Tag Archives: Dry Power Transformer

1 Introduction

Working temperature of electronic products have strict requirements. Power supply internal temperature too high

will result in temperature-sensitive semiconductor devices, electrolytic capacitors and other components fail. When

the temperature exceeds a certain value, the failure rate increases exponentially. Some statistics show that the

electronic component temperature is increased by 2 ℃, reliability, decreased by 10%; temperature 50 ℃, the life span

of only when the temperature is 25 ℃, 1 / 6. Therefore, control of electronic devices will encounter throughout the

chassis and internal component temperature requirements, and this is the thermal design of electronic equipment. The

high-frequency switching power supply with high power heating devices of this type of equipment, the temperature is

affecting their reliability and the most important factor, for the thermal design of the whole, there are strict

requirements. Complete thermal design includes two aspects: how to control the heat source of heat; how to heat the

heat spread it out. The ultimate goal is to reach thermal equilibrium after the temperature control of electronic

equipment within the allowable range.

2 heat control design

Switching power supply in the major heat components for the semiconductor switch (eg, MOSFET, IGBT, GTR, SCR,

etc.), high-power diode (such as ultra-fast recovery diode, Schottky diode, etc.), high frequency transformers, filter

inductors and other magnetic components and false loads. For each type of heating components have different methods of

heat control.

2.1 to reduce the power switch of the heat

High-frequency switching power supply switch is in heat one of the larger devices, reducing its heat can not only

improve the reliability of the switch itself, but also can reduce the machine temperature, increase the overall

efficiency and MTBF (MTBF). Switch during normal operation, was opened, off two states, the resulting loss can be

broken down into two critical state generated on-state losses and the resulting loss. One on-state losses by the

switch on-state resistance of the decision itself. You can choose a low on-state resistance of the switch to minimize

this loss. The MOSFET on-resistance than the IGBT s large, but it s high frequency, thus switching power supply design

is still the preferred device. Now IRL3713 IR s new series of HEXFET (hexagonal field-effect transistor) power MOSFET

on-resistance has been done 3mΩ, so that these devices have lower conduction loss, gate charge and switching losses.

APT U.S. companies have similar products. Turn on and off two kinds of critical state by selecting the switching loss

can also be faster, shorter recovery time to reduce the device. But more important is through the design of better

control and buffering techniques to reduce losses, higher switching frequency, this method can reflect the advantage

of more time to. Such as various soft-switching technology, allowing the switch at zero voltage and zero-current state

of open or off, thus greatly reducing the losses generated by these two states. Some manufacturers still use from a

cost to consider hard-switching technology, you can reduce all types of buffering techniques to switch the loss and

improve its reliability.

2.2 to reduce the power diode heat

High-frequency switching power supply, power diodes have many applications, the selected species are also

different. For 50Hz AC input into DC power rectifier diode and snubber circuit in the fast recovery diode, under

normal circumstances not have better control technology to reduce the loss, by selecting only high-quality devices,

such as the use of on-press Schottky diode drop lower or faster turn-off and soft recovery ultrafast recovery diode to

reduce losses, reduce the heat. High-frequency transformer secondary side of the rectifier circuit synchronous

rectification can also be used to further reduce heat loss and the rectifier voltage drop, but they will increase

costs. Therefore, manufacturers how to master the balance between performance and cost, to achieve cost-effective is a

very worthwhile study.

2.3 to reduce the high frequency transformer and filter inductor and other magnetic components of the heat

High-frequency switching power supply indispensable to apply a variety of magnetic components, such as filter

chokes, filter inductance energy storage, there are high-frequency isolated power transformer. Their work will produce

more or less copper loss, iron loss, these methods of heat loss comes out. Especially inductors and transformers, high

frequency current in the coil current due to skin effect, copper loss will increase exponentially, so that the

inductor, transformer losses resulting part can not be neglected. Therefore, to be used in the design of parallel

multi-strand thin enameled wire wound, or the use of wide and thin copper winding to reduce the impact of skin effect.

Generally used in high-quality ferrite core material, such as Japan s TDK magnetic materials production. Models of

choice to leave a margin to prevent magnetic saturation.

2.4 to reduce the heat load off

High-power switching power supply in order to avoid no-load state of the voltage rise, often with dummy load –

high power resistors, power supply with active PFC units especially. Switching power supply work, leave a small amount

of current through the load, will not only reduce the efficiency of switching power supply, and its heat is also a

factor affecting the thermal stability of the machine. Dummy load in the printed circuit board (PCB) are often

associated with the position on the use of electrolytic capacitor output filter are very close, while electrolytic

capacitors are extremely sensitive to temperature. Therefore it is necessary to reduce the false heat load. More

feasible approach is to leave the variable impedance load designed way. Through the switching power supply output

current to control the false detection of the size of the load impedance, when the power is in the normal load, exit

load consumption current state false; no-load, off load consumption current maximum. This will not affect the

stability of the no-load power, it will not reduce the power efficiency and generate a lot of unnecessary calories.

3 Thermal Design

3.1 The basic method of heat calculation

There are three basic ways heat: heat conduction, convection and thermal radiation.

1) heat conduction through direct contact with the object or objects within the heat transfer between the various

parts that are heat conduction. The mechanism is different from the temperature of the object or objects at different

temperatures between the various parts of the molecular kinetic energy with each other. Thermal conductivity and the

current concept is very similar, always from a high temperature heat transfer to the place where the temperature is

low, heat transfer in the process of thermal resistance as there is current flow in the process of the same

resistance. The heat flux Φ = [W], where Rt is the thermal resistance, τ is the temperature difference. The thermal

resistance Rt = [K / W], where δ is the conductor thickness, λ is the thermal conductivity, A is the conductor

cross-sectional area. Thus, in the switching power supply design, by the heat dissipation power source, find the

temperature rise τ = ΦRt. As the actual application, the heat flux from the heat starting to reach the radiator tend

to go through several different materials, heat conductor, namely, the existence of different resistance in series, in

the calculation, the total thermal resistance and thermal resistance for more than one.

2) Convection heat transfer through the heat the way it passed close together with the fluid layer, this layer of

fluid heated, volume expansion, density, smaller, upward mobility, the density of the surrounding fluid flow over the

filling, fill over expansion of the fluid absorbs heat upward mobility, and so on, continually remove heat from the

hot surface components, a process known as convection. The calculation of convective heat transfer is generally made

by using Newton s formula: Φ = αA (θ1-θ2) [W], where A is the fluid in contact with the wall area [m2], α for the

convective heat transfer coefficient, θ1 is the wall temperature [ K], θ2 is the average fluid temperature [K].

Thus, the heat flux Φ and the convective heat transfer coefficient α, cross-sectional area A and the solid surface

and fluid temperature difference (θ1-θ2) proportional to the product. Convection heat transfer is a complex process,

it is not only determined by the thermal process, but also depends on the gas dynamics. Simply put, the factors that

affect the convective heat transfer has two aspects: (1) the fluid physical properties, such as density, viscosity,

expansion coefficient, thermal conductivity, specific heat, etc.; (2) fluid flow, is a natural convection or forced

convection, laminar or turbulent flow is. Because the laminar flow, heat transfer mainly depends on the flow layer

between disparate thermal conductivity; and turbulent flow, the flow in the layer close to the bottom wall, the fluid

to produce vortex enhanced heat transfer effect. In general, other conditions being equal, turbulence of the laminar

flow heat transfer coefficient of heat transfer coefficient than the good times, and even more.

3) As the temperature difference caused by the thermal radiation propagation of electromagnetic waves known as

thermal radiation. It is the process of heat conduction and convection heat transfer is much more complex. It is part

of the object s thermal energy into electromagnetic energy, electromagnetic waves can pass through the media such as

air, vacuum, etc., spread out to four weeks, when faced with other objects, then is absorbed and then converted into

heat, the rest of the is reflected back. Distributed by a variety of objects, the infrared, that is, a thermal

radiation. In a vacuum or air, the radiation radiated objects Φ, depends on the nature of objects, surface conditions

(such as color, roughness, etc.), size and surface area of ​​the surface temperature. Φ = εσbA (T14-T24) which σb is

the Boltzmann constant, a value of 5.67 × 10-8, A is the radiation surface area [m2], T is the absolute temperature

of two surfaces [K], ε is the surface blackness . Surface darker, rougher, more intense radiation.

3.2 Switching Power Supply in the main heat source of heating mode

Switching power supply of the heat source, such as the rectifier bridge, the power switch, fast recovery diodes,

magnetic components as well as high-power dummy load resistance, etc., the heat generated by these components must try

to distribute out, generally used in thermal design of heat mainly through heat conduction and convection heat

transfer. That all components are first fixed to the heat radiator, the heat transfer by thermal conduction to the

radiator, radiator heat on the way through convection from the air out of chassis. The actual thermal conditions for

the three heat transfer of the integrated, you can use Newton s formula to unify the expression: Φ = KSτ, where S is

the heat dissipation surface area, K is the surface thermal coefficient. Surface heat coefficient is usually

determined by the test, in general engineering fluid mechanics data are available. It is the unity of all three forms

of heat up.

By Φ = KSτ, we can calculate the power dissipation in the future, depending on the allowable temperature rise τ

to determine the heat dissipation surface area S, and thus determine the radiator to be used. This calculation for

improving the reliability of switching power supply, power density, cost and so is of great significance. In a

considerable number of cases, the power module manufacturers to reduce costs, often using a common type of heat sink,

these radiators are not designed to be very appropriate. Specific requirements for communication with high reliability

for high-frequency switching power supply, specifically targeted to the heat sink design becomes very important. For

example, a New Zealand power system for communication rectifier modules Intergy R2948 (48V/60A) single module output

power of 2900W, its use of air cooling for the forward wind, the wind on the ramp way, the radiator is designed. Its

most prominent feature is the heat sink on the heat sink showed a certain angle, the air flow can be oriented

obliquely, the flow of hot air from bottom to top in line with the physical characteristics of flow, so that under the

same thermal power can reduce the air flow requirements. Meanwhile, cast aluminum heat sink for the matte appearance,

surface roughness, this shape in the bottom of the air flow rate, flow more easily into a turbulent layer, thereby

increasing the heat transfer coefficient. Combination of these two features, can greatly improve the radiator s

cooling efficiency, resulting in the same power output, and other external conditions, reducing the required fan

speed, if we take the size of the fan with variable speed power output can be increase fan life. The rectifier module,

the fan MTBF is the lowest of all the components, rectifier module has been improved MTBF restricting the bottleneck,

so take various measures to improve the cooling efficiency to extend fan life on a very positive meaning. Original

Huawei Electric Company, now part of Emerson Network Power products have a similar design, indicating that this design

approach is being used more and more power supply manufacturers. Because of this need for custom-made radiator,

according to user requirements mold, so the high cost of some, but to improve the reliability of power supply is still

quite useful.

4 Conclusion

In summary, the high-frequency switching power supply of thermal design, the need to consider two aspects of

heating and cooling conditions, lower priority to the use of various types of heating technology, while improving the

whole, especially the radiator cooling efficiency. This design idea from some manufacturers of power supply module has

been verified, is engaged in power supply design provides a design method that can learn.

1 Introduction

Working temperature of electronic products have strict requirements. Power supply internal temperature too high

will result in temperature-sensitive semiconductor devices, electrolytic capacitors and other components fail. When

the temperature exceeds a certain value, the failure rate increases exponentially. Some statistics show that the

electronic component temperature is increased by 2 ℃, reliability, decreased by 10%; temperature 50 ℃, the life span

of only when the temperature is 25 ℃, 1 / 6. Therefore, control of electronic devices will encounter throughout the

chassis and internal component temperature requirements, and this is the thermal design of electronic equipment. The

high-frequency switching power supply with high power heating devices of this type of equipment, the temperature is

affecting their reliability and the most important factor, for the thermal design of the whole, there are strict

requirements. Complete thermal design includes two aspects: how to control the heat source of heat; how to heat the

heat spread it out. The ultimate goal is to reach thermal equilibrium after the temperature control of electronic

equipment within the allowable range.

2 heat control design

Switching power supply in the major heat components for the semiconductor switch (eg, MOSFET, IGBT, GTR, SCR,

etc.), high-power diode (such as ultra-fast recovery diode, Schottky diode, etc.), high frequency transformers, filter

inductors and other magnetic components and false loads. For each type of heating components have different methods of

heat control.

2.1 to reduce the power switch of the heat

High-frequency switching power supply switch is in heat one of the larger devices, reducing its heat can not only

improve the reliability of the switch itself, but also can reduce the machine temperature, increase the overall

efficiency and MTBF (MTBF). Switch during normal operation, was opened, off two states, the resulting loss can be

broken down into two critical state generated on-state losses and the resulting loss. One on-state losses by the

switch on-state resistance of the decision itself. You can choose a low on-state resistance of the switch to minimize

this loss. The MOSFET on-resistance than the IGBT s large, but it s high frequency, thus switching power supply design

is still the preferred device. Now IRL3713 IR s new series of HEXFET (hexagonal field-effect transistor) power MOSFET

on-resistance has been done 3mΩ, so that these devices have lower conduction loss, gate charge and switching losses.

APT U.S. companies have similar products. Turn on and off two kinds of critical state by selecting the switching loss

can also be faster, shorter recovery time to reduce the device. But more important is through the design of better

control and buffering techniques to reduce losses, higher switching frequency, this method can reflect the advantage

of more time to. Such as various soft-switching technology, allowing the switch at zero voltage and zero-current state

of open or off, thus greatly reducing the losses generated by these two states. Some manufacturers still use from a

cost to consider hard-switching technology, you can reduce all types of buffering techniques to switch the loss and

improve its reliability.

2.2 to reduce the power diode heat

High-frequency switching power supply, power diodes have many applications, the selected species are also

different. For 50Hz AC input into DC power rectifier diode and snubber circuit in the fast recovery diode, under

normal circumstances not have better control technology to reduce the loss, by selecting only high-quality devices,

such as the use of on-press Schottky diode drop lower or faster turn-off and soft recovery ultrafast recovery diode to

reduce losses, reduce the heat. High-frequency transformer secondary side of the rectifier circuit synchronous

rectification can also be used to further reduce heat loss and the rectifier voltage drop, but they will increase

costs. Therefore, manufacturers how to master the balance between performance and cost, to achieve cost-effective is a

very worthwhile study.

2.3 to reduce the high frequency transformer and filter inductor and other magnetic components of the heat

High-frequency switching power supply indispensable to apply a variety of magnetic components, such as filter

chokes, filter inductance energy storage, there are high-frequency isolated power transformer. Their work will produce

more or less copper loss, iron loss, these methods of heat loss comes out. Especially inductors and transformers, high

frequency current in the coil current due to skin effect, copper loss will increase exponentially, so that the

inductor, transformer losses resulting part can not be neglected. Therefore, to be used in the design of parallel

multi-strand thin enameled wire wound, or the use of wide and thin copper winding to reduce the impact of skin effect.

Generally used in high-quality ferrite core material, such as Japan s TDK magnetic materials production. Models of

choice to leave a margin to prevent magnetic saturation.

2.4 to reduce the heat load off

High-power switching power supply in order to avoid no-load state of the voltage rise, often with dummy load –

high power resistors, power supply with active PFC units especially. Switching power supply work, leave a small amount

of current through the load, will not only reduce the efficiency of switching power supply, and its heat is also a

factor affecting the thermal stability of the machine. Dummy load in the printed circuit board (PCB) are often

associated with the position on the use of electrolytic capacitor output filter are very close, while electrolytic

capacitors are extremely sensitive to temperature. Therefore it is necessary to reduce the false heat load. More

feasible approach is to leave the variable impedance load designed way. Through the switching power supply output

current to control the false detection of the size of the load impedance, when the power is in the normal load, exit

load consumption current state false; no-load, off load consumption current maximum. This will not affect the

stability of the no-load power, it will not reduce the power efficiency and generate a lot of unnecessary calories.

3 Thermal Design

3.1 The basic method of heat calculation

There are three basic ways heat: heat conduction, convection and thermal radiation.

1) heat conduction through direct contact with the object or objects within the heat transfer between the various

parts that are heat conduction. The mechanism is different from the temperature of the object or objects at different

temperatures between the various parts of the molecular kinetic energy with each other. Thermal conductivity and the

current concept is very similar, always from a high temperature heat transfer to the place where the temperature is

low, heat transfer in the process of thermal resistance as there is current flow in the process of the same

resistance. The heat flux Φ = [W], where Rt is the thermal resistance, τ is the temperature difference. The thermal

resistance Rt = [K / W], where δ is the conductor thickness, λ is the thermal conductivity, A is the conductor

cross-sectional area. Thus, in the switching power supply design, by the heat dissipation power source, find the

temperature rise τ = ΦRt. As the actual application, the heat flux from the heat starting to reach the radiator tend

to go through several different materials, heat conductor, namely, the existence of different resistance in series, in

the calculation, the total thermal resistance and thermal resistance for more than one.

2) Convection heat transfer through the heat the way it passed close together with the fluid layer, this layer of

fluid heated, volume expansion, density, smaller, upward mobility, the density of the surrounding fluid flow over the

filling, fill over expansion of the fluid absorbs heat upward mobility, and so on, continually remove heat from the

hot surface components, a process known as convection. The calculation of convective heat transfer is generally made

by using Newton s formula: Φ = αA (θ1-θ2) [W], where A is the fluid in contact with the wall area [m2], α for the

convective heat transfer coefficient, θ1 is the wall temperature [ K], θ2 is the average fluid temperature [K].

Thus, the heat flux Φ and the convective heat transfer coefficient α, cross-sectional area A and the solid surface

and fluid temperature difference (θ1-θ2) proportional to the product. Convection heat transfer is a complex process,

it is not only determined by the thermal process, but also depends on the gas dynamics. Simply put, the factors that

affect the convective heat transfer has two aspects: (1) the fluid physical properties, such as density, viscosity,

expansion coefficient, thermal conductivity, specific heat, etc.; (2) fluid flow, is a natural convection or forced

convection, laminar or turbulent flow is. Because the laminar flow, heat transfer mainly depends on the flow layer

between disparate thermal conductivity; and turbulent flow, the flow in the layer close to the bottom wall, the fluid

to produce vortex enhanced heat transfer effect. In general, other conditions being equal, turbulence of the laminar

flow heat transfer coefficient of heat transfer coefficient than the good times, and even more.

3) As the temperature difference caused by the thermal radiation propagation of electromagnetic waves known as

thermal radiation. It is the process of heat conduction and convection heat transfer is much more complex. It is part

of the object s thermal energy into electromagnetic energy, electromagnetic waves can pass through the media such as

air, vacuum, etc., spread out to four weeks, when faced with other objects, then is absorbed and then converted into

heat, the rest of the is reflected back. Distributed by a variety of objects, the infrared, that is, a thermal

radiation. In a vacuum or air, the radiation radiated objects Φ, depends on the nature of objects, surface conditions

(such as color, roughness, etc.), size and surface area of ​​the surface temperature. Φ = εσbA (T14-T24) which σb is

the Boltzmann constant, a value of 5.67 × 10-8, A is the radiation surface area [m2], T is the absolute temperature

of two surfaces [K], ε is the surface blackness . Surface darker, rougher, more intense radiation.

3.2 Switching Power Supply in the main heat source of heating mode

Switching power supply of the heat source, such as the rectifier bridge, the power switch, fast recovery diodes,

magnetic components as well as high-power dummy load resistance, etc., the heat generated by these components must try

to distribute out, generally used in thermal design of heat mainly through heat conduction and convection heat

transfer. That all components are first fixed to the heat radiator, the heat transfer by thermal conduction to the

radiator, radiator heat on the way through convection from the air out of chassis. The actual thermal conditions for

the three heat transfer of the integrated, you can use Newton s formula to unify the expression: Φ = KSτ, where S is

the heat dissipation surface area, K is the surface thermal coefficient. Surface heat coefficient is usually

determined by the test, in general engineering fluid mechanics data are available. It is the unity of all three forms

of heat up.

By Φ = KSτ, we can calculate the power dissipation in the future, depending on the allowable temperature rise τ

to determine the heat dissipation surface area S, and thus determine the radiator to be used. This calculation for

improving the reliability of switching power supply, power density, cost and so is of great significance. In a

considerable number of cases, the power module manufacturers to reduce costs, often using a common type of heat sink,

these radiators are not designed to be very appropriate. Specific requirements for communication with high reliability

for high-frequency switching power supply, specifically targeted to the heat sink design becomes very important. For

example, a New Zealand power system for communication rectifier modules Intergy R2948 (48V/60A) single module output

power of 2900W, its use of air cooling for the forward wind, the wind on the ramp way, the radiator is designed. Its

most prominent feature is the heat sink on the heat sink showed a certain angle, the air flow can be oriented

obliquely, the flow of hot air from bottom to top in line with the physical characteristics of flow, so that under the

same thermal power can reduce the air flow requirements. Meanwhile, cast aluminum heat sink for the matte appearance,

surface roughness, this shape in the bottom of the air flow rate, flow more easily into a turbulent layer, thereby

increasing the heat transfer coefficient. Combination of these two features, can greatly improve the radiator s

cooling efficiency, resulting in the same power output, and other external conditions, reducing the required fan

speed, if we take the size of the fan with variable speed power output can be increase fan life. The rectifier module,

the fan MTBF is the lowest of all the components, rectifier module has been improved MTBF restricting the bottleneck,

so take various measures to improve the cooling efficiency to extend fan life on a very positive meaning. Original

Huawei Electric Company, now part of Emerson Network Power products have a similar design, indicating that this design

approach is being used more and more power supply manufacturers. Because of this need for custom-made radiator,

according to user requirements mold, so the high cost of some, but to improve the reliability of power supply is still

quite useful.

4 Conclusion

In summary, the high-frequency switching power supply of thermal design, the need to consider two aspects of

heating and cooling conditions, lower priority to the use of various types of heating technology, while improving the

whole, especially the radiator cooling efficiency. This design idea from some manufacturers of power supply module has

been verified, is engaged in power supply design provides a design method that can learn.