Texas Instruments | AN-2020 Thermal Design By Insight, Not Hindsight (Rev. C) | Application notes | Texas Instruments AN-2020 Thermal Design By Insight, Not Hindsight (Rev. C) Application notes

Texas Instruments AN-2020 Thermal Design By Insight, Not Hindsight (Rev. C) Application notes
Application Report
SNVA419C – April 2010 – Revised April 2013
AN-2020 Thermal Design By Insight, Not Hindsight
.....................................................................................................................................................
ABSTRACT
This application report provides an in-depth discussion of thermal design.
1
2
3
4
5
6
Contents
Introduction .................................................................................................................. 2
Definitions .................................................................................................................... 2
2.1
Example: Calculating Your Required θJA ........................................................................ 5
Rules of Thumb ............................................................................................................. 5
3.1
Rule 1: Board Size ................................................................................................. 5
3.2
Rule 2: Thermal VIAS ............................................................................................. 6
3.3
Rule 3: Copper Thickness ........................................................................................ 7
3.4
Rule 4: Avoid Breaks in the Thermal Path ...................................................................... 7
3.5
Rule 5: Heat Sink Placement is Just as Important as Selection ............................................. 8
3.6
Rule 6: Multiple Heat Sources: Superposition Almost Works ................................................ 9
Summary ..................................................................................................................... 9
Derivation of Rule 1 (Thermal Resistance from the Surface of the PCB to Ambient Air) ......................... 9
References ................................................................................................................. 11
List of Figures
1
IC Mounted On A Four-Layer Printed Circuit Board .................................................................... 2
2
Simplified Thermal Resistance Model For A Typical PCB............................................................. 3
3
Expanded Thermal Resistance Model For A Typical PCB
4
A Hot Spot Is Created When A Break Is Created In The Thermal Path ............................................. 8
5
Thermal Resistance Model With Heat Sink Attached To The Package Top ........................................ 8
6
Thermal Resistance Model With Heat Sink Attached To Board Bottom ............................................. 9
...........................................................
4
List of Tables
1
Typical Thermal Resistance Values ...................................................................................... 4
2
Calculating Coefficienct Of Heat Transfer Using Grashof Number. Used For Natural Convection
Calculations ................................................................................................................ 11
All trademarks are the property of their respective owners.
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
1
Introduction
1
www.ti.com
Introduction
All electronics contain semiconductor devices, capacitors and other components that are vulnerable to
thermally accelerated failure mechanisms. Thermal design becomes vital to improving the reliability of any
design. Unfortunately, thermal design can be very difficult because of the mathematical analysis of fluid
dynamics for complex geometries. Although this remains true for the foreseeable future, this application
report covers the basics of thermal design for DC-DC converters using a simplified resistor model of heat
transfer. Focus is on the thermal design for the semiconductor devices, but all of these techniques can be
applied to other components. The resistor model is very useful for quickly estimating your design
requirements, such as the PCB size and whether airflow is required. Finite element analysis software can
then be used to analyze the design in more detail. The listed reference material is home to additional data
and many useful thermal calculators, covering material that is beyond the scope of this document.
Our discussion of thermal design will begin with the definition of parameters used in data sheets such as
θJA and θJC, and end with some rules of thumb for the thermal design of a DC-DC converter, including their
derivation. An accompanying spreadsheet (see References) uses these derivations to quickly provide a
ballpark figure for the thermal performance of your design.
2
Definitions
Description of Thermal Terms
Parameters of interest : θJA, θJC, θCA, θJT
Silicon Die (Junction)
Exposed Pad (EP)
Package Body
EP Landing Pattern
2
Top Copper Layer
2
Inner Ground Layer
2 oz/ft (0.070 mm)
1 oz/ft (0.035 mm)
2
Second Inner Layer
1 oz/ft (0.035 mm)
2
Bottom Ground Layer
2 oz/ft (0.070 mm)
12 mil diameter
Thermal Vias
1.6 mm r 10%
Figure 1. IC Mounted On A Four-Layer Printed Circuit Board
The most commonly specified parameter, in data sheets, for thermal performance is θJA. θJA is defined as
the thermal impedance from the Junction, of the integrated circuit under test, to the Ambient environment.
If we describe it using a resistor model, it is the parallel combination of all the paths that heat can take to
move from the IC junction to the ambient air. The equation for this thermal resistance is:
TJA =
TJUNCTION - TAMBIENT
Power Dissipation
(1)
In our resistor model the heat transfer, measured in watts, takes the place of charge transfer measured in
amps, and the temperature potential between the junction and ambient temperatures replaces the voltage
potential. The heat that needs to be transferred away from the junction is the power dissipated by the IC.
2
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
Definitions
www.ti.com
Ambient Air Temperature
(TA)
Junction Temperature
(TJ)
TT
Case Top Temperature
(TT)
JT
TA
JC
CA
TA
TJ
TC
Exposed Pad/Case Temperature
(TC)
JA
TJ
TA
Figure 2. Simplified Thermal Resistance Model For A Typical PCB
There are two primary heat paths for a DC-DC converter represented above by their associated thermal
resistances. The first path travels from the junction of the IC to the plastic molding at the top of the case
(θJT) and then to the ambient air by convection/radiation (θTA). The second path is from the junction of the
IC to an exposed pad (θJC). The exposed pad is then connected to the PCB, where the heat travels to the
surface of the PCB and to the ambient air by convection/radiation (θCA).
There is one point of confusion that is common in defining θJC.
For DC-DC converters without an exposed pad, θJC is defined as the thermal impedance from the junction
to the top of the case. This is in direct conflict with our previous definition of θJC, being the thermal
impedance from the junction to the exposed pad. This confusion comes about because of the large
number of packages that DC-DC converters have been shoved into over the years. As newer packages
with exposed pads were released into the market it was decided that θJC should represent the lowest
thermal impedance path from the junction of the IC to the outside world.
Now that we have cleared up the terminology, we can discuss the usefulness of various parameters.
Use the value of θJA given in the data sheet to compare different packages, and use it along with the IC
power dissipation for a sanity check in your design. The high thermal resistance of the plastic packaging
ensures that most of the heat travels from the exposed copper pad to the PCB, which usually has a much
lower thermal resistance. A heat sink can be added to either the top of the package or directly beneath the
exposed pad on the backside of the PCB. Again, because of the high thermal resistance of plastic, a heat
sink will be more effective when connected to an exposed metal pad, either directly or, through thermal
vias.
Since most of the heat transfer is through the exposed pad to the PCB it becomes immediately apparent
that the value of θJA is highly PCB dependant. In other words, the most critical value to determine in any
design is thermal resistance of the PCB (θCA). Well what, exactly, is θCA and how is it calculated? θCA is the
equivalent resistance of a thermal resistive lattice that centers on the IC and ends at the surfaces of the
board. It is the final of your freshman year, Circuits 101 class, all over again. Figure 3 below shows the
details.
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
3
Definitions
www.ti.com
Ambient Air Temperature
(TA)
Junction Temperature
TJ __
SA
SA
SA
JC
TC
Cu
Cu
FR4
Cu
FR4
Cu
Cu
FR4
Cu
Cu
FR4
Cu
FR4
Cu
SA
FR4
Cu
FR4
Cu
FR4
Cu
VIA
Cu
FR4
SA
Cu
FR4
FR4
SA
Cu
SA
FR4
Cu
Cu
SA
SA
SA
Figure 3. Expanded Thermal Resistance Model For A Typical PCB
There are new terms to add to our ever growing lexicon of θ's. θCu is the thermal resistance of our board's
copper to lateral heat transfer. θFR4 is the thermal resistance between the copper planes provided by the
vertical resistance of FR-4 laminate. θVIA is the thermal resistance of the thermal vias placed directly
underneath the exposed pad. θSA is the thermal resistance from the surface of the PCB to the ambient air.
It is a combination of convective and radiative heat transfer. If we break the board into 1 cm squares, the
typical values for these resistances are listed below.
Table 1. Typical Thermal Resistance Values
Nam Value
e
Description
Conditions
θCu
Lateral thermal
resistance of
copper plane.
Length = 1 cm, width = 1 cm,
1 ounce copper thickness =
0.0035 cm, thermal
conductivity of copper (λCu) =
4 W / (cm °C)
θVIA
71.4 °C /
W
261 °C / W Thermal
resistance of
typical 12 mil
via.
Via length = 0.165 cm (65
mils), 0.5 ounce copper
plating thickness = 0.00175
cm, drill hole radius = 6 mil
(0.01524 cm), thermal
conductivity of copper (λCu) =
4 W / (cm °C)
Equation
TCu =
1 x Length
èCu
Area
°C cm
x 1 cm
W
1 cm x .0035 cm
0.25
=
(2)
1 x Length
OCu
TVIA =
Area
°C cm
x 0.165 cm
W
2
S x [(0.01524 cm) - (0.01524 cm-0.00175 cm)2]
0.25
=
(3)
θSA
θFR4
4
1000 °C /
W
13.9 °C /
W
Thermal
resistance from
the surface of a
1 cm square of
the PCB to the
ambient air due
to natural
convection.
1 cm square, a first order
approximation of the heat
transfer coefficient of PCB
Board to Air for natural
convection is (h) = 0.001 W /
(cm2 °C)
Vertical thermal
resistance of
FR-4 substrate.
1 cm square, FR-4 thickness
= 0. 032 cm (12.6 mil),
thermal conductivity of FR-4
(λFR4) = 0.0023 W / (cm °C)
2
°C cm
1
1000
W
h
=
TSA =
Surface Area 1 cm x 1 cm
TFR4 =
1 x Thickness
èFR4
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
Area
435
=
(4)
°C cm x 0.032 cm
W
1 cm x 1 cm
(5)
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
Rules of Thumb
www.ti.com
The amount of variability in PCB designs is significant. You can see that the thermal resistance depends
on board size, airflow, PCB thickness and many other parameters. For this reason, a series of JEDEC
standards (JESD51-1 to JESD51-11) were developed, which specify the PCB size and layout for testing
θJA for different types of packages. DC-DC converters that are tested to these standards can be directly
compared to one another. Always check the data sheet to see what PCB parameters were used to
measure θJA. Later on we will discuss some tips for designing your PCB, but the final design, and thus θJA,
depends on the end user.
2.1
Example: Calculating Your Required θJA
Calculate the θJA required for a DC-DC converter with an output voltage, VOUT, of 2.5V and output current,
IOUT, of 4 amperes. The converter efficiency, η, is 91.4%. The ambient temperature, TA, is 50°C. The
capacitors you selected are rated up to 100°C, and because of their proximity to the DC-DC converter you
decide that the maximum junction temperature, TJ, you would like the converter to reach is 90°C.
The power dissipated by the converter, PD, can be easily calculated.
PD = VOUT x IOUT x ( 1 -1)
PD = 2.5V x 4A x (
(6)
1 -1) = 0.94W
0.914
(7)
Assuming that all of the power is dissipated internally to the IC, (a fair assumption if using LDOs or
modules) we can now calculate the maximum value of θJA.
TJA 7
90°C - 50°C
7 42.5 °C/W
0.94W
(8)
Since a θJA of less than 42.5°C / Watt is required for your design, you won’t get there with a SOT-23 which
has a θJA closer to 179°C / Watt. You would need to select a package like the 14 Pin eMSOP with a θJA of
around 40°C / W. If you would like to increase the thermal margins for your design, pick a package like the
TO-PMOD-7 or TO-263, which have θJA's of around 20°C / W. Remember θJA is board dependant and all
of these numbers come from using a JEDEC standard test board measuring 3” x 4”. If your board varies
from the JEDEC board, which it will, you will need to get a better estimate of thermal performance by
using θJC, and deriving θCA.
A package with a low θJC will have very good heat transfer to the case, usually to an exposed pad. θJC is a
very good indicator of a packages thermal performance. A low value for θJA implies a low value for θJC.
The θJC for packages designed to dissipate large amounts of power can be less than 2°C / Watt. Typically,
packages with no exposed pad have a θJC of greater than 100°C / Watt. That means that for every watt
dissipated in the package the temperature difference between the junction of the IC and case will increase
100°C. If the value of θJC is not included in the data sheet it can often be requested from the manufacturer.
3
Rules of Thumb
PCB DESIGN to meet a given θCA
Parameters of interest: θCA
A low value for θJC (less than 10 °C / Watt) is a good start for a thermal design, but we still need to design
a PCB or heat sink to transfer the heat from the case to the ambient air. This section will provide simple
guidelines so that we can avoid calculating the full resistor model. The derivations for these rules of thumb
will be discussed later.
3.1
Rule 1: Board Size
a)
Board Area (cm2) 8 15.29
Board Area (in2) 8 2.37
cm2
x PD
W
in2
x PD
W
(9)
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
5
Rules of Thumb
www.ti.com
With only natural convection (i.e. no airflow), and no heat sink, a typical two sided PCB with solid copper
fills on both sides, needs at least 15.29 cm2 (≊ 2.37 in2) of area to dissipate 1 watt of power for a 40°C rise
in temperature. Adding airflow can typically reduce this size requirement by up to half. To reduce board
area further a heat sink will be required.
Several assumptions are made here. First, that any enclosure for the PCB does not restrict the natural
convection of either side of the PCB. Second, that the PCB provides a low thermal resistance from the IC
to the edges of the board. This can be accomplished by attaching the exposed pad to copper ground
planes that extend to the edge of the PCB. With a four layer board, the internal ground planes can also be
used to transfer the heat to the edges of the board. This can improve the thermal performance of even a
well designed two-layer board by up to 30 percent.
b) The following equation can be used to approximate the minimum board size if θJC is known.
500 °C x cm
W
Board Area (cm ) 8
JA JC
2
77.5 °C x in
W
Board Area (in ) 8
JA JC
2
2
2
(10)
where θJC is obtained from the datsheet, and θJA is calculated from the power dissipation, ambient
temperature, and maximum junction temperature as shown in the previous example.
3.1.1
Example: Calculating the Required Board Size to Hit a Target θJA
Using our previous DC-DC converter example with VOUT = 2.5, IOUT = 4A, and PD = 0.94W, we determined
that a θJA = 42.5°C / W was necessary for the design.
a) Using the rule of thumb, with no airflow the PCB should be at least
Board Area (in2) 8 2.37
in2
x 0.94W = 2.23 in2
W
(11)
b) Compare this to the equation using θJC. We determined that a 14 pin eMSOP would be a good choice
for this design because the θJA is around 40°C / W on the JEDEC board. A 14 pin eMSOP has a θJC of ≊
7.3°C / W. Using the equation we can calculate a minimum PCB area
2
77.5 °C x in
W
= 2.2 in2
Board Area (in ) 8
°C
°C
42.5
- 7.3
W
W
2
(12)
If we had picked a package that was not suitable to the task such a SOT-23 with a θJC ≊ 100°C / W, the
answer would be negative telling us that there is no amount of board area that could be used to heat sink
this device and meet the specifications.
However, if we had picked the TO-PMOD-7 package with θJC ≊ 1.9°C / W we could have reduced the
board size, or gained some margin for our design.
2
77.5
Board Area (in ) 8
42.5
3.2
°C x in
W
2
°C
°C
- 1.9
W
W
= 1.91 in
2
(13)
Rule 2: Thermal VIAS
åVIAS
°C
261 W
# of Thermal Vias
(14)
A typical 12 mil diameter thru hole via with 0.5 oz copper sidewalls has a thermal resistance of 261 °C /
Watt. Place as many thermal vias as will fit underneath the exposed pad to form an array, with 1mm
spacing. Connect the vias to as many layers of copper as possible to spread the heat away from the
package and to the PCB surface where it can transfer to the ambient air. For many DC-DC converters the
exposed pad is electrically connected to ground and thus, the internal ground layer and the bottom ground
layer are usually the most convenient copper planes for heat transfer. The thermal resistance is
significantly lowered by having as solid a bottom layer ground as possible.
6
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
Rules of Thumb
www.ti.com
The thermal resistance of a via can be calculated using the following equation
TVIA =
1 x Length
OCu
S x [(radius)2 - (radius ± plating thickness)2]
(15)
where λCu = 4 W / cm K, the length is the thickness of the board (0.1561 cm typ), the radius is the radius
of the drill hole (0.1524 cm typ), and the plating thickness = 0.0035 cm multiplied by the copper weight in
ounces.
3.2.1
Example: Thermal Impedance of VIA Array
Using the 14-pin eMSOP. The exposed pad measures 3.1 x 3.2 mm. This would allow us to fit 16 thermal
vias underneath the exposed pad of the device using 1mm spacing. Connect the vias to the copper
ground layers to spread the heat. The thermal resistance of this 4 x4 array would be approximately 261 /
16 = 16.3°C / W
There are many advocates of completely plating closed the thermal vias to improve the thermal
performance. A typical 8 mil diameter plated closed via has a thermal resistance of 128°C / W. The
thermal resistance of the 16 vias in parallel is 128 / 16 = 8°C / W. The improvement in heat transfer to the
bottom of the package is considerable, but it is usually not worth the cost. Plating closed the thermal vias
can double or triple the cost of your PCB design. A more economical option is to ask for 1 oz plating on
standard 12 mil vias, with perhaps a 10-20% cost adder. The thermal resistance of a single via is reduced
to 140°C / W, and for a 4 x 4 array the thermal resistance is only 8.75°C / W.
Another option is to use a package with a larger exposed pad such as the TO-PMOD-7. The exposed pad
measures 5.35 mm x 8.54 mm. This would allow us to place at least 40 vias. The thermal resistance of
the via array using standard 12 mil vias is only 261 / 40 = 6.525°C / W.
3.3
Rule 3: Copper Thickness
The thicker the copper in the board, the more easily heat can transfer away from the IC. The equation for
the thermal resistance of a copper plane to lateral heat transfer is
1 x Length
OCu
TCu =
Width x Thickness
(16)
where λCu = 4 W / cm K, Length and Width are in centimeters, and copper Thickness = 0.0035 cm
multiplied by the copper weight in ounces (0.5 oz. typical).
At least one ounce copper is recommended for all DC-DC converter designs. Two ounce copper is
recommended for designs that dissipate more than 3 watts. Four ounce copper is recommended for
designs that dissipate more than 6 watts.
To truly appreciate the value of thicker copper in a design we will look at the results for two "almost"
identical two layer boards. The only difference is in the thickness of the copper. The first board has one
ounce copper, and the second board has two ounce copper. Both boards measure 3 x 3" with minimal
copper on the top layer where the component is placed, and the bottom layer completely filled with copper
as a heat spreader. θJA = 28.3°C / W for the first board and 21.2°C / W for the board with thicker copper.
This is a 25% improvement in the thermal performance of the board, by only changing the copper weight.
3.4
Rule 4: Avoid Breaks in the Thermal Path
Maintain a copper ground plane on either the top or bottom copper layer with as few breaks as possible to
create a heat spreader on the PCB. Spreading the heat across the PCB provides a low impedance path to
the surface of the PCB and improves convective heat transfer. Traces perpendicular to the heat flow will
create high impedances (speed bumps) for the heat and create hot spots (traffic jams). If traces through
the copper heat spreader are unavoidable try to make them run parallel to the heat transfer, which flows
radially from the heat source. The thermal image below Figure 4 shows three almost identical boards with
the same power dissipation. The only difference is a wide cut made on the top copper that breaks the
thermal path. In the middle board, where the break is perpendicular to the heat flow this causes a 5.5 °C
rise in temperature on an otherwise identical board. On the right hand board, where the break is parallel to
the heat flow, the temperature rise is only 1.5 °C. This might make a serious difference if the board is
pushing the thermal limits of the design.
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
7
Rules of Thumb
www.ti.com
Figure 4. A Hot Spot Is Created When A Break Is Created In The Thermal Path
3.5
Rule 5: Heat Sink Placement is Just as Important as Selection
Heat sink selection must include the thermal resistance from the IC junction to the attachment point of the
heat sink to be effective. For best performance a heat sink should be attached to the lowest impedance
path to the IC junction.
3.5.1
Example: Heat Sink Performance for Different Locations
To understand rule 5 we will take a look at heat sink performance at two locations. If we again look at the
TO-PMOD7, the package has a very low θJC = 1.35 °C / W. The thermal resistance to the top of the
package is considerably higher because of the plastic interface. θJT can range from 50 to 200 °C / W for
packages with a plastic top. A heat sink on top of the package is connected in series with the high thermal
impedance of the plastic making the heat sink less effective. A heat sink on the bottom of the board is
connected in series with the low thermal impedance of the exposed pad and the relatively low thermal
impedance of the vias making the heat sink more effective. To determine the effectiveness of a heat sink
at the two locations, let's compare the resistive models.
TT
Ambient Air Temperature
(TA)
HEATSIN
K
Heat Sink Temperature
(THS)
Junction Temperature
(TJ)
Case Top Temperature
(TT)
JT
TA
JC
CA
TJ
TA
TC
Exposed Pad/Case Temperature
(TC)
JA
TJ
TA
Figure 5. Thermal Resistance Model With Heat Sink Attached To The Package Top
8
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
Summary
www.ti.com
Ambient Air Temperature
(TA)
Junction Temperature
(TJ)
TT
Case Top Temperature
(TT)
JT
TA
JC
CA
TA
TJ
VIAS
HEATSIN
K
Exposed Pad/Case Temperature
(TC)
TC
JA
TJ
TA
Heat Sink Temperature
(THS)
Figure 6. Thermal Resistance Model With Heat Sink Attached To Board Bottom
Newer MOSFET packages are improving the thermal conductivity to the top of the package by having an
exposed metal tab on the top. This style of package gives the user several heat sinking options.
3.6
Rule 6: Multiple Heat Sources: Superposition Almost Works
If there are multiple heat sources in your design, how do you accommodate that? Well the good news is
that superposition almost works. Almost? We are modeling the thermal environment as a resistive network
with current sources (thermal resistances and power dissipation); therefore, we can use the superposition
theorem, however, because some of those resistances have non-linear dependencies on temperature
there will be some error in our final result.
To apply superposition to a system with multiple heat sources, solve for one heat source at a time, while
leaving the others as open circuits in your system. Calculate the temperature rise at the locations of all the
heat sources due to the other heat sources. The results can then be added together to get an estimate of
the total temperature rise at the different locations.
4
Summary
Determining thermal performance is vital to any design, and should be considered before it becomes a
problem. We have seen how some rules of thumb, derived through insight into thermodynamic principles,
can be used at the beginning of the design process to help avoid drastic redesigns. Although this will not
replace the accuracy of modern finite element analysis software, it will give you a starting point for the
thermal design of your system.
5
Derivation of Rule 1 (Thermal Resistance from the Surface of the PCB to Ambient
Air)
Heat transfer from the board to the ambient air is primarily by convection and radiation. A widely published
figure for the heat transfer coefficient from the surface of the PCB (h) to air is 10 W / m2K. We will use this
as a starting point and look at the derivation of the heat transfer coefficient later.
To turn h into the thermal resistance from the board surface to the ambient air θSA, we take the inverse
and then convert from square meters to square inches and finally divide by the surface area.
2
2
in K
m K
1
2
155
0.1
1550 in
W
W
h
x
=
=
TSA =
2
Surface
Surface
Surface
1m
Area
Area
Area
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
(17)
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
9
Derivation of Rule 1 (Thermal Resistance from the Surface of the PCB to Ambient Air)
www.ti.com
The estimate used in rule one is easily derived from this if we assume that the thermal resistance from the
junction of the IC to the surface of the PCB is small compared to θSA. The resistive model simplifies θJA =
θJC + θSA. If there is convective heat transfer from both sides of the board, θSA is reduced by half. A typical
14 pin eMSOP has a θJC ≊ 7.3. Therefore, to limit the IC junction temperature to a 40 degree rise in
temperature for 1 W of power dissipation this translates to a required thermal resistance from board
surface to ambient air of
TSA =
TJA - TJC 40YC - 7.3YC
=
PD
1W
= 32.7
YC
W
(18)
The required board area is therefore
155
Board Area =
in2YC
W
2 x TSA
in2YC
W
= 2.37 in2
=
YC
2 x 32.7
W
155
(19)
Calculating h
Now that you have seen the derivation for board area, we will revisit the derivation of the thermal transfer
coefficient.
The thermal transfer coefficient can be calculated for forced convection with the following equation.
h=
Nu x èAIR
Length of
Board
(20)
where λAIR = thermal conductivity of air ≊ 0.024 W / m K.
Nu, the Nusselt number, named after Wilhelm Nusselt, is a ratio of the convective to conductive heat
transfer normal to a boundary layer. In our case the boundary is the surface of the PCB we are using to
heat sink the DC-DC converter. The following correlation can be used to calculate the Nusselt number
under laminar flow conditions.
(21)
Nu = 0.664 x Re
1/2
x Pr
1/3
for Pr 8 0.6
(22)
Pr, the Prandtl number, named after Ludwig Prandtl, is a ratio of Viscous diffusion to thermal diffusion.
The higher the numbers the better a fluid is at convective heat transfer as compared to conductive heat
transfer. Thus, a liquid metal such as mercury has a very low Prandtl number 0.015, and engine oil can be
as high as 40,000. The Prandtl Number for air is between 0.7 and 0.8.
Re, the Reynolds number is a ratio of the inertial to viscous forces for a given flow condition. The
Reynolds number can be used to determine the type of fluid flow. A laminar flow, characterized by lower
Reynolds numbers, is smooth and constant. A turbulent flow, characterized by higher Reynolds numbers,
where inertial forces are stronger than viscous forces, tends to have eddies and other flow instabilities.
The following equation can be used to calculate the Reynolds number.
Re =
Fluid Velocity x Density x Length
Viscosity
(23)
where Density = 1.184 kg / m3 and
Viscosity = 1.98e-5 kg / m s.
Strictly speaking the Reynolds number should not be used as a measure of turbulence for natural
convection. The Grashof number should be used (calculations below). However, I find that the simplicity of
using only one equation for both natural and forced convection often outways the advantage of improved
accuracy using the Grashof equation.
The velocity of air due to natural convection for a simplified vertical plate is
10
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
References
www.ti.com
VNC = 0.65 x [g x Length x
= 0.65 x [9.8
TBOARD - TA 1/2
]
TA
m
338K - 298K 1/2
]
x 0.0254m x
s2
298K
(24)
The velocity of air due to natural convection ranges from 0.1 to 0.7 m/s for many PCB’s, and for this
calculation is 0.118 m / s. In my experience using a relatively high velocity of Vc = 0.5 m / s is a good
starting point for a design and allows us to lump in the heat transfer from radiation and approximate the
result using the Grashof number without making additional calculations.
Finally, after several calculations we find that the heat transfer coefficient from the surface of the PCB to
the ambient air for a 0.0254m x 0.0254m (1 inch by 1 inch) board, with no forced airflow and air velocity
from natural convection = 0.118 m / s is
h = 7.484 W / m2K
(25)
Once radiation is added to the heat transfer, (calculation shown below) this value is very close to the
frequently published value (10 W / m2K) used for the derivation of rule one.
Table 2. Calculating Coefficienct Of Heat Transfer Using Grashof Number. Used For Natural
Convection Calculations
Name
Value
Description
Conditions
Gr
8.77 x
104
Simplified
equation for the
Grashof number.
Assumes ideal
gas propeties.
Ta = 298 °K. Tboard =
338 °K, gravity = 9.8
m/s. Kinematic viscosity
of air @ 25 °C = 15.68
x10-6 m2 / s. Length of
Board = 0.0254 m.
Nu
hradiation
14.39
Nusselt
Equation for the
two surfaces of
a horizontal
plate.
Pr = 0.7 for air at 25 °C.
Equation
4
4
g x [TBOARD - TA ] x Length
Gr =
TA x Kinematic Viscosity
9.8
=
3
2
4
4
3
m
2 x [(338 K) ± (298 L) ] x (0.0254 m)
s
2 2
-6
298 K x (15.68 x 10 m )
2
s
Nu = 0.54 x (Gr x Pr)
1/4
(26)
1/3
+ 0.15 x (Gr x Pr)
4
1/4
= 0.54 x (8.77 x10 x 0.7)
4
+ 0.15 x (8.77 x10 x 0.7)
1/3
(27)
0.78 W / Thermal transfer Ta = 298 °K. Tboard =
m2 K
due to radiation 338 °K, σ = 5.67 x 10-8,
e = 0.9, length of Board
= 0.0254 m, width of
Board = 0.0254 m.
4
4
e x ð x [TBOARD - TA ]
hRAD =
TBOARD
0.9 x 5.67 x 10
-8
=
W x [(338 K)4 ± (298 K)4]
2 4
m K
338 K
(28)
htotal
14.38 W
/ m2 K
Coefficient of
heat transfer
Thermal conductivity of
air (λAIR) = 0.024 W / (m
°C), length of Board =
0.0254 m
h=
=
6
Nu x thermal conductivity
Length
+ hRAD
W
14.39 x 0.024 m K
W
+ 0.78 2
m K
0.0254 m
(29)
References
AN-1520 A Guide to Board Layout for Best Thermal Resistance for Exposed Packages (SNVA183)
Cooling Zone Design Corner http://www.coolingzone.com/design_corner.php
" Forced Convection Heat Sink Thermal Resistance." Novel Concepts, Inc. 2012.
http://www.novelconceptsinc.com/calculators-forced-convection-heat-sink-thermal-resistance.cgi
Kollman, Robert, Constructing Your Power Supply - Layout Considerations.
www.ti.com/lit/ml/slup230/slup230.pdf
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
11
References
www.ti.com
LMZ1x/EXT Thermal Resistance Estimate (SNVU041)
Mathis, Miles. "The STEFAN-BOLTZMANN LAW a simplified derivation." The General Science Journal
http://www.wbabin.net/mathis/mathis64.pdf
12
AN-2020 Thermal Design By Insight, Not Hindsight
Copyright © 2010–2013, Texas Instruments Incorporated
SNVA419C – April 2010 – Revised April 2013
Submit Documentation Feedback
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and other
changes to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latest
issue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current and
complete. All semiconductor products (also referred to herein as “components”) are sold subject to TI’s terms and conditions of sale
supplied at the time of order acknowledgment.
TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TI’s terms
and conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessary
to support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarily
performed.
TI assumes no liability for applications assistance or the design of Buyers’ products. Buyers are responsible for their products and
applications using TI components. To minimize the risks associated with Buyers’ products and applications, Buyers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or
other intellectual property right relating to any combination, machine, or process in which TI components or services are used. Information
published by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty or
endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the
third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alteration
and is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altered
documentation. Information of third parties may be subject to additional restrictions.
Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or service
voids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.
TI is not responsible or liable for any such statements.
Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirements
concerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or support
that may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards which
anticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might cause
harm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the use
of any TI components in safety-critical applications.
In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TI’s goal is to
help enable customers to design and create their own end-product solutions that meet applicable functional safety standards and
requirements. Nonetheless, such components are subject to these terms.
No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the parties
have executed a special agreement specifically governing such use.
Only those TI components which TI has specifically designated as military grade or “enhanced plastic” are designed and intended for use in
military/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI components
which have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal and
regulatory requirements in connection with such use.
TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use of
non-designated products, TI will not be responsible for any failure to meet ISO/TS16949.
Products
Applications
Audio
www.ti.com/audio
Automotive and Transportation
www.ti.com/automotive
Amplifiers
amplifier.ti.com
Communications and Telecom
www.ti.com/communications
Data Converters
dataconverter.ti.com
Computers and Peripherals
www.ti.com/computers
DLP® Products
www.dlp.com
Consumer Electronics
www.ti.com/consumer-apps
DSP
dsp.ti.com
Energy and Lighting
www.ti.com/energy
Clocks and Timers
www.ti.com/clocks
Industrial
www.ti.com/industrial
Interface
interface.ti.com
Medical
www.ti.com/medical
Logic
logic.ti.com
Security
www.ti.com/security
Power Mgmt
power.ti.com
Space, Avionics and Defense
www.ti.com/space-avionics-defense
Microcontrollers
microcontroller.ti.com
Video and Imaging
www.ti.com/video
RFID
www.ti-rfid.com
OMAP Applications Processors
www.ti.com/omap
TI E2E Community
e2e.ti.com
Wireless Connectivity
www.ti.com/wirelessconnectivity
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2013, Texas Instruments Incorporated
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertising