Design Study of a Future 10kW Motor Controller

Full text

(1)LiU-ITN-TEK-G--08/036--SE. Designstudie av framtida 10kW växelriktare Daniel Eidborn 2008-06-17. Department of Science and Technology Linköping University SE-601 74 Norrköping, Sweden. Institutionen för teknik och naturvetenskap Linköpings Universitet 601 74 Norrköping.

(2) LiU-ITN-TEK-G--08/036--SE. Designstudie av framtida 10kW växelriktare Examensarbete utfört i elektriska drivsystem vid Tekniska Högskolan vid Linköpings universitet. Daniel Eidborn Handledare Eduardo Figueroa Karlström Examinator Carl-Magnus Erzell Norrköping 2008-06-17.

(3) Upphovsrätt Detta dokument hålls tillgängligt på Internet – eller dess framtida ersättare – under en längre tid från publiceringsdatum under förutsättning att inga extraordinära omständigheter uppstår. Tillgång till dokumentet innebär tillstånd för var och en att läsa, ladda ner, skriva ut enstaka kopior för enskilt bruk och att använda det oförändrat för ickekommersiell forskning och för undervisning. Överföring av upphovsrätten vid en senare tidpunkt kan inte upphäva detta tillstånd. All annan användning av dokumentet kräver upphovsmannens medgivande. För att garantera äktheten, säkerheten och tillgängligheten finns det lösningar av teknisk och administrativ art. Upphovsmannens ideella rätt innefattar rätt att bli nämnd som upphovsman i den omfattning som god sed kräver vid användning av dokumentet på ovan beskrivna sätt samt skydd mot att dokumentet ändras eller presenteras i sådan form eller i sådant sammanhang som är kränkande för upphovsmannens litterära eller konstnärliga anseende eller egenart. För ytterligare information om Linköping University Electronic Press se förlagets hemsida http://www.ep.liu.se/ Copyright The publishers will keep this document online on the Internet - or its possible replacement - for a considerable time from the date of publication barring exceptional circumstances. The online availability of the document implies a permanent permission for anyone to read, to download, to print out single copies for your own use and to use it unchanged for any non-commercial research and educational purpose. Subsequent transfers of copyright cannot revoke this permission. All other uses of the document are conditional on the consent of the copyright owner. The publisher has taken technical and administrative measures to assure authenticity, security and accessibility. According to intellectual property law the author has the right to be mentioned when his/her work is accessed as described above and to be protected against infringement. For additional information about the Linköping University Electronic Press and its procedures for publication and for assurance of document integrity, please refer to its WWW home page: http://www.ep.liu.se/. © Daniel Eidborn.

(4) Abstract This work has the approach for how an electric motor controller should be designed. In aircraft applications it is important that the system has a high power density, and that it is reliable. The target was to find out what is possible with technology of today, and with possibilities of tomorrow. The target is to be able to compare hydraulic systems with electrical ones. The type of controllers that was studied was controllers for permanent magnetized synchronous machines (PMSM). The reason for that choice is that PMSM has a high efficiency. Different transistor technologies were evaluated. Discrete IGBT was found to be the best option. Of the evaluated transistors has IRG4PSH71U operating with a SiC freewheeling diode the best efficiency. The benefit with discrete components is that they are easy to cool, for example if they are distributed on an aluminium heatsink with forced air cooling. To minimise losses and gain controllability on the motor should the inverter be controlled with some kind of vector control, such as DTC (Direct Torque Control). Keywords: Inverter, IGBT, Airplane, switching. Sammanfattning I det här arbetet studerades hur en elmotorstyrning i ett flygplan bör konstrueras. I flygapplikationer är det viktigt att systemet är lätt i förhållande till effekten, och att det är tillförlitligt. Målet var att ta reda på vad man kan uppnå med dagens teknik, och även med kommande teknik. Syftet är att kunna jämföra hydrauliska system med elektriska. Det som undersöktes var motorstyrningar till permanentmagnetiserade synkronmaskiner (PMSM), eftersom dessa motorer har en hög verkningsgrad. Olika transistortekniker utvärderades. Diskreta IGBT fanns vara det bästa valet i en sådan applikation. Av de transistorer som jämfördes har IRG4PSH71U tillsammans med en frihjulsdiod av SiC den bästa totala verkningsgraden. Det vill säga att summan av ledningsförluster och switchförluster är lägst för den kombinationen. Fördelen med diskreta transistorer är att de har en lägre termisk resistans och kan spridas på en kylfläns, vilket underlättar kylningen, t.ex. med en aluminiumkylfläns med forcerad luft kylning. För att minimera förluster och öka styrbarheten på motorn bör växelriktaren styras med någon typ av vektor kontroll exempelvis DTC (Direct Torque Control).. Nyckelord: Växelriktare, Frekvensomriktare, IGBT, Flygplan.

(5) Acknowledgements This report is the result of my degree project. The project was carried out at SAAB Aerosystems in Linköping with supervision of ITN on Linkoping’s University. This work however would not be possible without the help and support from many other persons. I would like to thank SAAB for the opportunity to make this degree project. The time at SAAB Areosystems has been a time when I have learned a lot about the work in the aviation industry. The TDGE department has been a good place to do the work at. I would like to thank Lars Austin for the supervision and the initiative to the work, and wish him the best for the future. I also want to thank Eduardo Figueroa for many tips along the way, both for this work and for my life. I would like to thank Andreas Johansson for tips along the writing, it has been a good help. My co worker Sebastian Fant also deserves thanks for rewarding discussions. I also want to thank my examiner Carl-Magnus Erzell and Lars Backström at ITN, for the help to reach this point in my education. I also want to thank my wife Johanna for the support during my time on the university. There are a lot of other people that has been important for me and my work, and those also deserve thanks. Linköping, June 2008 Daniel Eidborn.

(6) Contents Abbreviations ......................................................................................................................1 List of Symbols....................................................................................................................1 List of Figures......................................................................................................................2 List of tables ........................................................................................................................2 List of equations ..................................................................................................................2 1. Introduction .................................................................................................................3 1.1. Objective..............................................................................................................3 1.2. Background..........................................................................................................3 1.3. Research method..................................................................................................3 1.4. Delimitations .......................................................................................................3 2. Theory..........................................................................................................................4 2.1. Inverter Main Parts ..................................................................................................4 3. Components .................................................................................................................6 3.1. Transistors ...........................................................................................................6 3.1.1. Bipolar transistors........................................................................................6 3.1.2. FET ..............................................................................................................6 3.1.3. IGBT............................................................................................................7 3.1.4. Silicon Carbide Transistors .........................................................................7 3.2. Free Wheeling Diodes .........................................................................................7 3.2.1. Silicon Diodes .............................................................................................8 3.2.2. Silicon Carbide Diodes................................................................................8 3.3. Rectifier Diodes...................................................................................................8 3.3.1. Silicon Diodes .............................................................................................8 3.3.2. Silicon Carbide Diodes................................................................................8 3.3.3. Transistors ...................................................................................................8 4. Switching Strategy.......................................................................................................9 4.1. Scalar Control ......................................................................................................9 4.2. Direct Torque Control .........................................................................................9 5. Control and Driver Design ........................................................................................11 5.1. Bootstrap............................................................................................................11 5.1.1. Bootstrap Capacitor .......................................................................................12 5.1.2. Bootstrap Diode.........................................................................................12 5.2. Control Systems.................................................................................................12 5.3. Power Supply.....................................................................................................13 5.3.1. Control voltages.........................................................................................13 6. Losses and cooling ....................................................................................................15 6.1. Losses ................................................................................................................15 6.1.1. Rectifier .....................................................................................................15 6.1.2. Driver, Control and Internal Power Supply...............................................15 6.1.3. Transistor Bridge .......................................................................................15 6.2. Cooling ..............................................................................................................16 6.2.1. Thermal Resistance ...................................................................................16 6.2.2. Spreading...................................................................................................17.

(7) 6.2.3. Heatsink .....................................................................................................17 6.3. Manufacturing Methods for Heatsinks ..............................................................18 6.3.1. Casting.......................................................................................................18 6.3.2. Extrusion....................................................................................................18 6.3.3. Punch .........................................................................................................18 6.3.4. Milled ........................................................................................................18 6.4. Material..............................................................................................................19 6.4.1. Aluminum..................................................................................................19 6.4.2. Copper .......................................................................................................19 6.4.3. Zinc............................................................................................................19 6.5. Heat pipes ..........................................................................................................19 7. Chassis Design...........................................................................................................21 7.1. Cooling Scenarios..............................................................................................21 7.1.1. JAS 39 Gripen ...........................................................................................21 7.1.2. MERA-Windex..........................................................................................22 8. Results .......................................................................................................................23 8.1. Transistors .........................................................................................................23 8.1.1. Transistor Bridges .....................................................................................23 8.1.2. Discrete Transistors ...................................................................................23 8.1.3. Selection ....................................................................................................23 8.2. Freewheeling Diodes .........................................................................................25 8.3. Switching strategy .............................................................................................25 8.4. Control and driver circuits.................................................................................25 8.5. Cooling ..............................................................................................................25 8.6. Reliability ..........................................................................................................26 8.7. Estimations ........................................................................................................26 8.8. Inverter layout ...................................................................................................27 9. Discussion and Conclusions ......................................................................................29 10. Future Work...........................................................................................................29.

(8) Abbreviations AC Alternating Current DC Direct Current DTC Direct Torque Control EMI Electro Magnetic interference FET Field Effect Transistor IC Integrated Circuit IGBT Insulated Gate Bipolar Transistor PM Permanent Magnet PMSM Permanent Magnet Synchronic Motor PWM Pulse Width Modulation SiC Silicon Carbide. List of Symbols VPh VDC RDSon UF Î VGE VCE VAC TJ TC RΘJC RΘCH RΘHA TAMB TH PON Pton Ptot Ptoff Pf. Volt, Phase Volt Direct-Current Resistance Darin to Source with transistor open Forward voltage drop Top current Volt, Gate to Emitter Volt, Collector to Emitter Volt Alternating-Current Temperature, transistor Junction Temperature, transistor Case Thermal Resistance, Junction to Case Thermal Resistance, Case to Heatsink Thermal Resistance, Heatsink to Ambient Temperature, Ambient Temperature, Heatsink Power loss, conduction Power loss, turn on Power loss, total Power loss, turn off Power loss, freewheeling diode.

(9) List of Figures 2.1 3.1 3.2 4.1 4.2 5.1 5.2 5.3 6.1 6.5 7.1 8.1 8.7. Inverter overview from IRMCK341 datasheet. FET symbol IGBT symbol Scalar control stator current. [20] Direct Torque Control stator current. [20] Bootstrap transistor driver. [15] Block diagram of motor controller IRMCF341from data sheet no: PD60304 Typical fly back application. From data sheet TOPSwitch-II family [17] Thermal resistance model. From Wikipedia Heat pipe mechanisms. From Wikipedia Air requirements transistor linear model from Hitachi IGBT Module Application Manual [18] Inverter layout.. List of tables 6.1 8.1 8.2 8.3 8.4. Comparison of heatsink materials Loss estimation Weight estimation Price estimation Transistor compare. List of equations 8.1 8.2 8.3 8.4 8.5. Transistor on loss Transistor turn on loss Transistor turn off loss Transistor total loss Freewheeling diode loss. 2.

(10) 1. Introduction 1.1.. Objective. Saab Aerosystems aim is to gather a deeper knowledge about the technologies that are possible with modern power electronics, both with present and upcoming technologies. The matters of primary interest to assess are power density, efficiency, reliability, cooling needs and cost. This work concentrates on a 10 kW inverter intended for a general motor supply. The power supply is a 400 Hz, 230 VPh three phases or a ± 270 V DC source. Other power levels than 10 kW are also of interest. The things that are interesting to gain technical insight into are the most suitable components available for inverters, the way to use them, their expected efficiency and weight of the implemented solutions.. 1.2.. Background. It has always been a challenge to make aircrafts as light as possible. To day the goal is also to reduce the oil consumption. Therefore, much of engineering efforts are allocated to develop lighter and more fuel efficient aircrafts. At the same time, new and more advanced technical solutions with modern power electronics have been developed. SAAB Aerosystems wants to gain further knowledge on the possibilities and limitations with electrical systems in places where there are hydraulic or mechanical systems in use. This is part of the aims of SAAB to continue to be in the forefront of aircraft system developments that make their solutions to stay among the best on the market.. 1.3.. Research method. The research tools used consisted mainly of literature studies. The literature used was of different kinds, mainly articles, patents, datasheets, etc.. 1.4.. Delimitations. This work was not intended to cover any computer simulations or hardware models of an inverter. Neither was it aimed to consider EMC (Electromagnetic Compatibility) or EMI (Electromagnetic Interference) problems and their solutions. However a general overview of the market for IGBT components, the technical features, and a general concept for a controller are discussed.. 3.

(11) 2. Theory To control an AC motor an inverter is needed. The inverter is a device that generates an AC current with a variable frequency. The inverter converts a DC or AC source to a current that the motor needs for the moment.. 2.1.. Inverter Main Parts. The inverter can be described in terms of different blocks. From the power source to the inverter output the system can be described with the following blocks. (Left to right in figure 2.1). • The Power input First in the figure is the power input. An inverter can be supplied with different kind of voltages. In figure 2.1 the supply is an AC voltage. In this work the supply will be a three phase AC voltage.. • The Filter The filter is aimed mainly to protect the electrical environment on the supply grid from EMI disorders from the inverter. These disturbances can for instance be transients or harmonics leading to interferences that can result in problem in other electrical systems in the airplane.. • The Rectifier The diode bridge rectifies the input AC voltage to a DC voltage. Figure 2.1 shows a one phase rectifier, but in this study three phases are used.. • The Capacitors The capacitors are there to handle current transients and to keep the voltage relatively constant with a low ripple level. The capacitors act as an energy buffer between the AC source and the AC output.. • Control circuit To operate the inverter, a control circuit of some kind is needed. The control circuit has sensors for motor currents, voltages, position, etc. With the knowledge of these signals, the controller is able to accomplish a suitable switching pattern for the transistors. The control circuit opens the transistors in a suitable way to control the motor. In figure 2.1 the control circuit is represented with several different blocks.. 4.

(12) • Transistor bridge The transistor bridge is the main power block in the inverter. The transistors are switching the voltage to the motor windings. The bridge also includes freewheeling diodes which are there to handle the inductive currents arising in the reverse direction.. • Motor The motor used is a Permanent Magnet Synchronous Machine (PMSM). PMSM is the motor type that shows best efficiency. It is also controllable and reliable. The motor has no brushes that need maintenance or produce dust.. Figure 2.1. The figure shows a schematic Inverter circuit overview.. 5.

(13) 3. Components 3.1.. Transistors. There are a number of different transistors based on different technologies. The transistors considered here are those that have been designed for voltages and currents that make them a possible option for this application. For instance, Fuji IGBT modules application manual suggest a 1200V rated IGBT for a line in voltage of 400V AC. [4]. 3.1.1. Bipolar transistors The bipolar transistor is the most basic transistor type. The low gain on transistors for high currents results in a demand for a powerful driver circuit. This makes them unsuitable for power applications.. 3.1.2. FET FET (Field Effect Transistor) is a family of transistors components mainly developed for their use in power applications. A FET has a fast turn on-off characteristic that makes them suitable for switching applications. They have no forward voltage drop. Transistors for low voltages have a very low internal resistance. A FET is basically controlled with a voltage, when the gate capacitance is charged is no gate current needed to keep the voltage. Normally, a gate-emitter voltage of 15 VDC is used to control the transistor. This makes FET a perfect transistor technology for low voltages. Transistors designed for high voltages shows a forward resistance that is too high to make them an alternative. For example IRFSL3206PBF from IRF has an RDSon = 0.003 Ω but handle max 60 V. For transistors that handle higher voltages the resistance is higher. For transistors that handle 600 V with sufficient margin the RDSon is too high to be an alternative.. Figure 3.1 A schematic diagram of a FET transistor.. 6.

(14) 3.1.3. IGBT IGBT (Insulated Gate Bipolar Transistor) is a transistor technology that can handle higher voltages than FET. An IGBT can be seen as a bipolar transistor Darlington coupled with a FET on the base. The transistor can be controlled in the same way as a FET. An IGBT has a forward voltage drop like a bipolar transistor. An IGBT has a low forward conducting resistance, and is capable to handle high voltages. The switch time for an IGBT is longer than for a FET, furthermore the IGBT shows a current tail after switching off. This makes the switching losses higher for an IGBT than for a FET, but still the total losses, including conducting losses, are significantly lower for an IGBT than for transistors designed for these voltages. [13]. Figure 3.2 A schematic diagram of an IGBT transistor.. 3.1.4. Silicon Carbide Transistors Silicon carbides are materials exhibiting very good high temperature performance and therefore they have been in the focus for power semiconductor applications. The main difficulty in the development of silicon carbide components has been related to the how to make good quality components substrates. Today however there are a couple of manufacturers that makes commercially available SiC substrates. There is a fast growing quality improvement of SiC based components. There are a couple of companies that work with development of SiC transistors. SiC transistors are only commercially available for relatively low currents applications yet, and they are still most for evaluation purpose. SiC transistors are able to handle higher temperatures than Si. They will also be easier to use in parallel coupling, because of their negative temperature gain ratio, and positive temperature/resistance ratio. SiC will most likely be an important competitor to Si based IGBT in the future.. 3.2.. Free Wheeling Diodes. The freewheeling diodes conduct when the induction in the motor coil continues to force a current while the transistor is not conducting. If there is no diode to conduct this current there will be a high voltage peak, which would most probably destroy the inverter. The freewheeling diode has to be able to conduct the same current as the inverter transistors conduct.. 7.

(15) 3.2.1. Silicon Diodes Silicon is the most common substrate for diodes. Si diodes are cheaper than SiC diodes. For switching applications, Schottky diodes are the preferable option since they have a faster reverse recovery characteristic than ordinary diodes.. 3.2.2. Silicon Carbide Diodes A Silicon Carbide Schottky diode does not have a reverse recovery charge. This results in decreased switching losses in the diode. It also decreases the switching losses in the IGBT [1], [6], [7]. However, on the negative side it should be mentioned that SiC Schottky diodes are more expensive than a Si diode, furthermore, their conduction losses are larger, but the total losses in the inverter can be lowered with SiC diodes. The price for SiC components will almost certainly decrease and their performance will increase, which will make them an even better option in the future. The diodes can be separate discrete components or integrated in the same capsule as the transistor. A capsule with both transistor and diode is marketed as a CoPack. Discrete diodes however offer the possibility of choosing them freely and thereby it can be gained some flexibility in design. 3.3.. Rectifier Diodes. 3.3.1. Silicon Diodes Silicon rectifier diodes have a voltage drop around 1 V while operated at 25 °C and their nominal current. The input frequency in the studied example is 400 Hz in to the rectifier, and therefore at this low frequency reverse recovery is not a problem.. 3.3.2. Silicon Carbide Diodes The SiC diodes that are available today have a too low current rating to be a good option. The forward voltage drop is also higher than the one a Si diode has, and the price is higher. The better recovery performance is not a big advantage because of the low frequency. In the future, as the technology is developed is it possible that SiC will become a better option.. 3.3.3. Transistors Transistors can be used in an active rectifier. In an active rectifier, the transistors are controlled in a similar way that diodes behave, which make it possible to use transistors with lower forward conducting losses than the diodes. To make a rectifier for a 400 VAC current, the transistors needed have a voltage drop that is higher than an ordinary diode. That means that there are not lower losses with transistors. An active rectifier can also be used to control the voltage on the DC-bus. This has not been considered in this work.. 8.

(16) 4. Switching Strategy The inverter has to generate a sinus voltage to the motor with PWM. The strategy to generate the voltage to the motor affects the performance of the motor. This is an area were there has been made significant improvements in the last decades. Advanced and more complex control logics can provide significant improvements in performance and efficiency.. 4.1.. Scalar Control. The most common way to control a PMSM is through what is known as scalar control. With this method the voltage and frequency are controlled so that the voltage to frequency quotient is constant. As long as the load torque is below the breakdown torque value, the motor follows the current frequency. At low RPM the voltage has to be higher than the normal voltage to frequency ratio due to the larger influence of stator resistance. It is common to use a constant voltage at low speeds, to compensate for the resistance. The voltage can be controlled either by an active rectifier, with controls of the voltage at the DC bus, or by the PWM controller. Then the PWM controller, which generates the sinus voltage, also controls the voltage of it.. 4.2.. Direct Torque Control. With DTC the PWM modulator is eliminated. Instead there is a direct control of the flux vectors and current vectors used. This approach results in much better control performance. The difference to PWM is that there is no predetermined switching pattern. Each switching is determined to control the motor in the best way. The principle is to use a mathematical model of the motor in the controller. The input to this model is the real motor current, voltage and rotor angle. The model is used to calculate the flux vector in the connected motor. This knowledge of the motor dynamics is used to determine the next switching. The control can either use a fixed frequency for the switching, or a hysteresis band for the flux vector. The transistor bridge can only be switched to eight different combinations. Two of these combinations results in zero voltage to the motor. Each of the eight combinations can be described as a voltage vector. When a fixed frequency is used, the need for a switching is calculated at specific times (several thousand times per second). Each time the vector that will give the best flux vector during the next period is chosen. This can be the same vector as during the last period, this result in no switching.. 9.

(17) Otherwise, when a hysteresis band is used there is no predetermined switching frequency. The switching will happen when the flux vector leaves the hysteresis band. It should be noticed that a narrow hysteresis band will give a high switching frequency while a wide band will give a low frequency. DTC makes it possible to use a lower switching frequency than with PWM. The advantage is a gain due to lower switching losses. However, it will lead to more harmonics which is due to the fact that the switching is not synchronised with the motor frequency. One of the advantages of the DTC is that it allows a better control of the motor. With the direct torque control approach it is possible to use maximum torque without the risk to exceed the breakdown torque, which in turn makes it possible to use a smaller motor. There is also a very quick torque response which makes it possible to control the motor and the load much faster. Another advantage of the DTC is that the motor is running on its optimal flux. It is neither over nor under magnetized, this results that the motor not consuming or delivering any reactive effect. In other words it has an effect factor of 1, this result in less current in the motor and inverter. The reduction of current in turn results in less resistive losses. A DTC controller brings some new problems. The complexity of the controller becomes higher, both for the software and the logic. This makes the development cost higher. The software also needs a good model of the hardware it should control. If the motor is changed, the software needs to be updated in the controller. The diagrams below in fig 4.1 and 4.2 illustrate the differences in currents with scalar control and DTC. The load current is reduced from Î = 20 A to Î = 8 A, with the same load. Alpha and beta are scalar projection of the currents. Observe the different scales in the figures 4.1 and 4.2, [5], [20], [21].. Figure 4.1 Scalar control. Figure 4.2 Direct Torque Control. 10.

(18) 5. Control and Driver Design The inverter bridge consists of six transistors. They are grouped in three pairs, each pair form a half bridge. Each of these half bridges controls one phase of the motor. One of the transistors is connected to the DC bus low side and the other to the high side. In this description IGBT is used, but the design is similar for FET. In an inverter bridge the transistor gate is controlled by a voltage over the gate and emitter connections. Normally the voltage on an IGBT should be between 10 V and 20 V. A too low voltage results in larger conduction losses; furthermore, below 10 V it increases very fast. On then other side, a voltage higher than 20 V may result in permanent damage to the IGBT, therefore, usually a 15 V level is used. The emitter of IGBT is connected to the low voltage side of the DC-bus. This results in that the control voltage (VGE) should be 10-20V higher than the voltage of the low side of the DC-bus. The emitter of high side IGBT is connected to the motor windings. The voltage on the motor connections varies between the low side DC-bus and the high side DC-bus voltage. The low voltage occurs when the low side IGBT is conducting, while the high voltage arises when the high side IGBT turns to conducting mode. As a result of this feeding voltage the potential level of the control signal has to vary with the voltage of the motor windings to keep the offset. To be able to meet all these requirements a bootstrap circuit is found to be a solution [11].. 5.1.. Bootstrap. To provide the gate voltage to the high side of the IGBT a bootstrap circuit is used as a solution. The two “to load” connections in the figure 5.1 are connected to each other, and to one motor connection. The bootstrap capacitor charge up when the low side IGBT is conducting. The voltage on VS in figure 5.1 is the same as the low side IGBT collector voltage, and the capacitor charge from VCC (15 V) via the bootstrap diode. When the low side IGBT closes and the high side is opening, the voltage on VB lifts with the voltage on the output voltage. The voltage on VS becomes the same as the high side IGBT source voltage. The capacitor voltage can then be used to control the high side IGBT. The condition for this circuit to work is that the load is switched between high and low voltage. If there is no switching then there is no opportunity to charge the bootstrap capacitor [15].. 11.

(19) Figure 5.1 The Bootstrap transistor driver. In this figure the transistors are FET, but IGBT are controlled in the same way.. 5.1.1. Bootstrap Capacitor The bootstrap capacitor has to be able to keep the voltage up on the IGBT gate during the “on” period. The main facts that affect the minimum size of the capacitor are the minimum gate voltage, leakage currents and gate capacitance. [15]. 5.1.2. Bootstrap Diode The bootstrap diode must be able to block more than the full DC-link voltage in the inverter. This happens when the IGBT on the high side is on, the voltage over the bootstrap diode will then be the sum of full DC-link voltage and the bootstrap capacitor voltage. In this application it will be around 550 V therefore a diode rated to 1200 V will be a good choice. The average current through the diode is the IGBT gate charge multiplied by the switching frequency. The average gate current is approximately 4 mA (370 [nC] · 10 [kHz] = 3.7 [mA]).. 5.2.. Control Systems. To control the inverter some kind of logic is necessary. Several manufacturers have circuits that are designed to control an inverter. These circuits vary from microcontrollers with outputs designed to control gate drivers to dedicated circuits with no programmable options. The last type is mostly designed to control small motors in consumer applications such as in a CD-player or similar devices. One circuit that is in between these extreme cases is a circuit found from International rectifiers, IRMCF341. This circuit has both a microcontroller and an integrated programmable motor control device. This IC is able to control the motor and control all other functions in the inverter, such as error control and warning system [14]. This controller circuit is shown as an example and probably it is not the best option. It should be clear that more work has to be done to select the best controller for this application. 12.

(20) Figure 5.2 Block diagram of motor controller IRMCF341from data sheet no: PD60304. 5.3.. Power Supply. 5.3.1. Control voltages The control circuit and the IGBT drivers need a power supply. Typically 5 V or 3.3 V is needed. The IGBT driver needs a 15 V supply voltage. The IRMCF341 controller chip requires 3.3 V and 1.8 V up to 200 mW. To provide these voltages it is necessary to use some kind of step-down regulator. One possible solution is a TOP-Switch from Power Integrations which is a fly-back circuit. TOP-Switch is designed to convert 230 V AC to a low voltage DC. It can do this with an efficiency around 70 % - 80 %. To get both 15 V and the lover voltages, is it possible to use a TOP-Switch to generate a 15 V voltage and then use linear regulators for 3.3 V and 1.8 V logic voltages. The linear regulator will get a poor efficiency, around 20 %. However, the low current leads to quite low losses, around 1.5 W. It is also possible to make a middle connection on the secondary side of the switch transformer to get a 5V logic voltage. In general it must be said that there are many different solutions with the same principle where the TOP-Switch is just one example [16], [17]. 13.

(21) Figure 5.3 Typical flyback application. The TOP-Switch and other similar circuits are designed to be used with main voltage 110 – 230 VAC. That makes it necessary to use a separate rectifier connected to just one phase and zero. The secondary voltage can have a separate ground level, in this case the DC-bus low side.. 14.

(22) 6. Losses and cooling 6.1.. Losses. 6.1.1. Rectifier The losses in the rectifier bridge are almost solely due to forward voltage drop in the diodes since the low frequency makes the switching losses low. The losses in the diode is W = I · UF where UF is the forward voltage drop in the diode. The forward voltage drop is mainly due to the PN-junction and to some extent to resistive effects.. 6.1.2. Driver, Control and Internal Power Supply The driver and control circuit in the inverter needs power to work. All this power is considered as losses, because nothing of it is used to drive the motor. Current needs for different parts. • Average total gate current is 3.7 mA · 6 = 22 mA (from chap. 5.1.2) • The IGBT driver demands 0.2 mA [15] • Microcontroller supply 27 mA at 3.3 V and 67 mA at 1.8 V [14] This makes the total currents for the voltages below than 15 V less than 100 mA. This will probably be higher in a final product due to more functions however this is an acceptable hint of the order of magnitude of the power demands. The total losses will be around 1.5 W if all of this electronics is powered with linear regulators from the 15 V line. The step down converter to 15 V will have an efficiency of around 70 %, which leads to 0.6 W of losses. The total losses from transistor drivers, control circuits and internal power supply will be around 2.3 W.. 6.1.3. Transistor Bridge The losses in the transistor bridge consist of both conducting and switching losses. The conducting losses are a result of the voltage drop over the transistor. This voltage drop can be described as comprising two parts, one is a voltage drop over a P-N junction, as in a diode, and the other part is a resistive voltage drop. Thus, the switching losses are due to an energy loss at each on-off turn over. The product of the voltage and the current integrated over one on-off turn over yields the energy loss for one period. The average power loss is then given by the energy loss divided by the switching time period. Different transistors have different characteristics. A transistor with lower conducting loss has often higher switching loss. Therefore, the process of selecting a transistor results in a trade-off between its conducting and switching losses.. 15.

(23) 6.2.. Cooling. To keep the inverter at a suitable working temperature is it necessary to cool it. The IGBT core has to be below 125 °C. A rule of thumb is that the failure rate of Si semiconductors doubles for every 10 – 15 °C rise in temperature above a working temperature of 50 °C, [12]. Therefore, to have a reliable and long lasting electronic system it is mandatory to keep the electronic components cool. Different types of cooling systems have been designed and several are extensively used. For low power components it is enough with the surrounding air and the junctions. For components with higher power loss is it common to use a heatsink, with or without fan, it is also possible to use fluid based cooling systems or heat pipes. The demands on the cooling system used will depend on a number of parameters as for instance where and how it will be used, its total power, and its operation form. For example, it could be used intermittent or continuously. If the use is intermittent it is possible to let the heatsink and the casing to store some of the heat allowing the system being cooled during a longer cut-off period than the current operational periods of time. That makes it possible to use a somewhat under dimensioned heatsink that would not be capable of cooling all the loss power developed during operation intervals.. 6.2.1. Thermal Resistance While designing a correct cooling system for a power electronic device it is essential to know the thermal resistance of the components used, to be able to calculate a correctly dimensioned cooling system. The cooling system can be schematically represented by an equivalent electric circuit. The figure 6.1 shows a schematic of an equivalent electrical circuit for a simple cooling system. Basically, the voltage represents the temperature, and the current represent the heat flux. TJ is the junction temperature of the transistor and TC is the case temperature of the transistor. RΘJC is the transistors internal thermal resistance. TH is the heatsink temperature at the contact point, and RΘCH the thermal resistance between transistor and heatsink. RΘHA is the resistance from the heatsink to the ambient and finally, TAMB is the ambient temperature of the system. If the maximum TJ is 125 °C and TAMB is 70 °C and the losses in the transistor is 30 W the total thermal resistance must not exceed 1.8 K/W. Typical for a power transistor is 0.6 K/W which leaves 1.2 K/W to the isolation pad and the heatsink. If there are more than one heat source on one heatsink they will be connected in parallel in the schematic.. 16.

(24) Figure 6.1 Schematic of thermal resistance model. This model is suitable for a steady state heat flow. The power frequency in the inverter is rather high and therefore for practical purpose its average value is considered. If the inverter is designed for intermittent use, the thermal inertia of the system has to be taken into account in the model.. 6.2.2. Spreading When using a heatsink for more than one component, is it important to distribute the components to use the heatsink in an effective way. If the components are placed in a bad configuration it can result in hot spots in the heatsink leading to components that becomes hotter than expected.. 6.2.3. Heatsink A heatsink is a massive metal body chosen because of its low thermal resistance and high heat capacity, which is used to conduct heat away from the hot component. The heatsink is then cooled, usually with just circulating air. There are several parameters that affect the performance of a heatsink. The material in the heatsink that is used to conduct the heat is one of them. Different materials have different thermal conductivity and also different specific heat capacity. The mass and geometry of the heatsink affects the heat conducting performance as well as the heatsink capability to store heat. The area that the heatsink is in contact with the surrounding air affects the amount of heat that can be removed from it, as well as the shape of the heatsink affects the air flow around it. It is desirable to have a good air flow around all parts of the heatsink. Furthermore, the flow around the heatsink should be turbulent to enhance the thermal transfer capability from the heatsink to the circulating. 17.

(25) air. The color of the heatsink is also a parameter that affects the performance. A black oxidized passive heatsink will have around 25% lower thermal resistance than a similar with metallic surface (see black body radiation theory). The pressure of the surrounding air also affects the cooling (it will indirectly affect the Reynolds number since it affects density and viscosity). In avionic applications it should be regarded that at low pressure the air is less dens and can not handle the same amount of heat, furthermore, the air temperature is of course vital for an effective cooling, [12].. 6.3.. Manufacturing Methods for Heatsinks. 6.3.1. Casting With casting methods it is possible to make heatsinks that are shaped in three dimensions. This makes it possible to accomplish geometries that are optimal than those that can be obtained with other manufacturing methods. It is rather difficult to cast pure aluminum therefore it is often needed to add impurities to facilitate the casting process, [8].. 6.3.2. Extrusion Extrusion is a common way to manufacture heatsinks made of aluminum. It is a cheap production method but it associated with a rather high initial cost. The disadvantage is that it is only possible to shape in two dimensions with this method (cross section) and therefore the third dimension will always be straight. The most common material used in extruded heatsinks is aluminum, but copper or brass is also an option.. 6.3.3. Punch For small heath power removal demands punched heatsinks could be a cheap solution. In this method the heatsinks are punched out of sheet metal. Often the fins are bended in different directions. Even here the most common material used is aluminum. A heatsink can also be designed with stacked sheet metal parts. By bending the different layers in different ways it is possible to form a heatsink in three dimensions.. 6.3.4. Milled In manufacturing small series of heathsinks a milled method is an option. The manufacturing price will be high, but the initial cost is usually small. In aircraft applications the size of production series is often rather small, and the price per unit tolerance can be higher than in other branches. This makes this method a reasonable option.. 18.

(26) 6.4.. Material. 6.4.1. Aluminum Aluminum is light and has a good thermal conductivity. It is easy to manufacture aluminum heatsinks by extrusion. In comparison with copper, aluminum is a poorer heat conductor. But if heat conduction per weight is compared the aluminum will be the better option.. 6.4.2. Copper Copper is one of the best heat conductors there is. Only silver is slightly better. (4.5 %) Silver is both heavier (17 %) and more expensive, silver is not a good option. Copper is more than three times heavier than aluminum. In an aircraft application low weight is essential. Copper can be an option in the core of the heatsink where there is a high heat flux.. 6.4.3. Zinc Zinc is a worse heat conductor than both aluminum and copper. It is also heavier than aluminum. The advantage with zinc is that it is easy to cast. An aluminum alloy that is castable has a heat conductivity that is in the same level as zinc. This makes zinc a good option in some applications. But in this application is the high weight is a disadvantage. [8] Copper Thermal conductivity Density. Heat capacity Price 08-06-04. 400.00 8.96 385.00 7.85. W/(m·K) g·cm−3 J/(kg·K) $/kg. Aluminum 238.00 2.70 903.00 2.90. W/(m·K) g·cm−3 J/(kg·K) $/kg. Zinc 120.00 7.13 389.00 1.97. W/(m·K) g·cm−3 J/(kg·K) $/kg. Table 6.1 Comparison of heatsink materials [19] [22].. 6.5.. Heat pipes. Heat pipes consists of a sealed tube made of a material with high thermal conductivity filled with a medium that change its phase from liquid to gas in the hot end of the pipe. In the cooler part of the pipe the medium is condensed. The condensed fluid is then transported back with gravity or capillary forces. The heat pipes always work with a pressure that makes the boiling/condensing temperatures to be in between the temperatures in the hot and cool end of the pipe. This makes the media boil in the hot end and condense in the cool end, resulting in a good heat transport. This device, due to its basic principle, makes the heath transport to be much better than a solid piece of metal. A 6 mm diameter pipe, 150 mm in length is able to transport up to 65 19.

(27) W with a thermal resistance of just 0.03 K/W. A common medium in a heat pipe is water but it obviously does not work at temperatures below 0 °C. There are however other medium that can operate at even lower temperatures. A common media for temperatures below 0 °C is methanol. Gravity force in the same direction as the heat transport makes the heat pipe less effective [3]. In aircraft applications it can not be taken for granted that the gravity force is in a specific direction. But if there is a limitation for flying with negative g-forces, the heat pipe can still be a very god option. Heat pipes are able to transport heat from one hundred to several thousand times that of an equivalent piece of copper. The resistance to conduct the heat within the pipe is very low, with typical values of 0.2 °C/(W·cm2) for thermal resistance at the evaporator and condenser, and 0.02 °C/(W·cm2) for axial resistance.. Figure 6.5 Schematic heat pipe operation mechanisms and description. 20.

(28) 7. Chassis Design 7.1.. Cooling Scenarios. When discussing cooling systems in aircraft it is necessary to consider what kind of airplane the system is intended to be used in.. 7.1.1. JAS 39 Gripen In a cooling system in an aircraft such as Gripen, air at around 0 °C is used. This cooling system needs a significant amount of power to generate this cool air. Furthermore, the cooling system gets heavier if the needs are big. This makes it necessary to use as little air as possible to cool the inverter. The graph shows the required amount of cooling air to cool the inverter with 0 °C, 10 °C and 25 °C air temperature and various temperatures out from the cooling system. The loss is estimated to 300 W. The normal temperature variations in the input air are 0 °C – 25°C.. Air requirements 2. Air amount kg / min. 1.8 1.6. Incoming temperature. 1.4 1.2. 0 °C 10 °C 25 °C. 1 0.8 0.6 0.4 0.2 0 30. 40. 50. 60. 70. 80. 90. 100 110. Outgoing temperature °C Figure 7.1 Input cooling air requirements at 0, 10 or 25 °C. 21.

(29) 7.1.2. MERA-Windex In MERA-Windex the cooling solutions are different that in a fighter. MERA-Windex is an electrified sailplane. The air available is the same that surrounds the aircraft. This air is not cooled and can be warmer than in case considered for a fighter, on the other side, there is almost no limit of the amount of air available. The only energy needed for the cooling is to make an air flow into the system to be chilled. This can be done by using a fan or the airflow around the plane. The heatsink can also be a part of the plane surface.. 22.

(30) 8. Results 8.1.. Transistors. The conclusion is that the best transistor type for an inverter for this voltage is IGBT. In this work a large number of transistors have been evaluated, and the result of this evaluation is summarized in table 8.4. All the transistors that seamed to be a possible option were evaluated. In the evaluation both discrete transistors and complete bridges were checked. The transistors that have been checked can be found in table 8.4. The most interesting information in table 8.4 is the total power loss (see columns marked Plosstot). This is a calculation of the total power loss in the transistor bridge at different frequencies. The calculations are an estimation of the losses when the transistor is used in an inverter. The calculations use a simple model of the transistors and the inverter operation [4].. 8.1.1. Transistor Bridges Transistor bridges are easy to design with. One block contains all power components for the inverters usually they comprise six inverter transistors, freewheeling diodes, a brake transistor and a temperature sensor. In some cases the transistor drivers are also integrated in the module. The module is also electrical isolated from the heatsink.. 8.1.2. Discrete Transistors Discrete transistors in general have a lower thermal resistance than bridge modules. It is also possible to distribute them on a heatsink which makes them easier to cool. However, they are not electrically isolated from the heatsink, and therefore there has to be an isolated disk between the transistor and heatsink. This in turn makes the thermal resistance somewhat higher.. 8.1.3. Selection All potential transistors found were compared with a mathematical model. [18] Equations 8.1 to 8.4 are the model for the transistor loss, and equation 8.5 is the model for the freewheeling diode conduction loss [18].. Pton. { 2I. }. {. [. /(2π ) × {a + (2π )}× a + (π / 4)b × 2 I 0 + (π / 4) × cos Φ * a + (8b /(3π )) × 2 I 0 Eq 8.1 Eq 8.2 = 0.5 × Eton × f. PON =. 0. Ptoff = 0.5 × Etoff × f. Eq 8.3. Ptot = PON + Pton + Ptoff. Eq 8.4. 23. ]}.

(31) Definitions:. I0 a, b. Inverter phase output current rms value Linear approximate curve value represented by VCE (sat) = a + bi shown in fig. 8.1.. cos(Φ) Load power factor. Figure 8.1 linear approximations for the Current-Voltage characteristic curve.. Pf =. { 2I. 0. Definitions:. [. }{. /(2π ) × a + (π / 4)b × 2 I 0 − (π / 4) × cos θ × a + (8b / 3π ) × 2 I 0. I0 a, b. ]}. Eq 8.5. Inverter phase output current rms value Linear approximate curve value represented by VF = a + bi similar as in fig 8.1. cos(Φ) Load power factor. The total losses when the transistor was used in a bridge to generate a 400 V 15 A (rms) sinus voltage with 10 kHz switching frequency were calculated. The results can be found in table 8.4. The IRG4PSH71U is the transistor that seems to be the best option. This is a transistor without a freewheeling diode. A separate diode is easier to cool, and a SiC diode would lower the transistor losses even more. IRG4PSH71UD is the same transistor but with diode.. 24.

(32) 8.2.. Freewheeling Diodes. The choice of freewheeling diodes affects the switching losses both in the diode and in the transistor. A SiC diode lowers these losses. There is a little increase in conducting losses, but a SiC diode is still a god option. Today they are more expensive, but in an application like the one discussed in this work the better performance make it worth of using them. The generic conduction losses in the freewheeling diode can be calculated with equation 8.5 above.. 8.3.. Switching strategy. DTC or another kind of vector technique is a good option to control the inverter. This makes the working current level lower and the controllability of the motor better. This is of benefit if the controller is aimed to be used in a servo application. These benefits motivate the more complex control algorithms. The benefits were already discussed in chapter 4.. 8.4.. Control and driver circuits. The example circuits described earlier in this work (chapter 5) were selected just to illustrate the functions needed. They are probably not the most suitable circuits. All these and other possible circuits have to be evaluated for best performance, reliability, efficiency, etc.. 8.5.. Cooling. Aluminum is the best heatsink material for an airplane application because of its low thermal resistance and low density. The thermal resistance for the system is: • • • • • •. 0.36 W/°C junction to case 0.24 W/°C case to heatsink 0.3 W/°C isolator 0.24 W/°C insulator to heatsink 0.2 W/°C heatsink 0.4 kg Total 1.34 W/°C. [2] [2] [9] (same value as case to heatsink) [23]. Those values are for the case of an aluminum heatsink assuming that available air flow is enough to cool the circuit. With a power loss of about 27 W/IGBT, the air temperature has to be lower than 88 °C to keep the junction temperature below 125 °C. To get even better. 25.

(33) reliability, a lower design temperature should be preferred. The heatsink considered above is based on a commercially available heatsink [23].. 8.6.. Reliability. For aircraft application, reliability is extremely important since the consequences of a failure can be fatal. Since a single component error can result in a total loss of function, reliability has to be assessed carefully in every aeronautic application. Some parts or components can be made redundant, such as the internal power supply. Others parts, like sensors for instance, can allow the inverter to be designed to still work safe and properly even if some of the sensors fails. But other parts like the power components are harder to be made redundant in the system without big disadvantages in weight and cost. One way to make a fully redundant system can be simply to build two systems that work in parallel, then, if one of the systems goes down can the function be maintained with a parallel circuit, although with some decreased performance. A better way to make systems reliable is to reduce the failure risks. To achieve this, the circuit has to be deigned to handle the normal operation with margins. For example the recommended transistor should have a voltage rating that exceeds the double system voltage, and with even larger currents margins. As already mentioned, experience has shown that electronics double their failure rate for every 10°C rise above a working temperature of 50°C. This means that even if the silicon can handle 125°C, a design for a cooler working temperature is a god idea. Temperature cycling and variations are also known to be a stress factor for electronic components and therefore, a stable working temperature below 50°C is optimal for a high reliability.. 8.7.. Estimations. The estimations of losses were calculated using a very simple model as it was described in chapter 8.1.3. The results can be used to estimate the size and cooling needs for a future converter however, a construction project will demand a better calculation or a numerical simulation.. Loss estimations Power supply & driver & control. Rectifier diodes Transistors loss Freewheeling diodes. 2.3 W 30 W 120 W 5.2 W. Chap. 6.1.2 [10] Table 8.4 Table 8.4. Table 8.1 Estimated losses.. 26.

(34) Weight estimations Chassis Cover Cooling Connectors Transformer with Y & D Rectifier diodes Transistors Capacitors Drives and control PCB Total weight ca.. 500 g 500 g 4 000 g 100 g 20 000 g 40 g 36 g 5 000 g 40 g 200 g 30 500 g. Table 8.2 Weight estimations.. Hardware cost estimations Component IGBT Rectifier Logic PCB Heatsink Transformer SiC diode Driver PSU PSU other Capacitor Sum ca.. Component details URG4PSH71U 36MT120 dsPIC30F6010A 4layer 6*8" 80*80*40 0,2 °C/W 3 phase, 50Hz C2D20120D IR2131 TOP221. Price Kr 103.40 114.00 163.00 209.10 547.70 6 000.00 203.28 72.50 22.50 25.00 1 000.00. Amount 6 2 1 1 6 1 6 1 1 1 4. Sum Kr 620.40 228.00 163.00 209.10 3 286.20 6 000.00 1 219.70 72.50 22.50 25.00 4 000.00 16 000.00. Source Digikey ELFA ELFA PCB123.com Farnell Digikey ELFA ELFA ELFA ELFA. Table 8.3 Cost estimations.. 8.8.. Inverter layout. In figure 8.7 a possible layout for a 10kW inverter is suggested. The semiconductor components that have the largest power loss are distributed on a heatsink on each side of the inverter. The heatsinks are designed to be cooled with a forced air flow along the sides of the inverter.. 27.

(35) Figure 8.7 Inverter lay-out.. The power input section contains a 12pulse rectifier. The capacitors are distributed close to the transistors in the inverter bridge, to keep the inductances low. In the center part of the inverter are the electronics with lower cooling needs. This is one layout suggestion, but there are a number of things that influence the optimal layout. That is a work that has to be done with the specific product in mind. With its EMI, cooling and operational problems, etc.. 28.

(36) 9.. Discussion and Conclusions. It is likely that electrical drives will be more common in the future. The development in this area makes progress all the time. The power semiconductors performance increases for every year, and the SiC based semiconductors have been for years a very promising field for solid state technologies, a breakthrough has to be assumed much closer than in the past. The SiC diodes are already making possible excellent improvements in power inverters. This development is not something that will happen only in the aviation industry. The possibility to save energy, space and weight will be a great advantage to many different areas. This has been an interesting study to carry out. I have got an insight in both the power electronic development of today, the aviation industry, their interrelations and points of common interest.. 10. Future Work In this work the input filter is not considered. To use the system in an airplane is it necessary to use some kind of filter and, a 12 pulse rectifier would most probably be a suitable start. In any case this is matter that will need further assessments. Before any hardware constructing takes place it would be advisable to perform numerical simulations of the inverter. With a simulation study it would be possible to understand much better the operational figures of merit of the inverter.. 29.

(37) 11. References [1]. O’Neill Michael. The benefits of using a Cree Inc. IGBT/SiC Schottky Co-pack in AC inverter applications. Cree inc. September (2006). [2]. International Rectifiers. Insulated gate bipolar transistor. IRG4PSH71U Datasheet no: PD – 91685 [3]. CCI Heat Pipe Design Guide. CCI RD01 Heat Pipe Team. June 27, (2006). [4]. Fuji IGBT modules application manual. Fuji electric device technology co. LTD. REH984; February, (2004). [5]. Schmidt István, Vincze Katalin, Veszprémi Károly and Seller Balázs. Adaptive hysteresis current vector control of synchronous servo driver with different tolerance areas. Department of electrical machines and drives. Budapest University of Technology and Economics. June 29, (2000). [6]. Richmond Jim. Hard-switched silicon IGBTs? Cut switching losses in half with silicon carbide Schottky diodes. Cree Inc. (2005) [7]. Vieillard S. and Meuret R. High efficiency, high reliability 2 kW inverter for aeronautical application. Hispano-Suiza, Rond point rené ravaud, MoissyCamayel, France, ISBN: 9789075815108, (2007). [8]. Keller Kurtis P. Efficiency and Cost Tradeoff Between Aluminum and ZinkAluminum Die Cast Heatsinks. University of North Carolina, Computer Science Department. [9]. ELFA page 212 (2008) [10]. International Rectifiers. Three phase bridge Bulletin I2771 rev. D (1997-08) [11]. International Rectifier. Applications note AN-978. HV Floating MOS-Gate Driver IC’s, (2007). [12]. International Rectifier. Applications note AN-983. IGBT characteristics. [13]. International Rectifier. Applications note AN-990. Application characterization of IGBT’s. [14]. International Rectifier. Sensorless Motor Control IC for Appliances. IRMCF341 Data Sheet No. PD60304, (2006)..

(38) [15]. International Rectifier. High And Low Side Driver IR2213 Preliminary Data Sheet No. PD60030-M, (2002). [16]. Power Integrations, INC. Application Nota AN-29 TOPSwitch-GX Flyback Quick Selection Curves, (2003). [17]. Power Integrations, INC. Data Sheet for TOP221-227 TOPSwitch-II Family, (1997). [18]. IGBT MODULE APPLICATION Manual Hitachi, Ltd. Ref.No.IGBT-01 (Rev.2) [19]. Nordling Carl and Österman Jonny. Physics Handbook ISBN: 91-44-03152-1, (1999). [20]. Stulrajter Marek, Hrabovcová Valéria and Franko Marek. Permanent Magnets Synchronous Motor Control Theory. Journal of Electrical Engineering, Vol. 58 No. 2, pg. 79-84 (2007). [21]. United States Patent: 4 678 248. Manfred Depenbrock; October 18 (1985) [22]. www.finansportalen.se (2008-06-04 16:50) [23]. Heatsink datasheet, Manufacturer: HS MARSTON AEROSPACE LTD, Article number: CF1-0816-0400-0816A.

(39) Eo n. Pc on Eo @ d E @ 12 rr ff @ @ 5° 15 12 12 C A 5° 5° 12 (m C C Vc 5° J/ Vf c ( ( C es m m FW yc 50 J/ J/ at Pl Pl l e cy cy D Pl % )( os os @ Ps os cl cl e @ n M D 1 st st Ps w e s Pt M 2 s P u a ot ot )1 )1 )1 15 Ic @ t R 5 w t nu s o od o I Vd y R ° c Vt w d P t @ @ @ 1 t @ 5 A 5 5 C V cy fa t@ @ 0@ @ el @ @ A A ac A @ ce 10 14 12 14 10 @ @ & ct na c F 6 6 6 6 2 6 2 k le m Fs 15 15 15 15 W ur kH 15 kH 00 kH 00 kH kH 5° Kh 00 5° ag m ax 5°C 0°C er (W D C C w A A A A V A V V e e z z z z z z ) IRGPS60B120KD IRGPS60BTO-274AA 595 1200 120 60 1,8 1,30 5-40kHz 3,10 0,82 10,7 9,3 15,5 21,7 125 162 199 Si 0,039 1,1 0,032 0,75 IRG4PSH71UD IRF TO-274AA 350 1200 99 50 1,5 1,70 5-40 3,2 0,86 9,5 9,6 16,0 22,4 120 158 196 Si 0,033 1 0,035 0,775 IRGP30B120KD-E IRF TO-274AC 300 1200 60 30 2,3 1,50 5-40 2,8 3,97 14,3 8,4 14,0 19,6 160 194 227 Si 0,06 1,34 0,39 0,9 IRG4PSH71KD IRF TO-274AA 350 1200 78 42 2 1,7 4-20 4,9 1,15 12,6 14,7 24,5 34,3 171 230 288 Si 0,04 1,4 0,035 1,18 IRGPS40B120UD IRF TO-274AA 595 1200 80 40 2,6 1,50 5-40 2,85 1,02 16,6 8,6 14,3 20,0 157 191 226 Si 0,07 1,55 0,039 0,95 IRG4PF50WD IRF TO-247AC 200 900 51 28 2 2,60 20-100 2,5 1,80 10,8 7,5 12,5 17,5 120 150 180 Si 0,035 1,18 0,085 1,48 IRG4PH50KD IRF TO-247AC 200 1200 45 24 1,5 2,4 4-20 4,2 1,60 14,5 12,6 21,0 29,4 172 223 273 Si 0,045 1,63 0,06 1,5 IRG4PH50UD IRF TO-247AC 200 1200 45 24 2,7 2,4 5-40 5 1,62 14,9 15,0 25,0 35,0 189 249 309 Si 0,03 1,95 0,05 1,65 IRG4PH40U IRF TO-247AC 160 1200 41 21 2,3 5-40 2,8 0,00 14,4 8,4 14,0 19,6 137 170 204 0,04 1,7 IRGP20B120UD-E IRF TO-247AC 300 1200 40 20 3,4 1,5 5-40 2,1 1,05 21,9 6,3 10,5 14,7 176 201 226 Si 0,095 2 0,043 0,95 IRG4PH40UD IRF TO-247AC 160 1200 41 21 2,3 4 -40 3,8 2,78 14,7 11,4 19,0 26,6 173 219 264 Si 0,06 1,4 0,14 2,2 IRG4PH40K IRF TO-247AC 160 1200 30 15 2,53 4-20 2,6 0,00 16,8 7,8 13,0 18,2 148 179 210 0,075 1,5 IRG4PSH71U IRF TO-247AA 350 1200 99 50 2,4 8-40 2,11 2,81 9,5 6,3 10,6 14,8 112 137 162 0,033 1 0,26 0,83 IRGPS40B120U IRF TO-247AA 595 1200 80 40 2,6 1,5 0,00 16,6 4,5 7,5 10,5 127 145 163 0,07 1,55 IRGB15B60KD IRF TO-220AB 208 600 31 15 2,05 1,20 1,26 0,77 14,1 3,8 6,3 8,8 112 127 142 Si 0,06 1,3 0,025 0,77 2 x CID100512 CREE 2 x TO-247-3 200 1200 40 20 2,6 2,80 2,835 2,78 23,2 8,5 14,2 19,8 207 241 275 SiC 0,13 1,6 0,25 0,9 CID150660 CREE TO-220 208 600 31 15 2,05 3,25 1 1,89 13,8 3,0 5,0 7,0 112 124 136 Si 0,063 1,2 0,15 0,85 APT50GF120JRD APT SOT-227 460 1200 75 50 1,9 0,80 2,1 0,64 12,0 6,3 10,5 14,7 114 139 164 Si 0,052 1,1 0,023 0,62 APT50GF60BR APT TO-247 300 600 75 50 1,3 1,9 0,00 8,4 5,7 9,5 13,3 84 107 130 0,034 0,8 7MBI50N-120 FUJI 7 in one-packa 400 1200 50 1,7 1,2 -24kHz 2,25 2,7 1,8 0,83 11,3 20,3 33,8 47,3 194 275 356 Si 0,034 1,285 0,018 0,93 7MBR25SC120 FUJI 7 in one-packa 180 1200 35 2,1 1,5 -40kHz 1,8 1,26 0,81 1,01 13,8 11,6 19,4 27,1 159 205 252 Si 0,065 1,169 0,033 1 7MBR25SA140 FUJI 7 in one-packa 180 1400 35 2,1 1,75 -42kHz 2,88 2,25 1,35 1,15 13,8 19,4 32,4 45,4 207 284 362 Si 0,065 1,169 0,05 1 7MBR25SA120 FUJI 7 in one-packa 180 1200 35 2,1 1,5 -42kHz 1,89 1,17 0,81 0,99 13,8 11,6 19,4 27,1 159 205 252 Si 0,065 1,169 0,04 0,9 7MBR15SC120 FUJI 7 in one-packa 110 1200 25 2,65 1,8 -38kHz 2,07 1,17 0,72 1,17 17,3 11,9 19,8 27,7 182 230 277 Si 0,1 1,15 0,046 1,09 7MBR25UA120 FUJI 7 in one-packa 115 1200 25 1,9 1,8 -38kHz 2,34 1,44 1,17 1,18 12,7 14,9 24,8 34,7 173 232 291 Si 0,053 1,2 0,053 1 7MBR35UB120 FUJI 7 in one-packa 160 1200 35 1,5 1,35 -38kHz 1,89 1,44 1,17 0,94 9,5 13,5 22,5 31,5 144 198 252 Si 0,037 0,943 0,032 0,93 FS15R12VT3 Infineon SixPACK 86 1200 24 1,85 1,65 1,71 1,305 0,846 1,05 11,8 11,6 19,3 27,0 147 193 239 Si 0,071 0,729 0,054 0,81 FS15R12YT3 Fast Infineon SixPACK 110 1200 25 1,85 1,65 1,35 1,485 1,008 1,05 12,2 11,5 19,2 26,9 149 195 241 Si 0,078 0,683 0,053 0,84 BSM 15 GD 120 DN2Infineon SixPACK 145 1200 25 3,1 1,7 -84kHz 2,16 1,35 0,81 1,04 20,9 13,0 21,6 30,2 209 261 313 Si 0,142 1,025 0,096 0,31 FS25R12W1T4 Infineon SixPACK 205 1200 45 25 1,65 1,4 1,26 1,17 1,08 0,99 11,0 10,5 17,6 24,6 135 177 220 Si 0,054 0,9 0,03 1,02 FS25R12YT3 Infineon SixPACK 165 1200 40 1,5 1,3 1,35 1,53 1,26 0,91 10,5 12,4 20,7 29,0 143 193 242 Si 0,042 1,025 0,027 0,94 FS25R12KE3 G Infineon SixPACK 145 1200 40 1,55 1,35 1,35 1,98 1,62 0,96 9,8 14,9 24,8 34,7 154 213 272 Si 0,032 1,07 0,029 0,99 BSM 25 GD 120 DN2Infineon SixPACK 200 1200 35 2,4 1,4 2,07 1,35 3,24 0,98 15,5 20,0 33,3 46,6 219 299 379 Si 0,057 1,588 0,031 0,99 FS25R12KT3 Infineon SixPACK 145 1200 40 1,5 1,35 1,35 1,575 1,35 0,89 9,6 12,8 21,4 29,9 140 191 243 Si 0,042 0,875 0,037 0,8 FP15R12W1T3 Infineon 7 in one-packa 105 1200 25 1,8 1,95 1,98 1,26 0,945 1,27 12,3 12,6 20,9 29,3 157 207 257 Si 0,061 0,983 0,063 1,01 FP15R12W1T4_B3 Infineon 7 in one-packa 130 1200 28 2,15 2,1 1,62 1,17 0,63 1,46 14,7 10,3 17,1 23,9 158 199 241 Si 0,083 1 0,027 1,7 FP15R12KE3G Infineon 7 in one-packa 100 1200 25 2 0,9 1,98 1,35 0,99 1,09 13,0 13,0 21,6 30,2 163 214 266 Si 0,073 0,9 0,047 0,95 FP15R12KS4C Infineon 7 in one-packa 180 1200 30 3,7 0,9 1,8 0,99 0,9 1,04 24,5 11,1 18,5 25,8 220 264 308 Si 0,12 2 0,036 1,01 FP25R12KE3 Infineon 7 in one-packa 150 1200 40 1,55 2 1,71 1,35 1,26 0,90 10,0 13,0 21,6 30,2 143 195 247 Si 0,045 0,875 0,033 0,86 FP25R12KT3 Infineon 7 in one-packa 155 1200 40 1,4 1,3 1,8 1,17 1,215 1,28 8,5 12,6 20,9 29,3 134 184 235 Si 0,028 0,93 0,073 0,9. Table 8.4 IGBT (All calculations are per transistor).

(40)

No results found