives vapor through adiabatic
section to condenser.
 3-4 Vapor condenses, releasing heat to a heat sink.
 4-1 Capillary pressure created by menisci in wick
pumps condensed fluid into evaporator section.
 Process starts over.

(Faghiri, 1995)
Ideal Thermodynamic Cycle
(Faghiri, 1995)
Heat Pipe Applications
 Electronics cooling- small high performance components
cause high heat fluxes and high heat dissipation demands.
Used to cool transistors and high density semiconductors.

 Aerospace- cool satellite solar array, as well as shuttle
leading edge during reentry.

 Heat exchangers- power industries use heat pipe heat
exchangers as air heaters on boilers.

 Other applications- production tools, medicine and human
body temperature control, engines and automotive industry.

(Faghiri, 1995)
Types of Heat Pipes
 Thermosyphon- gravity assisted wickless heat pipe. Gravity is
used to force the condensate back into the evaporator. Therefore,
condenser must be above the evaporator in a gravity field.

 Leading edge- placed in the leading edge of hypersonic vehicles
to cool high heat fluxes near the wing leading edge. (Faghiri, 1995)


 Rotating and revolving- condensate returned to the evaporator
through centrifugal force. No capillary wicks required. Used to
cool turbine components and armatures for electric motors.

 Cryogenic- low temperature heat pipe. Used to cool optical
instruments in space. (Peterson, 1994)





Types of Heat Pipes
 Flat Plate- much like traditional cylindrical heat pipes but
are rectangular. Used to cool and flatten temperatures of
semiconductor or transistor packages assembled in arrays
on the top of the heat pipe.








(Faghiri,1995)
Types of Heat Pipes
 Micro heat pipes- small heat pipes that are noncircular and use angled
corners as liquid arteries. Characterized by the equation: rc /r
h
>1 where rc
is the capillary radius, and r
h
is
the hydraulic radius of the flow
channel. Employed in cooling
semiconductors (improve
thermal control), laser diodes,
photovoltaic cells, medical
devices.





(Peterson,1994)

Types of Heat Pipes
 Variable conductance- allows variable heat fluxes into the evaporator
while evaporator temperature remains constant by pushing a non-
condensable gas into the condenser when heat fluxes are low and
moving the gas out of the condenser when heat fluxes are high, thereby,
increasing condenser surface area. They come in various forms like
excess-liquid or gas-loaded form. The gas-loaded form is shown below.
Used in electronics cooling. (Faghiri,1995)


Types of Heat Pipes
 Capillary pumped loop heat pipe- for systems where the heat fluxes are
very high or where the heat from the heat source needs to be moved far
away. In the loop heat pipe, the vapor travels around in a loop where it
condenses and returns to the evaporator. Used in electronics cooling.








(Faghiri, 1995)
Main Heat Transfer Limitations
 Capillary limit- occurs when the capillary pressure is
too low to provide enough liquid to the evaporator
from the condenser. Leads to dryout in the
evaporator. Dryout prevents the thermodynamic
cycle from continuing and the heat pipe no longer
functions properly.

 Boiling Limit- occurs when the radial heat flux into
the heat pipe causes the liquid in the wick to boil
and evaporate causing dryout.


(Faghiri, 1995)


Heat Transfer Limitations
 Entrainment Limit- at high vapor velocities, droplets of liquid in the
wick are torn from the wick and sent into the vapor. Results in
dryout.

 Sonic limit- occurs when the vapor velocity reaches sonic speed
at the evaporator and any increase in pressure difference will not
speed up the flow; like choked flow in converging-diverging
nozzle. Usually occurs during startup of heat pipe.

 Viscous Limit- at low temperatures the vapor pressure difference
between the condenser and the evaporator may not be enough to
overcome viscous forces. The vapor from the evaporator doesn‟t
move to the condenser and the thermodynamic cycle doesn‟t
occur.

(Faghiri, 1995)




Heat Transfer Limitations
 Each limit has its own particular range in which it is important. However,
in practical operation, the capillary and boiling limits are the most
important. The figure below is an example of these ranges.
(Peterson,1994)
Heat Transfer Limitations
 Actual performance curves, capillary limit and boiling limit, are the
limiting factors.
http://www.electronics-cooling.com/Resources/EC_Articles/SEP96/sep96_02.htm
Capillary Limit
 For a heat pipe to function properly, the capillary pressure must
be greater or equal to the sum of the pressure drops due to
inertial, viscous, and hydrostatic forces, as well as, pressure
gradients.


 If it is not, then the working fluid is not supplied rapidly enough to
the evaporator to compensate for the liquid loss through
vaporization. If this occurs, there is dryout in the evaporator.


(Peterson, 1994)
Capillary Limit
 Equation for minimum capillary pressure:







(Peterson, 1994)
Boiling Limit
 The Boiling limit is due to excessive radial heat flux; all the other
limits are due to axial heat flux.
 The maximum heat flux beyond which bubble growth will occur
resulting in dryout is given by:




(Peterson, 1994)
Boiling Limit
 K
eff
given by the table below:

























 Kl=therm. cond. Liquid, kw=therm. cond. wick
 o =thickness of tube, c =wick porosity


Resistance Network
(Peterson, 1994)
Heat Pipe Resistance
 In certain applications the temperature difference between the evaporator and the
condenser needs to be known, such as in electronics cooling. This may be done
using a thermal circuit.

 The main resistances within the heat pipe are:

Resistance Order of Magnitude
Rw,a liquid-wick resistance in the adiabatic section 10
4
Rp,a axial resistance of the pipe wall 10
2
Rw,e liquid-wick resistance in the evaporator 10
1
Rw,c liquid-wick resistance in the condenser 10
1
Rp,e radial resistance of the pipe wall at the evaporator 10
-1
Rp,c radial resistance of the pipe wall at the condenser 10
-1


Other resistances exist but most are small relative to the above
resistances.

The external resistances – the resistances transferring the heat to and
from the heat pipe – are also important in some cases.



(Peterson, 1994)
Heat Pipe Resistance
 The liquid-wick combination for the three heat pipe
sections are given by:



K
eff
given on a previous slide

 The radial and axial resistances can be determined
from traditional resistance equations for cylindrical
shapes and flat plates depending on the shape of the
heat pipe.
eff
i o
w
K L
d d
R
_
2
) / ln(
_
t
=
(Peterson, 1994)
The Wick and its Design
 Main Purpose- provides structure and force that
transports the condensate liquid back to the
evaporator. Also, ensures working fluid is evenly
distributed over evaporator surface.


(Peterson, 1994)
Capillary Pressure
 The driving force that transports the condensed
working liquid through the wick to the evaporator
is provided by capillary pressure. Working fluids
that are employed in heat pipes have concave
facing menisci (wetting liquids) as opposed to
convex facing menisci (non wetting liquids).

 Contact angle is defined as the angle between
the solid and vapor regions. Wetting fluids have
angles between 0 and 90 degrees. Non wetting
fluids have angles between 90 and 180 degrees.

(Faghiri, 1995)



Capillary Pressure
http://cwx.prenhall.com/petrucci/medialib/media_portfolio/text_images/FG13_04.JPG
Water
Wetting liquid
Mercury
Non wetting liquid
θ
θ
 Wetting angle
Capillary Pressure
 The shape of a fluid‟s meniscus is dependent on the fluid‟s
surface tension and the solid-fluid adhesion force. If the adhesion
force is greater than the surface tension, the liquid near the solid
will be forced up and the surface tension of the liquid will keep the
surface intact causing the entire liquid to move up.
http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html


 When the liquid in the evaporator vaporizes, the radius of
curvature of the menisci in the wick decreases. As the vapor
condenses in the condenser, the radius of curvature of the
menisci in the wick increases. The difference in the radius of
curvature results in capillary pressure (Peterson,1994) . Capillary
pressure is also due to body forces and phase-change
interactions (Faghiri, 1995).



Capillary Pressure
 The capillary pressure created by two menisci of
different radii of curvature is given by





 Where R
I
and R
II
are radii of curvature and σ
is the surface tension.
 Called the Young-Laplace Equation


(Peterson,1994)

|
|
.
|

\
|
+ =
I I I
R R
P
cap
1 1
o
Capillary Pressure
 To maximize capillary pressure, the minimum radii is needed.
For a circular capillary the minimum radii is :


 Substituting these values into the formula for capillary pressure:



 For max capillary pressure theta must be zero


u cos
) R , (R
min II I
r
=
r
P
cap
o 2
max ,
=
r
P
cap
u o cos 2
=
(Peterson,1994)
Capillary Pressure
 Wetting fluids have a cosθ value that will be positive.
This results in a positive capillary pressure that
creates a pushing force on the liquid in the wick near
the condenser; this forces the liquid to move to the
evaporator.

 Non-wetting fluids will have cosθ values that are
negative, resulting in a negative capillary pressure that
creates a suction force on the liquid in the wick. The
liquid is prevented from moving to the evaporator.

 For this reason, the working liquid in heat pipes must
be a wetting liquid.

(Peterson,1994)
Wick Design
 Two main types of wicks: homogeneous and
composite.

 Homogeneous- made from one type of material or
machining technique. Tend to have either high
capillary pressure and low permeability or the other
way around. Simple to design, manufacture, and
install (Faghiri, 1995) .

 Composite- made of a combination of several types or
porosities of materials and/or configurations. Capillary
pumping and axial fluid transport are handled
independently (Peterson,1994) . Tend to have a higher capillary
limit than homogeneous wicks but cost more (Faghiri, 1995).

Wick Design
 Three properties effect wick design:

 1. High pumping pressure- a small capillary pore radius
(channels through which the liquid travels in the wick)
results in a large pumping (capillary) pressure.
 2. Permeability - large pore radius results in low liquid
pressure drops and low flow resistance.
 Design choice should be made that balances large
capillary pressure with low liquid pressure drop.
Composite wicks tend to find a compromise between
the two.
 3.Thermal conductivity - a large value will result in a
small temperature difference for high heat fluxes.
(Peterson,1994)
Wick Design
(Peterson,1994).
http://www.electronics-cooling.com/Resources/EC_Articles/SEP96/sep96_02.htm
Choosing the Working Fluid
 Heat pipes work on a cycle of vaporization and condensation of
the working fluid, which results in the heat pipe‟s high thermal
conductivity. When choosing a working fluid for a heat pipe, the
fluid must be able to operate within the heat pipe‟s operating
temperature range. For instance, if the operating temperatures
are too high, the fluid may not be able to condense. However, if
the operating temperatures are too low the fluid will not be able to
evaporate. Watch the saturation temperature for your desired
fluid at the desired heat pipe internal pressure.

 In addition, the working fluid must be compatible with the wick
and container material.

(Peterson, 1994).
Choosing the Working Fluid
Operating temperature ranges for various working fluids:

http://www.cheresources.com/htpipes.shtml
Choosing the Working Fluid
 Generally, as the operating temperature
range of the working fluid increases, the
heat transport capability increases.

 Choice of working fluid should also
incorporate the fluid‟s interactions with
the heat pipe container and wick.
(Peterson, 1994).
Choosing the Working Fluid
Chi(1976) developed a parameter of gauging
the effectiveness of a working fluid called the
liquid transport factor:


Where ì is the latent heat of vaporization and o is the surface
tension. Subscript refers to the liquid

For electronics cooling applications, occurring
in low to moderate temperatures, water is the
liquid with the highest liquid transport factor.
Another common fluid is ammonia.

l
l
l
N
µ
oì µ
=
l
(Peterson, 1994).
Container Design
 Things that should be considered for
container design:
 Operating temperature range of the heat pipe.
 Internal operating pressure and container structural
integrity.
 Evaporator and condenser size and shape.
 Possibility of external corrosion.
 Prevent leaks.
 Compatibility with wick and working fluid.

(peterson,1994)

Container Design
 Stresses:
 Since the heat pipe is like a pressure vessel
it must satisfy ASME pressure vessel
codes.

 Typically the maximum allowable stress at
any given temperature can only be one-
fourth of the material‟s maximum tensile
strength.

(peterson, 1994)

Container Design
 Typical materials:
 Aluminum
 Stainless steel
 Copper
 Composite materials
 High temperature heat pipes may use
refractory materials or linings to prevent
corrosion.

(Peterson, 1994)
Heat pipe Compatibility
 When designing a heat pipe, the working fluid, wick, and container must
function properly when operating together. For example, the working fluid
may not be wettable with the wick; or the fluid and container may
undergo a chemical reaction with each other.
(Peterson, 1994)
Heat pipe Compatibility
 Working fluid/
material
compatibility.
(Faghiri, 1995)

Heat Sink/Source Interface
 The contact resistance between the
evaporator and the heat source and between
the condenser and the heat sink is relatively
large and should be minimized.

 Methods used to join the parts include use of
thermally conductive adhesives, as well as,
brazed, or soldered techniques.
(Peterson, 1994)

Heat Pipes in Electronics Cooling
 Cooling of electronics has one primary goal: maintain
a component‟s temperatures at or below the
manufacturer‟s maximum allowable temperature. As
the temperature of an electronic part increases the
rate of failure increases.





(Peterson, 1994)
Heat Pipes in Electronics Cooling
 Heat pipes are excellent candidates for electronics
cooling because of their high thermal conductivity,
high heat transfer characteristics, they provide
constant evaporator temperatures with variable heat
fluxes, and variable evaporator and condenser sizes.

 Therefore, they are good alternatives to large heat
sinks, especially in laptops where space is limited.

 They are good alternative to air cooling because of
their better heat transport capabilities. Air cooling may
still be used to remove heat from the condenser.


(Peterson, 1994).

Heat Pipes in Electronics Cooling
 Common heat pipes used in electronics
cooling:
 Micro heat pipes
 Capillary looped heat pipes
 Flat plate heat pipes
 Variable conductance heat pipes

Heat Pipes in Electronics Cooling
 In single component cooling, the heat pipe‟s
evaporator may be attached to an individual heat
source (power transistor, thyristor, or chip).

 The condenser is attached to a heat sink to dissipate
the heat through free or forced convection.

(Peterson, 1994)

Heat Pipes in Electronics Cooling
 Cooling can also occur with multiple arrays of devices
or entire printed wiring boards.

(Peterson, 1994)

Heat Pipes in Electronics Cooling
 An arrayed heat pipe cooling system

(Peterson, 1994)
Heat Pipes in Electronics Cooling
 Heat pipe cooling a component set up in an array
(Peterson, 1994)

Heat Pipes in Electronics Cooling
 Since many semiconductors are small,
micro heat pipes may be used for cooling
individual semiconductors or an array.
Good for applications
where space is limited
like laptops.



(Peterson, 1994)

Heat Pipes in Electronics Cooling
 When the electrical power is high and
the heat rejection requirements large
and nucleate pool boiling occurs,
another method of cooling a heat source
may be employed.
 Nucleate pool boiling causes a large
temperature drop. To reduce the drop,
you can make the device a part of the
wick structure to ensure that fresh liquid
is always in contact with the heat
source. Further providing cooling to the
transistor.
 In the image to the right the heat source (a transistor
chip) is in contact with the working liquid and the
working liquid is being evaporated away, cooling the
transistor.

(Peterson, 1994)

Heat Pipes in Electronics Cooling
 Summary:
 Heat pipes enable devices with higher
density heat dissipation requirements and
greater reliability.
 Low cost
 Proven alternative to conventional methods
of electronics cooling.



(Peterson, 1994)


Current Research in Electronics Cooling
 Laptops today perform well and are small;
therefore, they have high heat dissipation
demands.

 Excess heat may slow down the processor‟s
speed or shut the laptop off. (Junnarkar, 2003)

 First time a heat pipe used in a laptop was in
1994. Current heat pipes move the heat from
the CPU to a small heat sink. (Ali et al., 1999)
Current Research in Electronics Cooling
 Because micro heat pipes are small they are
very useful in cooling of laptops where space
is highly restricted.

 Wang and Peterson (2003) have come up with
two different micro heat pipe setups for laptop
cooling:
 Micro heat pipes configured into flat plate shapes were
employed to cool a CPU. The condenser was attached to a
heat sink. The heat sink was smaller in size than one not
attached to a heat pipe because the base of the heat sink
attached to a heat pipe experiences more uniform
temperatures and therefore, an increased efficiency.
Current Research in Electronics Cooling
 Two different configurations were developed
 Both were 152.4 mm long and 25.4 mm wide
 Layers of copper screen mesh, with parallel wires and
two copper sheets were formed, in the shape of a flat
heat pipe, to form an enclosed space.
 No capillary wick structure needed because of the
micro heat pipe‟s sharp corners.
 The fan is strategically placed to provide forced
convection to the heat sink.


Current Research in Electronics Cooling
Current Research in Electronics Cooling
Main Results:
 In configuration 1, tilt angle effected the
amount of heat dissipated
 In configuration 2, tilt angle had no effect
on amount dissipated.
 Important because laptops experience
operation in many orientations.
Current Research in Electronics Cooling
Mesh number is defined as the number of openings per linear inch. (About,2006)
Current Research in Electronics Cooling
 Things that increased heat transport
capacity:
 Increasing mesh number
 Increasing wire diameter
Current Research in Electronics Cooling
 The thermal resistance from the heat sink to the device
junction, due to the cooling of the heat pipe with forced
convection, is greater for case 1 than case 2 at all air
velocities. The values were determined from the relation:



Where Q
c
is the heat dissipated through the heat sink
total c
a j
ja
R Q
T T
1
=
÷
= u
Current Research in Electronics Cooling
Current Research in Electronics Cooling
 Other discoveries:
 Within the CPU‟s operating temperature
limit, the heat capacity of a micro heat pipe
is restricted by the heat sink‟s ability to
transfer heat through convection
 Heat transfer not restricted by the capillary
limit.

Current Research in Electronics Cooling
The maximum heat transfer limit provided by the heat pipe, for the most part,
is not reached due to deficiencies in the heat sink‟s ability to transfer
heat through convection.
Current Research in Electronics Cooling
 Case 2 provided a lower thermal
resistance and a greater heat transport
capacity than Case 1.

 Case 2 transported 52W at 85°C and .85 °C /W resistance.
 Case 1 transported 24W at 85°C and 1.55 °C /W resistance.

References
 About, Inc. (2006). May 2006. http://composite.about.com/library/glossary/m/bldef-m3304.htm
 Ali, A., Dehoff, R., and Grubb, K., 1999. ”Advanced Heat Pipe Thermal Solutions For Higher Power Notebook
Computers”.
 Basilius, A., Tanzer, H., and McCabe, S., 1987 “Heat Pipes for Cooling of High Density Printed Wiring Boards,” Proc. 6
th

Int. Heat Pipe Conf., Grenoble, France, pp.531-536.
 The Chemical Engineer’s Resource Page. 2004. March 2006. http://www.cheresources.com/htpipes.shtml
 Chi, 1976, Heat Pipe Theory and Practice, McGraw-Hill, New York.
 Dunn, P.D., and Reay, D.A., 1982, HJeat Pipes, 3
rd
.ed., Pergamon, Oxford.
 Faghri, Amin, 1995. Heat Pipe Science and Technology. US: Taylor &Francis
 Junnarkar, Sandeep. “Laptops Cool off with „Smart‟ Heat Pipes.” Cnet News.com. January22, 2003. March 2006.
 Lightstream Photonics (2003). March 2006.
http://www.lightstreamphotonics.com/images/tech_orangecontainer_small.png
 Murase, T., Yoshida, K., Koizumi, T., and Ishida, N., 1982, “Heat Pipe Heat Sink „Heat Kicker‟ for Cooling of Semi-
Conductors,” Furukama Review, Tokyo, Japan, Vol.2, pp24-33.
 Nave, R. Hyperphysics -capillary action. 2006. March 2006. http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html
 Nelson, L., Sekhon, K., and Fritz, J. E., 1978, “ Direct Heat Pipe Cooling of Semiconductor Devices,” Proc. 3
rd
Int. Heat
Pipe Conf., Palo Alto, CA, pp373-376.
 Peterson, G.P., 1994. An Introduction to Heat Pipes. Canada: John Wiley & Sons, Inc.
 Petrucci, R.H., Harwood, W.S., Herring, G. General Chemisty Principles and Modern Applications. 2001, Prentice-Hall,
Inc. Compainion Website. Narayan S. Homane. March 2006.
http://cwx.prenhall.com/petrucci/medialib/media_portfolio/text_images/FG13_04.JPG
 Warner, S. (2006) May 2006. http://www.electronics-cooling.com/Resources/EC_Articles/SEP96/sep96_02.htm
 Wang, Y., and Peterson, G., 2003. “ Flat Heat Pipe Cooling Devices for Mobile Computers.” ASME International
Mechanical Engineering Congress, Washington, D.C., Nov. 15-21,2003.
 Zorbil, V., Stulc, P., and Polasek, F., 1988, “ Enhancement Cooling of the Boards with Integrated Circuits by Heat Pipes,
“ Proc. 3
rd
. Int. Heat Pipe Symp., Tsukuba, Japan, pp273-279.


Introduction
Things to be discussed:
           

Basic components Advantages Ideal thermodynamic cycle Applications Types Heat transfer limitations Resistance network Wick design Choosing the working fluid Container design Heat pipes in electronics cooling Current research in electronics cooling

Basic Components
Adiabatic section

evaporator

condenser

wick

http://www.lightstreamphotonics.com/images/tech_orangecontainer_small.png

A constant condenser heat flux can be maintained while the evaporator experiences variable heat fluxes.1994)  Power flattening. Less temperature difference needed to transport heat than traditional materials (thermal conductivity up to 90 times greater than copper for the same size) resulting. (Faghiri. 1995) . 1995) (Peterson. in low thermal resistance. (Faghiri. (Faghiri. 1995)  Efficient transport of concentrated heat.Advantages of Heat Pipes  Very high thermal conductivity.

1995) (Peterson. .1994) . (Faghiri. High heat flux inputs can be dissipated with low heat flux outputs only using natural or forced convection (Faghiri. The condenser and evaporator can have different areas to fit variable area spaces .Advantages of Heat Pipes  Temperature Control. The evaporator and condenser temperature can remain nearly constant (at Tsat) while heat flux into the evaporator may vary . 1995)  Geometry control.

Process starts over. 3-4 Vapor condenses. 2-3 Vapor pressure drives vapor through adiabatic section to condenser. 1995) . (Faghiri. releasing heat to a heat sink.Thermodynamic Cycle      1-2 Heat applied to evaporator through external sources vaporizes working fluid to a saturated(2‟) or superheated (2) vapor. 4-1 Capillary pressure created by menisci in wick pumps condensed fluid into evaporator section.

Ideal Thermodynamic Cycle

(Faghiri, 1995)

Heat Pipe Applications

Electronics cooling- small high performance components cause high heat fluxes and high heat dissipation demands. Used to cool transistors and high density semiconductors. Aerospace- cool satellite solar array, as well as shuttle leading edge during reentry. Heat exchangers- power industries use heat pipe heat exchangers as air heaters on boilers. Other applications- production tools, medicine and human body temperature control, engines and automotive industry.

(Faghiri, 1995)

Types of Heat Pipes

Thermosyphon- gravity assisted wickless heat pipe. Gravity is used to force the condensate back into the evaporator. Therefore, condenser must be above the evaporator in a gravity field. Leading edge- placed in the leading edge of hypersonic vehicles to cool high heat fluxes near the wing leading edge. (Faghiri, 1995)

Rotating and revolving- condensate returned to the evaporator through centrifugal force. No capillary wicks required. Used to cool turbine components and armatures for electric motors.
Cryogenic- low temperature heat pipe. Used to cool optical instruments in space. (Peterson, 1994)

much like traditional cylindrical heat pipes but are rectangular. (Faghiri.Types of Heat Pipes  Flat Plate.1995) . Used to cool and flatten temperatures of semiconductor or transistor packages assembled in arrays on the top of the heat pipe.

and rh is the hydraulic radius of the flow channel. Characterized by the equation: rc /rh1 where rc is the capillary radius. photovoltaic cells. (Peterson.1994) . laser diodes.small heat pipes that are noncircular and use angled corners as liquid arteries. Employed in cooling semiconductors (improve thermal control). medical devices.Types of Heat Pipes  Micro heat pipes.

1995) . Used in electronics cooling. The gas-loaded form is shown below. They come in various forms like excess-liquid or gas-loaded form. increasing condenser surface area.allows variable heat fluxes into the evaporator while evaporator temperature remains constant by pushing a noncondensable gas into the condenser when heat fluxes are low and moving the gas out of the condenser when heat fluxes are high. (Faghiri. thereby.Types of Heat Pipes  Variable conductance.

In the loop heat pipe. Used in electronics cooling. the vapor travels around in a loop where it condenses and returns to the evaporator.Types of Heat Pipes  Capillary pumped loop heat pipe. 1995) .for systems where the heat fluxes are very high or where the heat from the heat source needs to be moved far away. (Faghiri.

occurs when the radial heat flux into the heat pipe causes the liquid in the wick to boil and evaporate causing dryout. Boiling Limit. Dryout prevents the thermodynamic cycle from continuing and the heat pipe no longer functions properly. Leads to dryout in the evaporator.Main Heat Transfer Limitations  Capillary limit. (Faghiri.occurs when the capillary pressure is too low to provide enough liquid to the evaporator from the condenser. 1995)  .

at low temperatures the vapor pressure difference between the condenser and the evaporator may not be enough to overcome viscous forces.occurs when the vapor velocity reaches sonic speed at the evaporator and any increase in pressure difference will not speed up the flow. Results in dryout. 1995)   . (Faghiri. Usually occurs during startup of heat pipe. Viscous Limit.at high vapor velocities. like choked flow in converging-diverging nozzle. The vapor from the evaporator doesn‟t move to the condenser and the thermodynamic cycle doesn‟t occur.Heat Transfer Limitations  Entrainment Limit. droplets of liquid in the wick are torn from the wick and sent into the vapor. Sonic limit.

the capillary and boiling limits are the most important.Heat Transfer Limitations  Each limit has its own particular range in which it is important. in practical operation. However.1994) . (Peterson. The figure below is an example of these ranges.

htm . are the limiting factors.Heat Transfer Limitations  Actual performance curves. capillary limit and boiling limit.com/Resources/EC_Articles/SEP96/sep96_02.electronics-cooling. http://www.

pressure gradients. the capillary pressure must be greater or equal to the sum of the pressure drops due to inertial. and hydrostatic forces. (Peterson.  If it is not. there is dryout in the evaporator. viscous. then the working fluid is not supplied rapidly enough to the evaporator to compensate for the liquid loss through vaporization. 1994) . as well as. If this occurs.Capillary Limit  For a heat pipe to function properly.

1994) .Capillary Limit  Equation for minimum capillary pressure: (Peterson.

1994) . all the other limits are due to axial heat flux.Boiling Limit  The Boiling limit is due to excessive radial heat flux.  The maximum heat flux beyond which bubble growth will occur resulting in dryout is given by: (Peterson.

 =wick porosity . cond. cond. wick  =thickness of tube. kw=therm. Liquid.Boiling Limit  Keff given by the table below:   Kl=therm.

Resistance Network (Peterson. 1994) .

c liquid-wick resistance in the condenser Rp. This may be done using a thermal circuit.Heat Pipe Resistance  In certain applications the temperature difference between the evaporator and the condenser needs to be known.c radial resistance of the pipe wall at the condenser  Resistance 101 101 10-1 10-1 Other resistances exist but most are small relative to the above resistances.a axial resistance of the pipe wall Rw.e radial resistance of the pipe wall at the evaporator Rp. (Peterson. 1994) . The external resistances – the resistances transferring the heat to and from the heat pipe – are also important in some cases.e liquid-wick resistance in the evaporator Rw. such as in electronics cooling.a liquid-wick resistance in the adiabatic section Rp. The main resistances within the heat pipe are: Order of Magnitude 104 102 Rw.

Heat Pipe Resistance  The liquid-wick combination for the three heat pipe sections are given by: Rw _  ln( d o / d i ) 2L_ K eff Keff given on a previous slide  The radial and axial resistances can be determined from traditional resistance equations for cylindrical shapes and flat plates depending on the shape of the heat pipe. (Peterson. 1994) .

ensures working fluid is evenly distributed over evaporator surface.The Wick and its Design  Main Purpose. 1994) .provides structure and force that transports the condensate liquid back to the evaporator. Also. (Peterson.

1995)  . Contact angle is defined as the angle between the solid and vapor regions.Capillary Pressure  The driving force that transports the condensed working liquid through the wick to the evaporator is provided by capillary pressure. Working fluids that are employed in heat pipes have concave facing menisci (wetting liquids) as opposed to convex facing menisci (non wetting liquids). (Faghiri. Non wetting fluids have angles between 90 and 180 degrees. Wetting fluids have angles between 0 and 90 degrees.

com/petrucci/medialib/media_portfolio/text_images/FG13_04.Capillary Pressure  Wetting angle θ θ Water Wetting liquid Mercury Non wetting liquid http://cwx.prenhall.JPG .

html  When the liquid in the evaporator vaporizes. the radius of curvature of the menisci in the wick decreases.gsu. As the vapor condenses in the condenser.edu/hbase/hframe. the liquid near the solid will be forced up and the surface tension of the liquid will keep the surface intact causing the entire liquid to move up. If the adhesion force is greater than the surface tension. the radius of curvature of the menisci in the wick increases. . The difference in the radius of curvature results in capillary pressure (Peterson. Capillary pressure is also due to body forces and phase-change interactions (Faghiri. http://hyperphysics.Capillary Pressure  The shape of a fluid‟s meniscus is dependent on the fluid‟s surface tension and the solid-fluid adhesion force. 1995).phy-astr.1994) .

 Called the Young-Laplace Equation  (Peterson.1994) .Capillary Pressure  The capillary pressure created by two menisci of different radii of curvature is given by 1 1  Pcap     R R       Where RI and RII are radii of curvature and σ is the surface tension.

For a circular capillary the minimum radii is : (R I .max  r (Peterson.1994) .Capillary Pressure  To maximize capillary pressure. the minimum radii is needed. R II ) min   r cos Substituting these values into the formula for capillary pressure: Pcap  2 cos  r For max capillary pressure theta must be zero 2 Pcap.

Capillary Pressure  Wetting fluids have a cosθ value that will be positive. the working liquid in heat pipes must be a wetting liquid. For this reason.  Non-wetting fluids will have cosθ values that are negative. The liquid is prevented from moving to the evaporator.  (Peterson. this forces the liquid to move to the evaporator.1994) . This results in a positive capillary pressure that creates a pushing force on the liquid in the wick near the condenser. resulting in a negative capillary pressure that creates a suction force on the liquid in the wick.

Simple to design. Tend to have a higher capillary limit than homogeneous wicks but cost more (Peterson. 1995)   Composite. manufacture. Tend to have either high capillary pressure and low permeability or the other way around.1994) (Faghiri. and install .Wick Design  Two main types of wicks: homogeneous and composite. Homogeneous. 1995).made from one type of material or machining technique. . (Faghiri.made of a combination of several types or porosities of materials and/or configurations. Capillary pumping and axial fluid transport are handled independently .

Thermal conductivity . Permeability .large pore radius results in low liquid pressure drops and low flow resistance.1994) .  Design choice should be made that balances large capillary pressure with low liquid pressure drop.Wick Design  Three properties effect wick design:    1. (Peterson.a large value will result in a small temperature difference for high heat fluxes. 2.a small capillary pore radius (channels through which the liquid travels in the wick) results in a large pumping (capillary) pressure. High pumping pressure. 3. Composite wicks tend to find a compromise between the two.

1994).com/Resources/EC_Articles/SEP96/sep96_02.htm (Peterson. .Wick Design http://www.electronics-cooling.

Watch the saturation temperature for your desired fluid at the desired heat pipe internal pressure.Choosing the Working Fluid  Heat pipes work on a cycle of vaporization and condensation of the working fluid. if the operating temperatures are too low the fluid will not be able to evaporate. For instance.  . the working fluid must be compatible with the wick and container material. In addition. (Peterson. the fluid may not be able to condense. the fluid must be able to operate within the heat pipe‟s operating temperature range. which results in the heat pipe‟s high thermal conductivity. if the operating temperatures are too high. However. When choosing a working fluid for a heat pipe. 1994).

Choosing the Working Fluid Operating temperature ranges for various working fluids: http://www.cheresources.com/htpipes.shtml .

as the operating temperature range of the working fluid increases. . Choice of working fluid should also incorporate the fluid‟s interactions with the heat pipe container and wick.Choosing the Working Fluid  Generally.  (Peterson. 1994). the heat transport capability increases.

Choosing the Working Fluid Chi(1976) developed a parameter of gauging the effectiveness of a working fluid called the liquid transport factor:  l Nl  l Where is the latent heat of vaporization and  is the surface tension. occurring in low to moderate temperatures. Subscript l refers to the liquid For electronics cooling applications. . 1994). (Peterson. Another common fluid is ammonia. water is the liquid with the highest liquid transport factor.

Possibility of external corrosion. Evaporator and condenser size and shape.Container Design  Things that should be considered for container design:       Operating temperature range of the heat pipe. (peterson.1994) . Compatibility with wick and working fluid. Prevent leaks. Internal operating pressure and container structural integrity.

Container Design  Stresses:  Since the heat pipe is like a pressure vessel it must satisfy ASME pressure vessel codes. 1994) .  (peterson. Typically the maximum allowable stress at any given temperature can only be onefourth of the material‟s maximum tensile strength.

 (Peterson. 1994) .Container Design  Typical materials: Aluminum  Stainless steel  Copper  Composite materials  High temperature heat pipes may use refractory materials or linings to prevent corrosion.

or the fluid and container may undergo a chemical reaction with each other.Heat pipe Compatibility  When designing a heat pipe. (Peterson. wick. the working fluid may not be wettable with the wick. For example. and container must function properly when operating together. 1994) . the working fluid.

Heat pipe Compatibility  Working fluid/ material compatibility. 1995) . (Faghiri.

as well as. or soldered techniques. Methods used to join the parts include use of thermally conductive adhesives. (Peterson. 1994)  . brazed.Heat Sink/Source Interface  The contact resistance between the evaporator and the heat source and between the condenser and the heat sink is relatively large and should be minimized.

(Peterson. As the temperature of an electronic part increases the rate of failure increases.Heat Pipes in Electronics Cooling  Cooling of electronics has one primary goal: maintain a component‟s temperatures at or below the manufacturer‟s maximum allowable temperature. 1994) .

they are good alternatives to large heat sinks. especially in laptops where space is limited.   They are good alternative to air cooling because of their better heat transport capabilities. . (Peterson. and variable evaporator and condenser sizes. they provide constant evaporator temperatures with variable heat fluxes. Therefore. 1994). high heat transfer characteristics.Heat Pipes in Electronics Cooling  Heat pipes are excellent candidates for electronics cooling because of their high thermal conductivity. Air cooling may still be used to remove heat from the condenser.

Heat Pipes in Electronics Cooling  Common heat pipes used in electronics cooling:     Micro heat pipes Capillary looped heat pipes Flat plate heat pipes Variable conductance heat pipes .

1994) .  (Peterson. or chip). the heat pipe‟s evaporator may be attached to an individual heat source (power transistor.Heat Pipes in Electronics Cooling  In single component cooling. The condenser is attached to a heat sink to dissipate the heat through free or forced convection. thyristor.

(Peterson.Heat Pipes in Electronics Cooling  Cooling can also occur with multiple arrays of devices or entire printed wiring boards. 1994) .

Heat Pipes in Electronics Cooling  An arrayed heat pipe cooling system (Peterson. 1994) .

Heat Pipes in Electronics Cooling  Heat pipe cooling a component set up in an array (Peterson. 1994) .

Heat Pipes in Electronics Cooling  Since many semiconductors are small. Good for applications where space is limited like laptops. (Peterson. micro heat pipes may be used for cooling individual semiconductors or an array. 1994) .

1994) . another method of cooling a heat source may be employed. cooling the transistor.Heat Pipes in Electronics Cooling  When the electrical power is high and the heat rejection requirements large and nucleate pool boiling occurs. To reduce the drop. you can make the device a part of the wick structure to ensure that fresh liquid is always in contact with the heat source. (Peterson.  In the image to the right the heat source (a transistor chip) is in contact with the working liquid and the working liquid is being evaporated away.  Nucleate pool boiling causes a large temperature drop. Further providing cooling to the transistor.

 Low cost  Proven alternative to conventional methods of electronics cooling.  (Peterson.Heat Pipes in Electronics Cooling  Summary: Heat pipes enable devices with higher density heat dissipation requirements and greater reliability. 1994) .

Current Research in Electronics Cooling  Laptops today perform well and are small. Excess heat may slow down the processor‟s speed or shut the laptop off. therefore. (Junnarkar. 1999) . 2003)   First time a heat pipe used in a laptop was in 1994.. they have high heat dissipation demands. (Ali et al. Current heat pipes move the heat from the CPU to a small heat sink.

. Wang and Peterson (2003) have come up with two different micro heat pipe setups for laptop cooling:   Micro heat pipes configured into flat plate shapes were employed to cool a CPU. The heat sink was smaller in size than one not attached to a heat pipe because the base of the heat sink attached to a heat pipe experiences more uniform temperatures and therefore.Current Research in Electronics Cooling  Because micro heat pipes are small they are very useful in cooling of laptops where space is highly restricted. an increased efficiency. The condenser was attached to a heat sink.

.4 mm long and 25. to form an enclosed space. in the shape of a flat heat pipe.  No capillary wick structure needed because of the micro heat pipe‟s sharp corners. with parallel wires and two copper sheets were formed.  The fan is strategically placed to provide forced convection to the heat sink.Current Research in Electronics Cooling  Two different configurations were developed  Both were 152.4 mm wide  Layers of copper screen mesh.

Current Research in Electronics Cooling .

tilt angle had no effect on amount dissipated.  Important because laptops experience operation in many orientations. tilt angle effected the amount of heat dissipated  In configuration 2.Current Research in Electronics Cooling Main Results:  In configuration 1. .

(About.Current Research in Electronics Cooling Mesh number is defined as the number of openings per linear inch.2006) .

Current Research in Electronics Cooling  Things that increased heat transport capacity: Increasing mesh number  Increasing wire diameter  .

Current Research in Electronics Cooling  The thermal resistance from the heat sink to the device junction. due to the cooling of the heat pipe with forced convection. is greater for case 1 than case 2 at all air velocities. The values were determined from the relation:  ja  T j  Ta Qc  1 Rtotal Where Qc is the heat dissipated through the heat sink .

Current Research in Electronics Cooling .

the heat capacity of a micro heat pipe is restricted by the heat sink‟s ability to transfer heat through convection  Heat transfer not restricted by the capillary limit.Current Research in Electronics Cooling  Other discoveries: Within the CPU‟s operating temperature limit.  .

is not reached due to deficiencies in the heat sink‟s ability to transfer heat through convection.Current Research in Electronics Cooling The maximum heat transfer limit provided by the heat pipe. for the most part. .

Current Research in Electronics Cooling  Case 2 provided a lower thermal resistance and a greater heat transport capacity than Case 1. Case 1 transported 24W at 85C and 1.85 C /W resistance.55 C /W resistance. Case 2 transported 52W at 85C and .   .

electronics-cooling. 2003. pp24-33. V.2003. March 2006. and Fritz. http://www. Heat Pipe Conf.References                  About. March 2006.htm Ali. “Heat Pipe Heat Sink „Heat Kicker‟ for Cooling of SemiConductors.” Proc.. “ Flat Heat Pipe Cooling Devices for Mobile Computers. R. Zorbil..com/petrucci/medialib/media_portfolio/text_images/FG13_04. HJeat Pipes. D. Tokyo. J. F. Prentice-Hall. R.png Murase. and McCabe. The Chemical Engineer’s Resource Page. McGraw-Hill. Inc. March 2006. A. General Chemisty Principles and Modern Applications. 6 th Int.. and Polasek.. L..D. and Reay. New York.” ASME International Mechanical Engineering Congress. Compainion Website. Nov. Heat Pipe Science and Technology. 1994.. France. 3rd. Amin. .phy-astr.S. Vol...shtml Chi.P.. and Grubb. 3rd Int. ”Advanced Heat Pipe Thermal Solutions For Higher Power Notebook Computers”. Y. (2006). Sandeep. 2004. Hyperphysics -capillary action. Int.. 2006. (2006) May 2006.lightstreamphotonics. K. May 2006. Koizumi..about. http://www..JPG Warner. S. S. Tsukuba. http://cwx..com/images/tech_orangecontainer_small.. Basilius.. and Ishida. G. G. Harwood.. 1976. Petrucci.. 1987 “Heat Pipes for Cooling of High Density Printed Wiring Boards. http://www. pp273-279. US: Taylor &Francis Junnarkar. A. pp373-376. 1982. “ Direct Heat Pipe Cooling of Semiconductor Devices. Lightstream Photonics (2003). “Laptops Cool off with „Smart‟ Heat Pipes. pp.531-536. H. 3rd. E. W. Yoshida. Tanzer. N.. 1995.A... 1988. Canada: John Wiley & Sons.com/htpipes. T. Sekhon. Faghri. Stulc. Herring.. Palo Alto. K. “ Enhancement Cooling of the Boards with Integrated Circuits by Heat Pipes. Inc... R. P. “ Proc. T. 1999. Japan. http://hyperphysics.” Furukama Review. P.gsu. Inc.” Proc.H. 1978. Heat Pipe Conf. Peterson. http://composite. G.com..com/Resources/EC_Articles/SEP96/sep96_02. Homane. March 2006. Oxford..htm Wang. 2003. D. 15-21. 2001.2.prenhall.ed. CA. Washington. Dunn. and Peterson. K. Narayan S.” Cnet News. 1982.C. Nave.. Japan.html Nelson.. Grenoble.cheresources.edu/hbase/hframe.. Dehoff. March 2006. Heat Pipe Symp. Pergamon. Heat Pipe Theory and Practice. An Introduction to Heat Pipes. January22.com/library/glossary/m/bldef-m3304.

.02.. ..9550 904731:/ .3072:89 1:3.70.9-9  03/0833..0.3/.250 904731:/ 2.3072.47901:/.439.94390.-0990.3/.4907  !0907843  .9550425.943574507034507.0.93940907 470. :3/074.439.349-0099.

9550425.0.9-9  4731:/.

425.9-9 . 2. .7  .907.

0.9$3.

$4:7..43/:..9.43/03807.0.947.3/900.08 .3/900..439.039071.70.8  -7.983870.7983.547.36:08  !0907843    .9.0  %0.984:7./08.:/0:8041 9072.970889.0-0900390 0.3..3/84:/-02320/ 094/8:80/9443905.0/ 4784/070/90.3/-09003 90.80.0 .0.

70.-0902507.74.9!50830.80890 7.39.8443  4434100.4.808 !0907843  .2.42543039 8902507.9:707 82.9743.3:1..8.793.70.9743.8430572..9743.3 .9:708.5.300.9:7041.947-0490 2.0.9:70 8 90902507.2:2.90411.:703.

90789..9743.7044/..700.38107..90814700..5945807085.08290/ %0.:8041 907-099070.-00.9!50830..5..947902507.947.3.094..0039.7...9073.9 838 0850.9.700.8443  0.9073.3//.9743.390.:80419079072.385479.9.0894..7.7.9174290..547.43/03807808  %0701470 90.95508.443-0.443-0.7044/./0 ..9  0.7.4432.4389.8 90574.-00.3/.997.9:7089.9 1:08 .00.8 .547.43/:.3/..43/03807 !0907843     . 89-0:80/947024.-908 7.9.0.997.

00.95508:80/300.95.740.9.43/:.3.9!50830.74450/0.900.95508 '.0.95508 .9743.5.-0.95508 .8 .7.8443  422430.95508 .443     .9743.

98394/885.8443  3830.0 540797.42543039.9974:170047147.0/.943   !0907843   .0/94.443 900.43/038078...9550 8 0.90 900.0.9472.9!50830..0.9 84:7.33/.0.0.-0.43.5 %0.547.388947 978947 47./:.9743.0/94.99.99.

844.:792:950.841/0.9743.8443  443.0.08 470397057390/73-4.77..3.9!50830.7/8  !0907843   ...

8443  3.0/0.9550.0.77.9!50830.9743.44388902 !0907843  .

9743..3.443.0.9!50830.9550.42543039809:53. !0907843   .8443  0.77.

8443  $3.9438 07085.0.947847.802.7082..08290/ 0.59458 !0907843   .  44/147.77.955082.02.-0:80/147.3.  2.43/:.443 3/./:.43/:.740.9!50830.9743.9478..3802.55.

-002540/ :.8443   039000.0.349072094/41.9:70/745 %470/:.984:7.32.9!50830.5.54078. .44390 97.90544-43.388947 .439.0..0.794190 ..5 83..99904736:/.388947  !0907843   .:808.4439490 97..984:7.0949079900.0 ..090/745  4:..3/3:..70 .:78  .83.3/90 4736:/8-030.917086:/ 8.943706:7020398.70 902507.388947 3902.443.97.97.439..897:.547.3/ 900./3..0.90544-434..90/.9:7094038:709.9700.9 84:7.99900.090/0..0.0 :7907574.0 2.9743.

08907 /03890.8.90770.443   !0907843  .9.9743.2094/8 4100.7 0.943706:7020398.489  !74.43.9743.094.-9  4.9550803..-0/0.8443  $:22.9073.9!50830.03.03943.9/885.3/ 70.0.

  .7082.943 /02.92. 90701470 90.:77039#080.50714720..0.7.7.3/.7     789920.3/8  ..0880.5945.9550824.8443  .91742 90!&94.84/4390574.30.983  09.82.9550:80/3.5945411  :33.5945894/.00.08847 8 8500/478:990.9743.0.83  :770390.0900.9/885.

443   .90/ .9550050703.882.740.. .44.5945 .:77039#080.073809....0.3/90701470 . 0.44341.9550-0.0/94.082470:31472 902507.33.43/03807.03.!& %0.8.95508.3/!0907843   .908.80/011.9743.70 .7.95.7082.0/94.95508.740.740..5945807085..99.0.07:801:3.3.9:708.70.431:70/3941.420:59 94/11070392.0.508070 02540/94.90.0 870897.8443  0.:8090-.983 %00.:802.30.3430349 .9550809:58147.983.99.983 .0/94..99.8041900.

9550 941472.473078 %01.5.7.3/ 94.7.9438070/0...431:7.983 .0708.0 4.0.1.:804190 2.90./0147.303.30.38897.9:70300/0/-0.75.9743.8443      %4/1107039.5041.480/85..43.897:.7003208 95.455078009807014720/ 3908.94394900..455078.3/ 22/0 .07841.0/ .7.:77039#080.5.740.0/94574.9 0.0450/ 49070 2243.9550 88..

8443 .30.:77039#080.9743.7.

0 4507.3#08:98  3.9 43.300110.431:7.90/  3.943 99.9438 .8443 .431:7..94332.24:39410.30.9/885./340110.:77039#080.59458050703.347039.90/90 .90/  25479.:80.24:39/885.7.39-0.9743.30.943 99.

9743.30.  -4:9   .73.8903:2-074145033850730.7.8443 083:2-078/0130/.:77039#080.

997.20907  .832083:2-07  3...7.8443  %389.8370/.30.93.70.385479 .80/0.:77039#080.9 3.70.9743.5.70.

9839490/0.943 870...7.90/974:900.8443  %09072.95509147.809.44341900.43.. 070".983 .9743.0 :3.9/885. 8900.907147.0/ .7 .   #949.01742900.3.30.  %  %.908 %0.943 7 ..:77039#080.0.. ".:08070/0907230/17429070.04.943 /:09490.3.80.9.70889.

30.9743.8443 .:77039#080.7.

943  0.9743.983 8.43.7.740.381070.:77039#080.2.90/-900.5.9550 870897...0.9:70 29 900.7 29 .8443   907/8.5.9.0708 9390!& 84507.9974:..30.997.90/-90.-994 97.941.93902507.4.3810734970897.

/0/-900.943 .38107 0.:77039#080.30.0//:094/01.983 8.997.0.9550 1479024895.03.083900.79 834970.7.9974:.8443 %02.3810729574.9743.43.2:20.-99497..

5. 70889.30.9H.7.3/.0.9743.4079072.8097.3.80574.997.99.9070.3854790/.3/ H.70..80 ./0/.8443  .385479 ..3.:77039#080.

3.70889.3854790/.3/ H.9H.0 .8097.

70889.3.0   .

  995.08                  -4:9 3.    .#010703.

.

.42.4254890 .-4:9 .

-7.7.

7.488.

2.

7.$4:94384707!407490-44 425:9078  .3.0!.9!50%072.7/8 !74.0   .0 55   %002.0/0..33007 8#084:7. 9 39 0.9!508147443410389!7390/734.   /.307  .3.  995.-/01 2  92   0411 # .-0 $  0..9!50431 7034-0 7.3/..3/7:-.8:8  %.

.

42. .07084:7.08 .

    0.7.3..42 .3/!7.9!508 30908 .9!50%047.  98970.7   .7 23  0.  995.7.3/005  .8   .59458444119 $2.9!50$.0 .7 $.79 0.8 :33.  047 :33 !  .9.2!4943.344 &$%.95508 892   0.7.03.3/#0.7.9!508 7/ 0/ !07.243  147/  .47 7.3:.0.3/%0.

.

8 .254943.42. 98970.

08.2.

  995.30. 53 :7..307*82.*47.5.07 14744341$02 43/:.5.    0.7.943   .80 % 48/..0 %44 .9478 :7:.0 # 50758.9!500.8 .9$3 0.7.2.3/8/.3 '4  55   .439.#0.9.90..  4:2 % .

.

8 5 .50758.897 8: 0/:.

-.80.

90..943940.  3.3..43/:.17.3$ 42.30 .508./. 425.430 $438 3.3/79     70.744/  $ 0773  0307. !097:..   995.3/4/07355.0 .9438   !7039.20 92 0843  $043  .9!508  .3430-890 ..0289!73.494  55  !0907843  !  33974/:. #  .9470.9!5044341$02.7.7.9 !50431 !. 7/ 39 0.08 !74.

.

. .42. 5703.

.5097:..

-.20/.

*5479144.20/.

08.909*2.

  995.7307 $    .*  ! .

.

9743. 00.443 .42.8 .

08.#084:7.

*79.08.

$!.

805*  92 .0203944341904. 7/ 39 0.5. ! .9!50$25 %8::-. 0.081474-0425:9078 $39073.9!508  !74.90/7.3/!4.80    3..     47- ' $9:.3  . .330073437088 .943..3/!0907843     .3 55  .:98-0.9!504430.3.90.83943   4..7/893907.3.

Sign up to vote on this title
UsefulNot useful