Low Power Extended Range DC Motor Controller: Design and Implementation, Study Guides, Projects, Research of Electrical and Electronics Engineering

Download Low Power Extended Range DC Motor Controller: Design and Implementation and more Study Guides, Projects, Research Electrical and Electronics Engineering in PDF only on Docsity! Low Power Extended Range DC Motor Controller ECE 445 Final Paper May 3, 2005 Team Members: Chris Groesch Chris Swanson TA: Bryan Dobbs Abstract As power electronics become more widely used and understood, as well as cheaper in cost, the use of DC motors in everyday devices will increase. This is why we chose to build a DC motor controller for our project. Our goal was to show that power electronics can be used in a cost effective manner to control machines that cannot be connected to the AC power grid because of their mobility. Our circuit was implemented using to sub-circuits controlled by two IC chips. The first circuit is a boost converter with its switching proscribed by the UC3825 chip. We designed the converter to increase the input from a lead-acid battery to a constant 15V. The second half of our circuit is an H- Bridge with a LS7260 chip to control the gates of the four MOSFET and therefore the output voltage to the motor. In this manner, the LS7260 is maintaining a user specified output voltage; the motor speed will also be determined by the user.  Circuit Breaker protection 2. System Block Diagram Figure 1. DC Motor Controller Block Diagram 2.1 12V Battery The input source is taken directly from a conventional 12V lead-acid battery. The reserve capacity of the battery is fully dependent on the quality and cranking amps. The ouput to the boost converter is 10-15 Vdc with a current range of 0-56 amps depending on the speed setting. 2.2 Boost Converter The boost converter, or step-up converter, is a switching dc/dc converter that produces an output voltage greater than the source(ranging from 10-15Vdc). The general layout of a boost converter is shown in Figure 2. For the worst case scenario, a 10V input voltage will have to be boosted to the maximum output of 15V. The design of the boost converter entails four main parts: Inductor, MOSFET, MOSFET Control, and a Diode. Figure 2. Basic Boost Converter Design. 2.2a Inductor One of the major aspects of the boost converter is the design of the inductor. Special consideration must be taken in choosing a core in order to accommodate at least 50A. The problem that arises is saturation and hysterisis at such high current levels. The power lab contained several iron powder cores which had a relative permeability of 75μo. Iron powder composed cores are known to possess a flux density of at least one Tesla and are also much smaller in size as opposed to a ferrite core of similar specifications. The toroidal shape by geometry tends to minimize leakage flux and allows easy access to the window area for winding. Upon recommendation, a T225-26 core was chosen for its permeability and size. As mentioned the 26 mix has a permeability of 75μo. The core has an outer diameter of 2.25 inches, an inner diameter of 1.405 inches and a height of .55 inches. As shown in equations (1) and (2), the inductance needed for a frequency of 30kHz came out to be approximately 60μH. A frequency of 30kHz was chosen for the mere fact that the core is not stable at higher frequencies with such a large current. If the frequency were chosen any larger the losses would significantly increase which in turn decreases the efficiency. t i LV    (1) H A s VL   05.60 22.2 33.13 *10  (2) The equivalent resistance provided by the core is shown in (3). The minimum amount of turns needed to achieve at least 60μH is calculated in equation (4). The core was actually wrapped with 30 turns to ensure proper minimal ripple. Equation (5) calculates the maximum flux based on Figure 3 and the dimensions of the core. With the maximum flux, the maximum current was then calculated in equation (6). Ideally, the maximum current is 50A so the core should not saturate due to its maximum current rating of 71.24A. Wb turnsA m m A l R cor c    7 27 10*029.1 5.150*10*4*75 14593.  (3) turns Wb turnsA HLRN c 94.2410*029.1*05.60 7     (4) WbmTAB c 42 max 10*0769.25.150*38.1*  (5) AWb turnsA Wb N R i c 24.71 30 )10*029.1(*)10*0769.2(* 74      (6) Figure 3. BH Curve for Inductor Core. In the worst case scenario the wire used in wrapping the inductor would need to carry about 60A taking losses into account. According to the National Electric Code (NEC), 8awg can carry a maximum of 70A in free air. Since the wire is pvc coated, it can withstand approximately another 5A above the free air rating. This will be the gauge of wire used throughout the circuit wherever full current can flow. Due to the fact that 8 gauge is rather difficult to wrap around the window area of 44.18cm2, a flat braided copper was chosen for its flexibility and ease of wrapping. The flat braided copper implemented has a width of a quarter of an inch which is equivalent in cross sectional area to that of an 8 gauge wire. The flat braided copper had to be insulated since the wire is not coated. Since the wire used was not a typical round gauge wire, the number of turns determined may be off by a bit due to the wire being braided within itself. The core was wrapped so the surface area was completely covered which took 30 turns. The inductance resulted in 90μH which is good since at least 60μH was needed. The next component modeled in the boost converter is the MOSFET which is used to vary the duty cycle. 2.2b MOSFET 2.3 H-Bridge The H-Bridge has the ability to change the direction of the motor and stop it completely. The control unit adjusts the inputs to the MOSFETs to accommodate the user defined action. 2.3a MOSFET The MOSFETs for the H-Bridge needed to be able to handle more than 70 A, so the STB200NF04 was used for the NFETs and the STB80PF55 was used for the PFETs. . Both were contained in a T0-220 package. The NFETs had ratings of 120A and 40V. The PFETs have ratings of 80A, 55V, and a minimal on state resistance of 1.8mΩ. 2.3b MOSFET Control The control portion of the circuit is primarily responsible for ultimately regulating the input voltage to the motor. The control unit also adjusts the inputs to the MOSFETs of the H-Bridge to vary the motor direction and has built in dead time to assure that all four MOSFETs are not on at the same time within the H-Bridge. Figure 6 shows the pin layout of the LS7260. Figure 6. LS7260 Pin Layout. The LS7260 was used to control the MOSFETs in the H-Bridge and there by the voltage output and the motor speed. The Commutation Select capability of the LS7260 was not needed for the H-Bridge application. The gates of the MOSFETs were driven using four of the six outputs on the chip. Out 1-3 were specified to drive PFETs, and Out 4-6 were to drive NFETs. By using this chip, the H-Bridge can be disabled using the Enable or Brake switch. This would cause the voltage to the motor to vanish and therefore stop it. It also allows for either forward or reverse operation of the motor through the use of a switch. The Sense inputs were not used for our application. The Oscillator determined what the switching frequency would be given a specific resistor and capacitor combination. The V Trip was connected to a potentiometer to allow the user to set the voltage across the output and therefore the speed as well. The Overcurrent function was an added safety feature to give the motor some protection from large currents. The Overcurrent could also be set using a potentiometer based on how much current the applied motor can handle. 2.4 Motor A basic dc motor, brush type or brushless, supplied by the user, is used for loads up to a peak of 500W for one minute with a continuous amount of 250W. This was not available in the testing stage but the circuit should properly function based on specifications and component ratings. 2.5 Performance Requirements 1. Input Voltage ranging from 10Vdc to 15Vdc. 2. Output voltage ranging from 0 Vdc to 15 Vdc specified by user. 3. Motor loads ranging up to a continuous 250W or 500W for one minute. 4. Efficiency greater than 85% (excluding motor). 5. Voltage ripple less than %2 . 3. System Design Schematic T i t l e S i z e D o c u m e n t N u m b e r R e v D a t e : S h e e t o f D C M o t o r C o n t r o l l e r A 1 1T u e s d a y , A p r i l 1 9 , 2 0 0 5 I N - 1 I N + 2 E A O U T 3 C L K / L E B 4 P G N D 1 2 I L I M 9 O U T A 1 1 V C 1 3 O U T B 1 4 V C C 1 5 C T 6 R A M P 7 R T 5 S S 8 V R E F 1 6 U 1 U C 3 8 2 5 D 1 M B R 3 0 6 0 L Q 1 S T B 2 0 0 N F 0 4 Q 2 S T B 2 0 0 N F 0 4 Q 3 A S T B 8 0 P F 5 5 Q 3 B S T B 8 0 P F 5 5 Q 4 S T B 2 0 0 N F 0 4 V 1 V d c 1 2 L 1 1 0 u H C S I O u t 1 O u t 2 O u t 3 C o m m o n O u t 4 O u t 5 O u t 6 B r a k e E n a b l e V d d S 3 S 2 S 1 F W D / R E V C S 2 V s s V t r i p O s c i l l a t o r L S 7 2 6 0 O v e r c u r r e n t U 2 2 0 1 9 1 8 1 7 1 6 1 5 1 4 1 3 1 2 1 1 1 2 3 4 5 6 7 8 9 1 0 R 1 1 k C 1 1 n R 2 2 5 k C 2 . 1 u F C 3 . 1 u F C 4 . 1 u F C 5 1 n C 6 4 . 7 n F 1 0 0 0 0 R 3 1 0 k R 4 1 0 k R 5 1 0 k R 6 1 0 k R 7 1 0 k R 8 1 M R 9 U s e r D e f i n e d R 1 0 1 0 0 k C 7 1 0 p F 0 0 0 0 0 0 0 0 0 0 Figure 7. DC Motor Controller System Schematic 4. Verification 4.1 Testing and Tolerance To ensure that the boost converter was continuously outputting 15V, various inputs ranging from 10-15V were used. As seen in Figure 8 the voltage was nearly 15.2V for all inputs resulting in near 100% regulation. The voltage was only off by about 1.3% due to the voltage divider feedback not having a precise reference output of 5.1V. This Figure 10. Outputs for an 11V input and 5V Output. The same values were used for the input and potentiometer voltage but with the LS7260 set to output a reverse voltage(grounding pin 19). Figure 11 is the output that corresponds to the reverse voltage. The waveforms are almost exactly the same except with negative voltage and current across the motor terminals. Again the switching signal would be much cleaner if a small resistor was placed in parallel with the boost capacitor. Figure 11. Outputs for an 11V input and -5V Output To ensure the circuit was working efficiently, a 10 and 8.33 load was used on the output. Unfortunately, a motor was not obtainable so the circuit could not be completely tested. In place of a motor, numerous 500Ω resistors were put in parallel to reduce the resistance and outline a motor equivalent. The 500Ω resistor boxes used from the power lab have a minimal amount of inductance within them so the output across the load terminals is primarily resistive with a small amount of series inductance. The input was varied and input/output power values were recorded. Figure 12 and 13 show the efficiency curves for the 10Ω and 8.33Ω loads. The efficiency for the smaller load had an average efficiency of about 92% while an increased load resulted in approximately 93%. This is well within specification where the efficiency had to be greater than 85%. This circuit is anticipated to have slightly larger efficiency at the larger power levels where the voltage drops are at a minimum. Figure 12. Efficiency for 10Ω Load. Efficiency (%) 100 95 30 85 80 75 70 65 60 55 50 1 8 20 22 24 26 28 Pout WW) Figure 13. Efficiency for 8.33Q Load.